February 23, 2017 | Author: lethanhvan186 | Category: N/A
AMCA Publication 203-90 (R2007)
Field Performance Measurement of Fan Systems
AIR MOVEMENT AND CONTROL
ASSOCIATION INTERNATIONAL, INC. The International Authority on Air System Components
AMCA PUBLICATION 203-90 (R2007)
Field Performance Measurement of Fan Systems
Air Movement and Control Association International, Inc. 30 West University Drive Arlington Heights, IL 60004-1893
© 2007 by Air Movement and Control Association International, Inc. All rights reserved. Reproduction or translation of any part of this work beyond that permitted by Sections 107 and 108 of the United States Copyright Act without the permission of the copyright owner is unlawful. Requests for permission or further information should be addressed to the Executive Director, Air Movement and Control Association International, Inc. at 30 West University Drive, Arlington Heights, IL 60004-1893 U.S.A.
Forward The original edition of Publication 203 was released in 1976. This, the second edition, updates much of the information that was presented. Annex K (estimating the power output of three phase motors) and Annex L (estimating belt drive losses) were rewritten and adjusted based on new information received from motor and drive manufacturers. Over four hundred belt drive loss tests were analyzed. New axial fan System Effect Factors were established based on a test project conducted and underwritten by AMCA. These factors were incorporated in their respective, applicable field test examples shown in Annex A. The intent of this publication is to provide information from which test procedures can be developed to meet the conditions and requirements encountered in specific field test situations. They include the proper procedure for determining various System Effect Factors. Numerous examples of actual field tests are presented in detail in Annex A. These examples provide sufficient guidance for the proper field testing of most fan system installations. Authority AMCA Publication 203 was approved by the Air Movement Control Association Membership in 1990. It was reaffirmed July, 2007. AMCA 203 Review Committee Robert H. Zaleski, Chairman
Acme Engineering & Manufacturing Corp.
Narsaiah Dasa
TLT-Babcock, Inc.
James L. Smith
Aerovent, Inc.
Jack E. Saunders
Barry Blower/SnyderGeneral Corp.
Erling Schmidt
Novenco, Inc.
Gerald P. Jolette
AMCA Staff
Disclaimer AMCA uses its best efforts to produce standards for the benefit of the industry and the public in light of available information and accepted industry practices. However, AMCA does not guarantee, certify or assure the safety or performance of any products, components or systems tested, designed, installed or operated in accordance with AMCA standards or that any tests conducted under its standards will be non-hazardous or free from risk. Objections to AMCA Standards and Certifications Programs Air Movement and Control Association International, Inc. will consider and decide all written complaints regarding its standards, certification programs, or interpretations thereof. For information on procedures for submitting and handling complaints, write to: Air Movement and Control Association International 30 West University Drive Arlington Heights, IL 60004-1893 U.S.A. or AMCA International, Incorporated c/o Federation of Environmental Trade Associations 2 Waltham Court, Milley Lane, Hare Hatch Reading, Berkshire RG10 9TH United Kingdom
Related AMCA Standards and Publications
Publication 200
AIR SYSTEMS System Pressure Losses Fan Performance Characteristics System Effect System Design Tolerances
Air Systems is intended to provide basic information needed to design effective and energy efficient air systems. Discussion is limited to systems where there is a clear separation of the fan inlet and outlet and does not cover applications in which fans are used only to circulate air in an open space. Publication 201
FANS AND SYSTEMS Fan Testing and Rating The Fan "Laws" Air Systems Fan and System Interaction System Effect Factors
Fans and Systems is aimed primarily at the designer of the air moving system and discusses the effect on inlet and outlet connections of the fan's performance. System Effect Factors, which must be included in the basic design calculations, are listed for various configurations. AMCA 201-02 and AMCA 203-90 are companion documents. Publication 202
TROUBLESHOOTING System Checklist Fan Manufacturer's Analysis Master Troubleshooting Appendices
Troubleshooting is intended to help identify and correct problems with the performance and operation of the air moving system after installation. Publication 203
FIELD PERFORMANCE MEASUREMENTS OF FAN SYSTEMS Acceptance Tests Test Methods and Instruments Precautions Limitations and Expected Accuracies Calculations
Field Performance Measurements of Fan Systems reviews the various problems of making field measurements and calculating the actual performance of the fan and system. AMCA 203-90 and AMCA 201-02 are companion documents.
TABLE OF CONTENTS
1.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
2.
Scope
3,
Types of Field Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
4.
Alternatives to Conducting Field Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2
5.
System Effect Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2
6.
Fan Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2
7.
Referenced Planes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2
8.
Symbols and Subscripts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
9.
Fan Flow Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
9.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 9.2 Velocity measuring instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 9.3 Location of traverse plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 9.4 The traverse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 9.5 Flow rate calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 9.6 Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8 10. Static Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8 10.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8 10.2 Pressure measuring instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 10.3 Static pressure measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 10.4 Static pressure calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10 10.5 Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 11. Fan Power Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 11.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 11.2 Power measurement methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 11.3 Power measuring instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 11.4 Power transmission losses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13
11.5 Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 12. Fan Speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 12.1 Speed measuring instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 12.2 Speed measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 13. Densities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 13.1 Locations of density determinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 13.2 Data required at each location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 13.3 Additional data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 13.4 Density values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 13.5 Temperatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 13.6 Barometric pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 13.7 Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 14. Conversion Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16 15. Test Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16 16. Precautions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17 17. Typical Fan-System Installations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18 17.1 Free inlet, free outlet fans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18 17.2 Free inlet, ducted outlet fans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 17.3 Ducted inlet, ducted outlet fans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 17.4 Ducted inlet, free outlet fans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 17.5 Air handling units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 Annex A
Field Test Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21
Annex B
Pitot-Static Tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97
Annex C
Double Reverse Tube . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .98
Annex D
Pitot-Static Tube Holder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .99
Annex E
Static Pressure Tap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .100
Annex F
Pitot-Static Tube Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .101
Annex G
Manometer Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .102
Annex H
Distribution of Traverse Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .104
Annex J
Instrumentation Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .106
Annex K
Phase Current Method for Estimating the Power Output of Three Phase Fan Motors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .108
Annex L
Estimated Belt Drive Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .110
Annex M
Density Determinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .112
Annex N
Density Charts and Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .117
Annex P
Diffusion at Fan Outlets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125
Annex R
Diffusion at Fan Outlets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .126
Annex S
Typical Format for Field Test Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .130
Annex T
Uncertainties Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .131
AMCA INTERNATIONAL, INC.
AMCA 203-90 (R2007)
Field Performance Measurement of Fan Systems
Fans for Rating
1. Introduction
The recommendations and examples in this publication may be applied to all types of centrifugal, axial, and mixed flow fans in ducted or nonducted installations used for heating, ventilating, air conditioning, mechanical draft, industrial process, exhaust, conveying, drying, air cleaning, dust collection, etc. Although the word air is used when reference is made in the general sense to the medium being handled by the fan, gases other than air are included in the scope of this publication.
Performance ratings of fans are developed from laboratory tests made according to specified procedures on standardized test setups. In North America, the standard is ANSI/AMCA Standard 210 / ANSI/ASHRAE 51 Laboratory Methods of Testing Fans for Rating. In actual systems in the field, very few fans are installed in conditions reproducing those specified in the laboratory standard. This means that, in assessing the performance of the installed fansystem, consideration must be given to the effect on the fan’s performance of the system connections, including elbows, obstructions in the path of the airflow, sudden changes of area, etc. The effects of system conditions on fan performance is discussed in Section 5, and more completely in AMCA Publication 201, Fans and Systems. A major problem of testing in the field is the difficulty of finding suitable locations for making accurate measurements of flow rate and pressure. Sections 9.3 and 10.3 outline the requirements of suitable measurement sections. Because these problems and others will require special consideration on each installation, it is not practical to write one standard procedure for the measurement of the performance of all fan-systems in the field. This publication offers guidelines to making performance measurements in the field which are practical and flexible enough to be applied to a wide range of fan and system combinations. Because of the wide variety of fan types and systems encountered in the field, Annex A includes examples of a number of different field tests. In most cases, these examples are based on actual tests which have been conducted in the field. Before performing any field test, it is strongly recommended that the following AMCA publications be carefully reviewed: AMCA Publication 200 - Air Systems AMCA Publication 201 - Fans and Systems AMCA Publication 202 - Troubleshooting AMCA Standard 210 - Laboratory Methods of Testing
2. Scope
Measurement of sound, vibration, and stress levels are not within the scope of this publication.
3. Types of Field Tests There are three general categories of field tests: A) General Fan System Evaluation - A measurement of the fan-system’s performance to use as the basis of modification or adjustment of the system. B) Acceptance Test - A test specified in the sales agreement to verify that the fan is achieving the specified performance. C) Proof of Performance Test - A test in response to a complaint to demonstrate that the fan is meeting the specified performance requirement. As acceptance and proof of performance tests are related to contract provisions, they are usually subject to more stringent requirements and are usually more costly than a general evaluation test. In the case of large fans used in industrial applications and of mechanical draft fans used in the electrical power generation industry the performance of a field test may be part of the purchase agreement between the fan manufacturer and the customer. In addition to Publication 203, AMCA Standard 803 Site Performance Test Standard-Power Plant and Industrial Fans defines the conditions which must be met to achieve higher accuracy of measurement. In new installations of this type, it is desirable to include a suitable measuring section in the design. Agreement must be reached on the test method to be used prior to performance of the test.
1
AMCA 203-90 (R2007)
4. Alternatives to Field Tests In some cases, considerations such as cost and problems of making accurate measurements may make the following alternative methods of testing worth investigation: A) Testing the fan before installation in a laboratory equipped to perform tests in accordance with AMCA Standard 210. Limitations in laboratory test facilities may preclude tests on full size fans. In this case, the full size fan can be tested at the installation site in accordance with AMCA Standard 210. This will usually require the installation of special ductwork. B) Testing a reduced scale model of the fan in accordance with AMCA Standard 210 and determining the performance of the full size fan as described in AMCA Publication 802, Power Plant Fans – Establishing Performance Using Laboratory Methods. C) Testing a reduced scale model of the complete fan and system using the test methods outlined in this publication. Tests conducted in accordance with AMCA Standard 210 will verify the performance characteristics of the fan but will not take into account the effect of the system connections on the fan’s performance (see Section 5).
5. System Effect Factors AMCA Publication 201, Fans and Systems, deals in detail with the effect of system connections on fan performance. It gives system effect factors for a wide variety of obstructions and configurations which may affect a fan’s performance. System Effect Factor (SEF) is a pressure loss which recognizes the effect of fan inlet restrictions, fan outlet restrictions, or other conditions influencing fan performance when installed in the system. SYSTEM EFFECT FACTORS (SEFs) ARE INTENDED TO BE USED IN CONJUNCTION WITH THE SYSTEM RESISTANCE CHARACTERISTICS IN THE FAN SELECTION PROCESS. Where SEFs are not applied in the fan selection process, SEFs must be applied in the calculations of the results of field tests. This is done for the purpose of allowing direct comparison of the test results to the design static pressure calculation. Thus, for a field test, the fan static pressure is defined as: Ps = Ps2 - Ps1 – Pv1 + SEF 1 + SEF 2 + …+ SEF n 2
Examples of the application of SEFs in determining the results of field tests are included in Annex A. In field tests of fan-system installations in which system effects have not been accounted for, it is important that their sources be recognized and their magnitudes be established prior to testing. The alternative to dealing with a large magnitude SEF is to eliminate its source. This requires revisions to the system. This alternative course of action is recommended when swirl exists at the fan inlet (see Publication 201, Figure 9.8). The effect on fan performance as a result of swirl at the inlet is impossible to estimate accurately as the system effect is dependent upon the degree of swirl. The effect can range from a minor amount to an amount that results in the fan-system performance being completely unacceptable.
6. Fan Performance Fan performance is a statement of fan flow rate, fan total or static pressures, and fan power input at stated fan speed and fan air density. Fan total or static efficiencies may be included. The fan air density is the density at the fan inlet. The fan flow rate is the volume flow rate at the fan inlet density.
7. Referenced Planes Certain locations within a fan-system installation are significant to field tests. These locations are designated as follows: Plane 1: Plane of fan inlet Plane 2: Plane of fan outlet Plane 3: Plane of Pitot-static tube traverse for purposes of determining flow rate Plane 4: Plane of static pressure measurement upstream of fan Plane 5: Plane of static pressure measurement downstream of fan The use of the numerical designations as subscripts indicate that the values pertain to those locations.
AMCA 203-90 (R2007)
8. Symbols and Subscripts SYMBOL A D De FLA H HL Hmo kW L N NLA NPH NPV Ps Psx Pt Ptx Pv Pvx pb pe pp px Q Qi Qx SEF T td tw V ΔPx,x’ ΔPs ρ ρx Σ
UNIT
9.1 General
Area of cross-section Diameter Equivalent diameter Full load amps Fan power input Power transmission loss Motor power output Electrical power Length Speed of rotation No load amps Nameplated horsepower Nameplated volts Fan static pressure Static pressure at Plane x Fan total pressure Total pressure at Plane x Fan velocity pressure Velocity pressure at Plane x Barometric pressure Saturated vapor pressure at tw Partial vapor pressure Absolute pressure at Plane x Fan flow rate Interpolated flow rate Flow rate at Plane x System effect factor Torque Dry-bulb temperature Wet-bulb temperature Velocity Pressure loss between Planes x and x’ Pressure loss across damper Fan gas density Gas density at Plane x Summation sign
ft2 ft ft amps hp hp hp kilowatts ft rpm amps hp volts in. wg in. wg in. wg in. wg in. wg in. wg in. Hg in. Hg in. Hg in. Hg cfm cfm cfm in. wg lb-in. °F °F fpm
Determine fan flow rate using the area, velocity pressure, and density at the traverse plane and the density at the fan inlet. The velocity pressure at the traverse plane is the root mean square of the velocity pressure measurements made in a traverse of the plane. The flow rate at the traverse plane is calculated by converting the velocity pressure to its equivalent velocity and multiplying by the area of the traverse plane.
Airflow direction
---
SUBSCRIPT c r x 1 2 3 4 5
DESCRIPTION
9. Fan Flow Rate
in. wg in. wg lbm/ft3 lbm/ft3 ---
DESCRIPTION
Value converted to specified conditions Reading Plane 1, 2, 3, ..., as appropriate Plane 1 (fan inlet) Plane 2 (fan outlet) Plane 3 (plane of Pitot-static traverse for purpose of determining flow rate Plane 4 (plane of static pressure measurement upstream of fan) Plane 5 (plane of static pressure measurement downstream of fan)
9.2 Velocity measuring instruments Use a Pitot-static tube of the proportions shown in Annex B or a double reverse tube, shown in Annex C, and an inclined manometer to measure velocity pressure. The velocity pressure at a point in a gas stream is numerically equal to the total pressure diminished by the static pressure. The Pitot-static tube is connected to the inclined manometer as shown in Annex F. The double reverse tube is connected to the inclined manometer as shown in Annex C. 9.2.1 Pitot-static tube. The Pitot-static tube is considered to be a primary instrument and need not be calibrated if maintained in the specified condition. It is suited for use in relatively clean gases. It may be used in gases that contain moderate levels of particulate matter such as dust, water, or dirt, provided certain precautions are employed (see Section 15). 9.2.2 Double reverse tube. The double reverse tube is used when the amount of particulate matter in the gas stream impairs the function of the Pitot-static tube. The double reverse tube requires calibration. It is important that the double reverse tube be used in the same orientation as used during calibration. Mark the double reverse tube to indicate the direction of the gas flow used in its calibration. 9.2.3 Inclined manometers. Inclined manometers are available in both fixed and adjustable range types. Both types require calibration. The adjustable range type is convenient in that it may be adjusted at the test site to the range appropriate to the velocity pressures which are to be measured. It is adjusted by changing the slope to any of the various fixed settings and by changing the range scale accordingly. Each setting provides a different ratio of the length of the indicating column to its indicated height. Adjustable range type manometers in which the slope may be fixed at 1:1, 20:1, and intermediate ratios are available (see Figure 10 in Annex G). 3
AMCA 203-90 (R2007) The accuracy of the manometer used in the measurement of velocity pressures is of prime importance. Select a manometer that will provide an acceptable degree of accuracy; consider the range, slope, quality, scale graduations, indicating fluid of the instrument and the range of the velocity pressures to be measured. The graph in Annex G indicates the effect of expected resolution of manometer readings on the accuracy of velocity determinations. The basis for this graph is described in Section 9.6. Determine velocities in the very low range more accurately by using a manometer with a slope of 20:1. Due to practical limitations in length, its use is restricted to measurements where the velocities are very low. Also, errors in velocity determinations made by using a Pitot-static tube and manometer exceed normally acceptable values at velocity pressure readings less than 0.023 in. wg. This corresponds to a velocity of approximately 600 fpm for air of 0.075 lbm/ft3 density. 9.2.4 Low velocity instruments. Normally, velocities encountered in the field test situations are well in excess of 600 fpm. Therefore, recommendations regarding alternate test procedures and instrumentation for use for velocities less than 600 fpm are not presented in this publication. Descriptions of various types of instruments used to determine range velocities are presented in Annex J. Most of the instruments require frequent calibration, and some are not suited for use in high temperature, dirty, wet, corrosive, or explosive atmospheres. If it is necessary to use one of these instruments, the procedure for its use, its calibration, and the expected accuracy of results should be agreed upon by all interested parties.
9.3 Location of traverse plane For field tests, suitable test measurement station locations must be provided in the system. When suitable locations are not available, consider making temporary or permanent alterations to the ducting for improved test accuracy. For free inlet, free outlet fans, convert a free inlet, free outlet fan to a ducted inlet, free outlet fan by the addition of a temporary duct. Estimate free inlet, free outlet fan flow rate by measuring other parameters and interpreting certified ratings performance (see Section 17.1).
4
than 75% of the velocity pressure measurements are greater than 1/10 of the maximum measurement (see Figure 9.1) 2) The flow streams should be at right angles to the traverse plane. Variations from this flow condition as a result of swirl or other mass turbulence are considered acceptable when the angle between the flow stream and the traverse plane is within 10 degrees of a right angle. The angle of the flow stream in any specific location is indicated by the orientation of the nose of the Pitot-static tube that produces the maximum velocity pressure reading at the location. 3) The cross-sectional shape of the airway in which the traverse plane is located should not be irregular. Proper distribution of traverse points and accurate determination of the area of the traverse plane are difficult to achieve when the airway does not conform closely to a regular shape. 4) The cross-sectional shape and area of the airway should be uniform throughout the length of the airway in the vicinity of the traverse plane. When the divergence or convergence of the airway is irregular or more than moderate in degree, significantly nonuniform flow conditions may exist. 5) The traverse plane should be located to minimize the effects of gas leaks between the traverse plane and the fan. 6) When it is necessary to locate the traverse plane in a converging or diverging airway (not recommended), note that the traverse plane and area is located at the tip of the Pitot-static tube. A location well downstream in a long, straight run of uniform cross-section duct will usually provide acceptable conditions for the Pitot traverse plane. When locating the traverse plane close to the fan, as is often done in order to minimize the effect of leakage, flow conditions upstream of the fan are usually more suitable. In some installations, more than one traverse plane may be required in order to account for the total flow (Annex A contains examples).
A Pitot traverse plane suitable for the measurements used to determine flow rate are as follows:
When a field test is anticipated, particularly when the requirement for a field test is an item in the specifications, the system designer should provide a suitable traverse plane location in the system.
1) The velocity distribution should be uniform throughout the traverse plane. The uniformity of distribution is considered acceptable when more
When the fan is ducted outlet and the traverse plane is to be located downstream from the fan, the
AMCA 203-90 (R2007)
Pv MAX
Pv MAX 10
A: IDEAL Pv DISTRIBUTION
Pv MAX
Pv MAX 10
B: GOOD Pv DISTRIBUTION (ALSO SATISFACTORY FOR FLOW INTO FAN INLETS. MAY BE UNSATISFACTORY FOR FLOW INTO INLET BOXES - MAY PRODUCE SWIRL IN BOXES)
Pv MAX
Pv MAX 10
Pv MAX
Pv MAX 10
60% 80%
C: SATISFACTORY Pv DISTRIBUTION - MORE THAN 75% OF Pv READINGS GREATER THAN: Pv MAX 10 (UNSATISFACTORY FOR FLOW INTO FAN INLETS OR INLET BOXES)
Pv MAX 10
D:
DO NOT USE
UNSATISFACTORY Pv DISTRIBUTION LESS THAN 75% OF Pv READINGS GREATER THAN: Pv MAX 10 (UNSATISFACTORY FOR FLOW INTO FAN INLETS OR INLET BOXES)
Pv MAX
Pv MAX 10
Pv MAX
40%
35%
20%
35%
E: DO NOT USE UNSATISFACTORY Pv DISTRIBUTION LESS THAN 75% OF Pv READINGS GREATER THAN: P MAX v
10 (UNSATISFACTORY FOR FLOW INTO FAN INLETS OR INLET BOXES)
F: DO NOT USE UNSATISFACTORY Pv DISTRIBUTION LESS THAN 75% OF Pv READINGS GREATER THAN: P MAX v
10 (UNSATISFACTORY FOR FLOW INTO FAN INLETS OR INLET BOXES)
Figure 9.1 - Typical Velocity Pressure Distributions Encountered in Velocity Pressure Measurement Planes in Fan-System Installations 5
AMCA 203-90 (R2007)
MEASUREMENT PLANE De MIN. 2 12 in. MIN.
Y
WHERE: De =
4YZ π
INLET BOX DAMPERS
Z
Note: The measurement plane should be located a minimum of ½ De from the inlet cone, but not less than 12 in. from the leaving edge of the damper blades. Figure 9.2
STACK
VELOCITY PROFILE
Note: Spiral vortex may form when fan discharges directly into a stack or similar arrangement. Figure 9.3 6
AMCA 203-90 (R2007) traverse plane should be situated a sufficient distance downstream from the fan to allow the flow to diffuse to a more uniform velocity distribution and to allow the conversion of velocity pressure to static pressure. Annex P provides guidance for the location of the traverse plane in these cases. The location of the traverse plane on the inlet side of the fan should not be less than ½ equivalent diameter from the fan inlet. Regions immediately downstream from elbows, obstructions and abrupt changes in airway area are not suitable traverse plane locations. Regions where unacceptable levels of swirl are usually present, such as the region downstream from an axial flow fan that is not equipped with straightening vanes, should be avoided. Swirl may form when a fan discharges directly into a stack or similar arrangement (see Figure 9.2). 9.3.1 Inlet box location. When the traverse plane must be located within an inlet box, the plane should be located a minimum of 12 inches downstream from the leaving edges of the damper blades and not less than ½ equivalent diameter upstream from the edge of the inlet cone (see Figure 9.3). Do not locate traverse points in the wake of individual damper blades. In the case of double inlet fans, traverses must be conducted in both inlet boxes in order to determine the total flow rate. 9.3.2 Alternative locations. On occasion, an undesirable traverse plane location is unavoidable, or each of a limited number of prospective locations lacks one or more desirable qualities. In such cases, the alternatives are: 1) Accept the most suitable location and evaluate the effects of the undesirable aspects of the location on the accuracy of the test results. In some instances, the estimated accuracy may indicate that the results of the test would be meaningless, particularly in acceptance tests and proof of performance tests. 2) Provide a suitable location by modifying the system. This course of action is recommended for acceptance tests and proof of performance tests. The modifications may be temporary, permanent, minor or extensive, depending on the specific conditions encountered. When the inlet side of the fan is not ducted but is designed to accept a duct, consider installing a short length of inlet duct to provide a suitable traverse plane location. This duct should be of a size and shape to fit the fan inlet, a minimum of 2 equivalent diameters long and equipped with a bell shaped or flared fitting at its inlet. The traverse plane should be located a minimum of ½ equivalent diameters from the fan inlet and not less than 1½
equivalent diameters from the inlet of the duct. Where the duct is small, its length may necessarily be greater than 2 equivalent diameters in order to ensure that the tip of the Pitot-static tube is a minimum of 1½ equivalent diameters from the duct inlet. This short length of duct should produce no significant addition to the system resistance, but in some cases it may alter the pattern of flow into the fan impeller, and thereby affect the performance of the fan slightly.
9.4 The traverse Annex H contains recommendations for the number and distribution of measurement points in the traverse plane. If the flow conditions at the traverse plane are less than satisfactory, increase the number of measurement points in the traverse to improve accuracy. Since the flow at a traverse plane is never strictly steady, the velocity pressure measurements indicated by the manometer will fluctuate. Each velocity pressure measurement should be mentally averaged on a time-weighted basis. Any velocity pressure measurement that appears as a negative reading is to be considered a velocity pressure measurement of zero and included as such in the calculation of the average velocity pressure. When it is necessary to locate the traverse plane in a converging or diverging airway, orient the nose of the Pitot-static tube such that it coincides with the anticipated line of the flow stream. This is particularly important at measurement points near the walls of the airway (see Annex A-1A). No appreciable effect on Pitot-static tube readings occur until the angle of misalignment between the airflow and the tube exceeds 10 degrees.
9.5 Flow rate calculations 9.5.1 Flow rate at traverse plane. The flow rate at the traverse plane is calculated as follows: Q3 = V3A3 Where: A3 = the area of the traverse plane V3 = the average velocity at the traverse plane = 1096 (Pv3/ρ3)0.5 ρ3 = the density at the traverse plane Pv3 = the root mean square velocity pressure at the traverse plane = [∑(Pv3r)0.5 / number of readings]2 7
AMCA 203-90 (R2007) Pv3r is the velocity pressure reading, corrected for manometer calibration and where applicable, corrected for the calibration of the double reverse tube. It is important that the calibration of the double reverse tube be applied correctly. The use of the calibration of the double reverse tube is described in Annex C. 9.5.2 Continuity of mass. The calculations of fan flow rate are based on considerations of continuity of mass, and as such, it is assumed that no mass is added or removed from the gas stream between the traverse plane and the fan inlet. In the general application, having determined the flow rate and density at the traverse plane, the flow rate at any location, (x), in the fan-system installation may be calculated, providing the density at this location is known and the assumption noted above is valid, i.e.: Qx = Q3 (ρ3/ρx) 9.5.3 Fan flow rate, single traverse plane. Where a single traverse plane is used, the calculation of the fan flow rate is: Q = Q1 = Q3 (ρ3/ρ1) Where: Q3 and ρ3 = as described in Section 9.5.1
ρ1 = the density at the fan inlet 9.5.4 Fan flow rate, multiple traverse planes. When it is necessary to use more than one traverse plane in order to account for the total flow: Q = Q1 = Q3a (ρ3a/ρ1) + Q3b (ρ3b/ρ1) + ... + Q3n (ρ3n/ρ1)
9.6 Accuracy The performance item of major concern in most fansystem installations is the flow rate. Every effort should be made to improve the accuracy of the flow rate determination. The uncertainty analysis presented in Annex T indicates that the uncertainties in flow rate determinations will range from 2% to 10%. This range is based on considerations of the conditions that are encountered in most field test situations. This includes instances in which the conditions at the Pitot traverse plane do not conform to all of the qualifications indicated in Section 9.3. The graph in Annex G provides guidance for improving the accuracy of the flow rate 8
determinations. This graph indicates the effect of expected resolution of velocity determinations. This effect is shown for several manometer slope ratios. For all ratios, the expected resolution used as a basis for the graph is the length of indicating column equivalent to 0.05 in. wg in a manometer with slope ratio of 1:1. As indicated in the graph, reading resolution uncertainty can be significant. However, this uncertainty can be controlled by selecting a manometer with a slope suited to the velocity pressures to be measured and by avoiding regions of very low velocity in the selection of the traverse plane location. Reading resolution uncertainties exceed normally acceptable values at velocity pressures less than 0.023 in. wg. This corresponds to a velocity of approximately 600 fpm for air of 0.075 lbm/ft3 density. Generally, ducts are sized for velocities considerably in excess of 600 fpm. Velocities less than 600 fpm may exist in certain sections of the system in some installations, but these sections can usually be avoided. Do no use a Pitot-static tube and manometer to determine velocities in the low ranges associated with filters and cooling coils in air conditioning, heating, and ventilating units. In some instances, the uncertainties incurred in the determinations of low velocity flows may be acceptable. For example, an uncertainty of 15% in the determination of the flow rate in a branch duct that accounts for 20% of the total flow rate for the system affects the accuracy of the total flow rate determination by only 3%. In addition to low range velocities, other conditions may exist at the traverse plane which can significantly affect the accuracy of the flow rate determination. These include nonuniform velocity distribution, swirl, and other mass turbulence. Improve the accuracy of the flow rate determination by avoiding these conditions in the selection of the traverse plane location, or improve the conditions by modifying the system.
10. Fan Static Pressure 10.1 General Determine fan static pressure by using the static pressures at the fan inlet and outlet, the velocity pressure at the fan inlet, and applicable System Effect Factors. The use of System Effect Factors in the determination of fan static pressure is described in Section 5. The velocity pressure at the fan inlet is the calculated average velocity pressure at this location, and as such, its determination is based on the fan flow rate, the density at the fan inlet, and the fan inlet area. The static pressures at the fan inlet and outlet may be obtained directly by making pressure measurements at these locations; or they may be
AMCA 203-90 (R2007) determined by making pressure measurements at other locations, upstream and downstream of the fan. In the latter case, the determinations must account for the effects of velocity pressure conversions and pressure losses, as may occur between the measurement planes and the planes of interest.
10.2 Pressure measuring instruments This section describes only the instruments for use in measuring static pressure. Instruments for use in the other measurements involved in the determination of fan static pressure are described in Section 13. Use a Pitot-static tube of the proportions shown in Annex B, a double reverse tube as shown in Annex C, or a side wall pressure tap as shown in Annex E, and a manometer to measure static pressure. 10.2.1 Pitot-static tube. The comments that appear in Section 9.2 regarding the use and calibration of the Pitot-static tube are applicable to its use in the measurement of static pressures. 10.2.2 Double reverse tube. The double reverse tube cannot be used to measure static pressure directly. It must be connected to two manometers and the static pressure for each point of measurement must be calculated. Both the manometer connections and the method of calculation are shown in Annex C. 10.2.3 Pressure tap. The pressure tap does not require calibration. Use no fewer than four taps located 90 degrees apart. In rectangular ducts, a pressure tap should be installed near the center of each wall. It is important that the inner surfaces of the duct in the vicinities of the pressure taps be smooth and free from irregularities, and that the velocity of the gas stream does not influence the pressure measurements. 10.2.4 Manometers. A manometer with either vertical or inclined indicating column may be used to measure static pressure. Inclined manometers used to measure static pressures require calibration and should be selected for the quality, range, slope, scale graduations, and indicating fluid necessary to minimize reading resolution errors.
10.3 Static pressure measurements It is important that all static pressure measurements be referred to the same atmospheric pressure, and this atmospheric pressure be that for which the barometric pressure is determined. Make static pressure measurements near the fan inlet and the fan outlet, and where the airway
between the measurement plane and the plane of interest is straight and without change in crosssectional area. Then the duct friction loss between the measurement plane and the plane of interest is usually insignificant, and considerations of velocity pressure conversions and calculations of pressure losses for duct fitting and other system components can be avoided. When a system component is situated between the measurement plane and the plane of interest, the pressure loss of the component must be calculated and credited to the fan. The calculation of the pressure loss is usually based on the component’s performance ratings, which may be obtained from the manufacturer of the item. If there is a change in area between the measurement plane and the plane of interest, then the calculation of the static pressure at the plane of interest must account for velocity pressure conversion and include any associated pressure loss. When the change in area is moderate and gradual, the conversion of velocity pressure is considered to occur without loss and the static pressure is calculated on the basis of no change in total pressure between the measurement plane and the plane of interest. This assumes that the duct friction loss between the two planes is negligible. When the change in area is an abrupt and sizable enlargement, as in a duct leading into a large plenum, the loss is considered to be equivalent to the velocity pressure in the smaller area, and the static pressure at the plane of interest is considered to be the same as the static pressure at the measurement plane. This assumes that the velocity pressure in the larger area and the duct friction loss are negligible. 10.3.1 Location of the measuring plane. When the fan is ducted outlet, the static pressure measurement plane downstream of the fan should be situated a sufficient distance from the fan outlet to allow the flow to diffuse to a more uniform velocity distribution and to allow the conversion of velocity pressure to static pressure. See Annex P for guidance in locating the measurement plane in these cases. In general, pressure taps should be used if it is necessary to measure static pressure in the immediate vicinity of the fan outlet. The static pressure at this location is difficult to measure accurately with a Pitot-static tube due to the existence of turbulence and localized high velocities. If the surface conditions or the velocities at the duct walls are unsuited for the use of pressure taps, then a Pitot-static tube must be used with extreme care, particularly in aligning the nose of the tube with the lines of the flow streams. The location of the static pressure measurement 9
AMCA 203-90 (R2007) plane upstream of the fan should not be less than ½ equivalent diameter from the fan inlet. In the event that static pressure measurements must be made in an inlet box, the measurement plane should be located as indicated in Figure 9.2. In the case of double inlet fans, static pressure measurements must be made in both inlet boxes in order to determine the average static pressure on the inlet side of the fan. In general, the qualifications for a plane well suited for the measurement of static pressure are the same as those for the measurement of velocity pressure, as indicated in Section 9.3:
negative. By definition, positive values are those measured as being greater than atmospheric pressures; negative values are those measured as being less than atmospheric pressure. In all of the equations in this publication, the values of static pressures must be entered with their proper signs and combined algebraically. 10.4.1 Static pressure at measuring planes. The static pressure at a plane of measurement (x) is calculated as follows:
Psx = 1) The velocity distribution should be uniform throughout the traverse plane.
∑P
sxr
number of readings
Where: 2) The flow streams should be at right angles to the plane. 3) The cross-sectional shape of the airway in which the plane is located should not be irregular. 4) The cross-sectional shape and area of the airway should be uniform throughout the length of the airway in the vicinity of the plane. 5) The plane should be located such as to minimize the effects of leaks in the portion of the system that is located between the plane and the fan. A long, straight run of duct upstream of the measurement plane will usually provide acceptable conditions at the plane. Regions immediately downstream from elbows, obstructions, and abrupt changes in airway area are generally unsuitable locations. Regions where unacceptable levels of turbulence are present should be avoided. If in any fan-system installation the prospective locations for static pressure measurement planes lack one or more desirable qualities, the alternatives are to accept the best qualified locations and evaluate the effects of the undesirable aspects of the conditions on the accuracy of the test results or provide suitable locations by modifying the system.
10
Psxr = the static pressure reading, corrected for manometer calibration 10.4.2 Static pressure at fan inlet or outlet. The static pressure at the fan inlet, Ps1, and the static pressure at the fan outlet, Ps2, may be measured directly in some cases. In most cases, the static pressure measurements for use in determining fan static pressure will not be made directly at the fan inlet and outlet, but at locations a relatively short distance upstream from the fan inlet and downstream from the fan outlet. These static pressure measurements are designated Ps4 and Ps5, respectively. Static pressure at the fan inlet, Ps1, is derived as follows: Pt4 = Pt1 + ΔP4,1 Where: Pt4 = the total pressure plane of measurement Pt1 = the total pressure at the fan inlet ΔP4,1 = the sum of the pressure losses between the two planes These losses (ΔP) include those attributable to duct friction, duct fittings, other system components, and changes in airway area. Although ΔP represents a loss in all cases, it is considered a positive value as used in the equations in this publication. By substitution and rearrangement:
10.3.2 When using a Pitot-static tube or a double reverse tube to measure static pressure, a number of measurements must be made throughout the plane. Use Annex H to determine the number and distribution of the measurement points. When using pressure taps, a single measurement at each of the taps located at the plane is sufficient.
Similarly, for static pressure at the fan outlet, Ps2:
10.4 Static pressure calculations
Pt2 = Pt5 + ΔP2,5
Static pressure measurements may be positive or
Ps2 = Ps5 + Pv5 - Pv2 + ΔP2,5
Ps1 = Ps4 + Pv4 - Pv1 - ΔP4,1
AMCA 203-90 (R2007) Where:
10.5 Accuracy
The velocity pressures at the various planes can be determined from the following general equations for the velocity pressure at a plane of measurement (x):
The uncertainty analyses in Annex T indicate that the uncertainties in fan static pressure determinations are expected range from 2% to 8%. This range is based on considerations of the conditions expected to be encountered in most field test situations.
Pvx = Pv3 (A3/Ax)2 (ρ3/ρx) Or: Pvx = (Qx/1096Ax)2 ρx Locate the static pressure measurement planes such that the pressure losses between the measurement planes and the planes of interest are insignificant. This will eliminate the uncertainties involved in the determination of the pressure losses, and the equations for Ps1 and Ps2 reduce to the following: Ps1 = Ps4 + Pv4 - Pv1 Ps2 = Ps5 + Pv5 - Pv2 These equations may be used when changes in area between the measurement planes and the planes of interest are moderate and gradual, and the pressure losses associated with conversions of velocity pressure to static pressure are negligible.
Improve the accuracy of the fan static pressure determination by avoiding static pressure measurement plane locations where turbulence or other unsteady flow conditions will produce significant uncertainties in the mental averaging of pressure readings. Other reading resolution uncertainties are not as significant in the fan static pressure determination as in the determination of flow rate. Generally, static pressure measurements are much greater in magnitude than velocity pressure measurements, and the selection of a manometer that will provide reasonably good accuracy is not usually a problem. The uncertainty analyses in Annex T and the resulting anticipated uncertainty range do not account for uncertainties that may occur in the following: •
Determinations of velocity pressure conversions occurring between the measurement planes and the planes of the fan inlet or fan outlet. The area and density values that are involved in these determinations are usually obtained without significant uncertainties. However, pressure losses associated with velocity pressure conversions are often difficult to determine accurately.
•
Determinations of other pressure losses occurring between the measurement planes and the fan inlet or fan outlet. This includes pressure losses in ducts, duct fittings, and other system components. The calculations of these losses are based on the assumption of uniform flow conditions. This assumption may not be valid, and the calculated pressure loss values may be significantly inaccurate.
•
Determinations of the values of System Effect Factors. These determinations are based on limited information, and as such, are subject to uncertainty.
If, in addition to the losses being negligible there are no changes in the areas between the measurement planes and the respective planes of interest, then the equations are further reduced to: Ps1 = Ps4 Ps2 = Ps5 These equations may also be used when the only losses between the measurement planes and the planes of interest are those associated with changes in area that are abrupt and sizable enlargements in the direction of flow. This assumes that the velocity pressure in the larger area is negligible. 10.4.3 Fan static pressure. The equation for fan static pressure is: Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2 + ... + SEF n Where: SEF 1, SEF 2, ... SEF n = System Effect Factors that account for the various System Effects that are uncorrected and exist at the time of the field test.
Avoid situations requiring these determinations, thereby eliminating them as sources for uncertainties. The uncertainties involved in determining the values of System Effect Factors can be avoided only by correcting the causes of the System Effects. This requires alterations to the system. 11
AMCA 203-90 (R2007)
11. Fan Power Input 11.1 General Fan power input data included as part of the fan performance ratings are normally defined and limited to either: •
power input to the fan shaft
•
the total of the power input to the fan shaft and the power transmission loss
The losses in fan shaft bearings are included in either case. Since the results of field tests are usually compared to the rated performance characteristics of the fan, field test values of fan power input should be determined on the same basis as that used in the fan ratings. For belt driven fans, the rated fan power input may or may not include belt drive losses. The information regarding the basis of the rated fan power input accompanies the rating data or is otherwise available from the fan manufacturer. In most instances, when a power transmission loss occurs, the loss will have to be determined and subtracted from the motor output in order to obtain the fan power input.
11.2 Power measurement methods In view of the fact that accuracy requirements for field test determinations of fan power input vary considerably, a number of test methods are recommended. These methods are intended to provide economical and practical alternatives for dealing with various levels of accuracy requirements. 11.2.1 Phase current method. This method for estimating the power output of three phase motors is based on the relationship of motor current and motor power output. The method, described in Annex K, requires measurements of the phase currents and voltages supplied to the motor while driving the fan. Depending on the operating load point of the motor, it may also involve the measurements of the no load phase currents. The phase current method is convenient and sufficiently accurate for most field tests. In this method, the closer the actual phase current is to the motor nameplate value of full load amps, the greater the accuracy. Since fan motors are normally selected for operation at or near the full load point, this method provides a reasonably accurate estimate of the power output of the fan motor. Determine fan power input by using the motor power output and, where applicable, the power transmission loss. 12
11.2.2 Typical motor performance data. Typical motor performance data may be used to determine fan power input. These data, which are referred to as typical in that the data and the actual performance of the motor are expected to correspond closely, can usually be obtained from the motor manufacturer. The data provided can be in a variety of forms, but are sufficient to determine motor power output based on electrical input measurements. It is important that the power supplied to the motor during the field test be consistent with that used as the basis for the motor performance data. The phase voltage should be stable and balanced, and the average should be withing 2% of the voltage indicated in the performance data. Depending on the form of the typical motor performance data, motor power output is determined by one of the following methods: 1) Given the typical motor performance chart of watts input versus motor power output at a stated voltage. Hmo, is the value in the typical motor performance data that corresponds to the field test measurement of watts input to the motor. 2) Given the typical motor performance chart of watts input versus torque output and speed at a stated voltage. Use the field test measurement of watts input and the corresponding typical motor performance data values of torque output and speed; the motor power output is calculated as: Hmo =
T ×N 63025
3) Given the typical motor performance chart of watts input versus motor efficiency at a stated voltage. Use the field test measurement of watts input and the corresponding typical motor performance data value of motor efficiency, the motor power output is calculated as: Hmo =
watts input × motor efficiency 746
4) Given the typical motor performance chart of amps versus power factor and motor efficiency at a stated voltage. Use the field test measurements of amps input and volts, and the typical motor performance data values of power factor (pf) and motor efficiency, corresponding to the measured amps input; the motor power output is calculated as:
AMCA 203-90 (R2007)
Hmo =
amps × volts × pf × motor efficiency 746
Or, for three phase motors: Hmo =
(3)0.5 × amps × volts × pf × motor efficiency 746
In both equations, amps and volts are the field test measurement values and, in the case of three phase motors, are the averages of the measured phase values. The fan power input is the motor power output minus the power transmission loss, where applicable. 11.2.3 Calibrated motors. A calibrated motor may be used to determine fan power input. When intending to use this method, it is usually necessary to specify in the motor purchase arrangements that the motor be calibrated since an additional cost is normally involved. Calibration data are similar to typical motor performance data with the exception that, instead of being merely typical, the calibration data represent the performance of a specific motor, based on a test of the motor. The motor is calibrated over a range of operation. Electrical input data and other data sufficient for the determination of power output are obtained in the calibration. The calibration normally provides data for operation at nameplate voltage, but may include data for operation at voltages 10% greater and 10% less than nameplate voltage. It is important that the power supplied to the motor during the field test be consistent with that used in its calibration. The phase voltage should stable and balanced, and the average should be within 2% of the voltage at which the motor was calibrated. The field test measurements and the calculations involved in the determination of motor power output are the same as those described in Section 11.2.2 for use with typical motor performance data. The fan power input is the motor power output minus the power transmission loss, where applicable. A calibrated motor provides accurate data to determine motor power output. However, the cost of the calibration is a limiting factor in the use of this method in field tests. For low horsepower applications, the fan manufacturer may be able to calibrate a motor. 11.2.4. Torquemeters. Another method to determine fan power input involves the use of a torquemeter installed between the fan and the driver. The use of a torquemeter requires some prearrangement with the purchaser, who would normally have specified such equipment, so that site conditions can be altered to
accommodate its installation. The torquemeter is extremely limited in field test application. This is due mainly to is high cost and the cost of its installation. In addition, the length of the shut down time and the revisions to site conditions required for its installation are usually undesirable. For practical considerations, it is not normally used in cases where the fan is belt driven and where the fan impeller is installed directly on the motor shaft.
11.3 Power measuring instruments Measurement of current, voltage, watts, and power factor can be obtained by using an industrial type power analyzer of good quality. This type of instrument is available with accuracies of 1% full scale for volts, amps and power factor, and 2% full scale for watts. Normally, the higher levels of accuracy requirements can be met by using this type of instrument, providing the measurements are well up on the scales. In many cases, accuracy level requirements will permit the use of a clip-on type ammeter-voltmeter. Clip-on instruments with accuracies of 3% full scale are available.
11.4 Power transmission losses Several types of power transmission equipment are used in driving fans. Those in which power transmission losses should be considered in the determination of fan power input include belt drives, gear boxes, fluid drives, and electromechanical couplings. Information as to whether the fan power input ratings include power transmission losses is included in the published performance ratings or is otherwise available from the fan manufacturer. It is important that this be established and that the fan power input be determined accordingly in order to provide a valid comparison of field test results to the fan performance ratings. In most cases, fan power input ratings do not include power transmission losses. 11.4.1 Estimating belt drive losses. In view of the lack of published information available for use in calculating belt drive losses, a graph is included in Annex L for this purpose. As indicated in the graph, belt drive loss, expressed as a percentage of motor power output, decreases with increasing motor power output and increases with increasing speed. This graph is based on the results of over 400 drive loss tests provided to AMCA by drive manufacturers. The graph serves as a reasonable guide in evaluating belt drive losses. The calculation of belt drive loss, using this graph, is included in many of the examples in Annex A. 13
AMCA 203-90 (R2007) 11.4.2 Estimating other transmission losses. For other types of power transmission equipment, consult the fan manufacturer to establish whether transmission losses are included in the fan ratings, and if so, request the magnitudes of the losses allowed in the ratings. Otherwise, it will be necessary to consult the manufacturer of the power transmission equipment for the information regarding transmission losses.
11.5 Accuracy The uncertainty analyses presented in Annex T indicate that the uncertainties in fan power input determinations are expected to range from 4% to 8%. This range is based on considerations of the conditions encountered in most field test situations, estimated accuracies of the various test methods presented in this publication and allowances for uncertainties in the determinations of power transmission losses.
12. Fan Speed 12.1 Speed measuring instruments Measure speed with a revolution counter and chronometer, a stroboscopic tachometer, an electronic counter-timer, or any other precision type tachometer which has a demonstrated accuracy of 0.5% of the measured value. Friction driven and magnetic type pickups should not be used in low fan power ranges where they can influence speed and fan power input measurements.
12.2 Speed measurements Establish the speed by averaging a minimum of three measurements made during the test determination period. The variation in the measurements should not exceed 1% for any single point of operation.
13. Densities 13.1 Locations of density determinations Determine the densities of the gas stream for Plane 1, the fan inlet; and for Plane 3, the velocity pressure measurement plane. In addition, the density at Plane 2, the fan outlet, must be determined whenever the fan total pressure, the fan velocity pressure, or an SEF at the outlet side of the fan is required.
13.2 Data required at each location The pressure and temperature of the gas stream must be obtained for each plane at which a density 14
determination is required. The pressures at Planes 1 and 2 are based on the static pressure measurements made for the purpose of determining the fan static pressure. The pressure at Plane 3 is obtained by averaging static pressure measurements made concurrent with the velocity pressure measurements made in a traverse of Plane 3. The absolute pressure at a plane is calculated by using the static pressure at the plane and the barometric pressure. For this reason, it is important that the barometric pressure be determined for the atmosphere to which static pressure measurements are referred. The temperatures used in density determinations are measured at the planes of interest.
13.3 Additional data Additional data required in the determination of density depends on the gas stream as indicated below: 1) For air, the wet-bulb temperature is required unless it is otherwise known that the air is saturated with water vapor or that the water vapor content of the air is insignificant. It should be noted that incorrect assumptions as to whether the air is dry or saturated can result in substantial errors in density determinations. 2) For gases other than air, the normal procedure is to rely on process personnel for the data necessary to determine the density of the gas. The information provided will include density or data sufficient to calculate the density, which should be for stated conditions of temperature and pressure.
13.4 Density values Gas stream density can be established when the pressure, temperature, and additional data, as indicated in Section 13.3, have been obtained. Procedures for establishing density are described in the examples in Annex M and are further illustrated in the field test examples in Annex A. Although the pressure and temperature of the gas stream must be obtained for each plane at which a density value is required, it is usually necessary to obtain additional data, such as the wet-bulb temperature, for only one plane in order to establish the densities at all planes. The densities at the planes for which the additional data is not obtained can be calculated, providing the gas stream does not change composition or undergo a change in phase between planes. The calculation is based on density being directly proportional to absolute pressure and
AMCA 203-90 (R2007) inversely proportional to absolute temperature. 13.4.1 Example calculation - ρ3 from ρ1. Use Figure N.1 of Annex N to establish the density of air at Plane 1 based on the test determinations of barometric pressure, pb, and the following Plane 1 values: Ps1, static pressure, in. wg td1, dry-bulb temperature, °F tw1, wet-bulb temperature, °F The following data are obtained for Plane 3: Ps3, static pressure, in. wg td3, dry-bulb temperature, °F Calculate the density at Plane 3 as follows: ⎛ P + 13.6 pb ⎞ ⎛ t d1 + 460 ⎞ ρ3 = ρ1 ⎜ s3 ⎟ ⎟⎜ ⎝ 13.6 p1 ⎠ ⎝ t d3 + 460 ⎠
thermometer should be accurate within 5°F of the measured value and readable to 5°F or finer. The temperature determination should be representative of the average temperature of the gas stream throughout the plane of interest. When the temperature varies with time or temperature stratification exists at the measurement plane, several temperature measurements may be necessary in order to obtain a representative average. At elevated temperatures, the thermometer may have to be shielded to prevent radiation effects from exposed heat sources. Locate the wet-bulb thermometer downstream from the dry-bulb thermometer in order to prevent the drybulb temperature measurement from being adversely affected. The wet-bulb thermometer wick should be clean, closely fitted, and wetted with fresh water. The velocity of the air over the wick should be between 700 and 2000 fpm. Use a sling psychrometer to obtain dry and wet-bulb air temperature measurements at the fan inlet for free inlet fans.
Where:
13.6 Barometric pressure p1 = the absolute pressure, in. Hg at Plane 1, calculated as follows: p1 = pb + (Ps1/13.6) In this manner, ρ3 can be calculated without having to measure the wet-bulb temperature at Plane 3. These equations can be used for gases other than air and can be adapted for use in calculations involving any two planes, subject to the limitations noted earlier. In the example calculation of ρ3, pb is determined for the atmosphere to which the measurements of Ps1 and Ps3 are referred. Refer static pressure measurements to a common atmosphere. When the pressures cannot be referred to a common atmosphere, the absolute pressure for each plane is calculated by using the static pressure measurement at the plane and the barometric pressure for the atmosphere to which the static pressure measurement is referred. However, for the purposes of accuracy, static pressure measurements that are used in the determination of fan static pressure must be referred to a common atmosphere.
13.5 Temperatures Measure temperatures with mercury-in-glass, dial, or thermocouple type thermometers. For temperatures through 220°F, the thermometer should be accurate within 2°F of the measured value and readable to 1°F or finer. For temperatures above 220°F, the
Use a portable aneroid barometer for field test determinations of barometric pressure when an acceptable site barometer is not available. The barometer should be accurate within 0.05 in. Hg of the measured value. Determine the test value of barometric pressure by averaging measurements made at the beginning and end of the test period. When the test value of barometric pressure is to be based on data obtained from a nearby airport, it is important that the data include the barometric pressure for the airport site and the elevation for which the pressure was determined (often the barometric pressure is corrected to sea level). This pressure value must then be corrected to the test site elevation. Barometric pressure decreases approximately 0.1 in. Hg for every 100 ft increase in elevation
13.7 Accuracy As indicated in Annex T, uncertainties in density determinations are expected to be less than 3%. However, care must be exercised in obtaining representative test measurements in order to prevent the uncertainties from exceeding this value.
14. Conversion Calculations Generally, the test fan will be operating at a speed and inlet density that are somewhat different from the 15
AMCA 203-90 (R2007) fan performance rating values of fan speed and inlet density. In order to provide a common basis for comparing the field test results to the fan performance ratings, each of these two items must be the same in both sets of data. This can be accomplished by converting the results of the field test to the speed and density conditions of the fan performance ratings. The equations for the conversion are as follows. Qc = Q (Nc / N) Psc = Ps (Nc / N)2 (ρc / ρ) Ptc = Pt (Nc / N)2 (ρc / ρ) Pvc = Pv (Nc / N)2 (ρc / ρ) Hc = H (Nc / N)3 (ρc / ρ) Where the subscript c designates values converted to specified conditions, and items without the subscript c are field test values. These conversion equations do not account for the effect of the compressibility of the gas stream. However, since the test fan usually operates at conditions of speed and inlet density that are reasonably close to the quoted fan performance, the conversion calculations usually result in small changes from field test values and the effect of the compressibility of the gas stream is considered to be negligible. Where test conditions are considerably different than design conditions, the effect of compressibility may need to be considered.
Work required to measurements (drilling installation of static thermometer wells, etc.) prior to the test date.
accommodate test of traverse holes, pressure taps and should be completed
4) System Effect Factors, if any, must be established prior to the conduct of the test. 5) The expected test uncertainties must be agreed upon prior to the test (see Annex T). 6) Responsibility for the cost of the test or any fansystem modifications required as a result of the test should be established. 7) Prior to testing, an inspection must be made to ensure that the fan is installed in accordance with the fan manufacturer’s recommendations. The duct system should also be inspected for compliance with design specifications, conditions of filters, abnormal duct restrictions, etc. 8) The majority of fan field performance tests cover a single point of operation, namely, the design duty. If it is deemed necessary to cover several points of operation, provision must be made in advance for changing the system resistance. The means used to vary the system resistance must not cause adverse flow conditions in the vicinities of the fan and measurement planes. If the fan cannot be tested at the quoted system design point, then it is sufficient for the evaluation of fan field performance to establish the proximity of the field test point to any portion of the fan performance rating curve within the limitations of the uncertainty analysis (see Annex T).
15. Test Preparation 15.1 The following items should be agreed upon by all interested parties prior to the start of a field performance test: 1) AMCA Publication 200, Air Systems, AMCA Publication 201, Fans and Systems, and AMCA Publication 202, Troubleshooting, should be reviewed and implemented before starting the field test. 2) Personnel conducting field tests on fans must be technically competent and fully conversant with all four parts of the AMCA Fan Application Manual. The person responsible for conducting the test should be designated and agreed upon by all parties. 3) The test instrumentation and locations of test measurement planes should be established. 16
9) It must be established that the system remains constant for the duration of the test. Modulating dampers should be set in a fixed position, no process changes shall be undertaken, etc. Variable inlet vane controls or inlet box dampers must be set in the full open position for the duration of the test, except when testing for control characteristics. 10) All precautions to ensure the safety of test personnel must be observed. 11) The fan-system should be operated for a length of time sufficient to ensure steady state conditions prior to the start of the test. 12) It is advisable that representatives of all parties interested in the test results be present at the time of the test to cover their areas of responsibility.
AMCA 203-90 (R2007)
15.2 It is recommended that as a minimum, the following equipment be taken to or be otherwise available at the job site: 1) Pitot-static tubes of suitable lengths for the maximum duct size to be traversed. Considerations should be given to the use of a double reverse tube in dirty atmospheres. 2) Manometers suitable for measuring static pressures. Manometer fluids other than water are acceptable, provided the specific gravity is known. A spare bottle of manometer fluid is advisable. 3) Inclined manometer suitable for measuring velocity pressures. 4) Flexible tubing of suitable length to enable manometers to be installed at a convenient location. 5) Tubing couplings and “T” type tubing connectors. 6) Thermometers to cover the range of anticipated temperatures. 7) Sling psychrometer for obtaining dry-bulb and wet-bulb temperatures. 8) Clip-on ammeter-voltmeter, power analyzer, or other suitable electrical measurement instruments for the determination of fan power input. 9) Fan speed measurement instrument. 10) Aneroid barometer. 11) Flashlight, tape, measuring rule, hand tools, coveralls, etc. 12) Test data sheets, calculator, and necessary drawings. 13) Complete AMCA Fan Application Manual containing Publications 200, 201, 202, and 203.
2) Static and total pressure manometer tubing must be “pinched off” prior to inserting or removing the Pitot-static tube from the test duct. Release both legs of the tubing simultaneously after the Pitotstatic tube is inside the test duct and properly oriented. Failure to release simultaneously may result in manometer fluid being blown from the manometer. 3) Loop the manometer tubing well above the manometer so that any fluid which is inadvertently blown from the gauge will drain back into the manometer. 4) The Pitot-static tube is intended for measuring pressures in relatively clean gases. When using Pitot-static tubes in dirty, wet, or corrosive atmospheres, both legs of the Pitot-static tube must be cleaned out frequently during the test. Since fan pressure readings are never strictly steady, absence of fluctuations is an indication of a plugged Pitot-static tube. Consider using a double reverse tube in these situations. 5) When making measurements in wet gas streams, continually check for the presence of moisture in the tubing. Clear plastic tubing is ideal from this standpoint. If moisture collects in the tubing, immediately remove the Pitot-static tube and clean the inside of the tubing and Pitotstatic tube before proceeding with the test. 6) Before performing any work inside a fan, ductwork, or other system components, make certain that the fan motor starter is “locked out.” 7) The area at the plane of flow measurement should be measured internally to account for internal insulation or other obstructions. 8) Do not rely on damper control indicators to ensure that dampers are fully open. Check visually. 9) Measure temperatures on both sides of double inlet fans as temperature differences may exist between each side.
16. Precautions The following precautions should be observed when conducting a field test: 1) Connect the Pitot-static tube to the manometers according to anticipated pressures, i.e., whether the pressures are positive or negative, and the magnitudes of pressures.
10) When measuring in high temperature, corrosive or explosive atmospheres, instruments should be selected for suitability for such atmospheres.
17. Typical Fan-System Installations A fan assembly may include any number of appurtenances: variable inlet vanes, inlet boxes, inlet 17
AMCA 203-90 (R2007) box dampers, outlet dampers, inlet screens, belt guards, inlet bells, diffusers (evasés). Alternately, these items may be included in the fan-system installation, but not be a part of the fan assembly. In order to determine the proper field test procedure and to provide a valid basis for comparing field test results to the fan performance ratings, it is important to establish which of these items are considered a part of the fan and which are considered a part of the system. The fan performance ratings may be assumed to include the appurtenances that are established as being a part of the fan assembly. The locations of the fan inlet and fan outlet depend on whether specific appurtenances are considered to be a part of the fan assembly. If the assembly includes an inlet box, the fan inlet is the inlet to the inlet box. For a fan assembly that includes a diffuser, the fan outlet is the outlet of the diffuser. In the case of heating, ventilating, and airconditioning equipment, the field test procedure will depend on whether the equipment is a factory assembled central station unit, a built-up unit, or a packaged unit (see Section 17.4). The performance ratings for a fan that includes inlet box dampers, variable inlet vanes or outlet dampers cover operation of the fan with these items in the full open positions. In order to be able to compare the field test results to the fan performance ratings, it is essential that these items be fixed in their full open positions for the duration of the test. In addition, when the loss through a damper must be calculated, it is essential that the damper blades be fixed in their full open positions during the test since this is the condition on which the damper pressure loss ratings are based. This consideration arises when a damper, which is not considered a part of the fan is located between a static pressure measurement plane and the fan. In order to determine the fan static pressure, the loss through the damper must be calculated. In these cases, the calculation of the loss is based on the performance ratings for the damper.
a) The operations of ovens, furnaces, paint booths, air conditioning equipment, other fans, and similar items that may supply or exhaust air from the building in intermittent or modulating fashions. b) The use of doors providing access to the building. The effect is most significant when large doors that are normally closed are kept open for extended periods such as in loading operations. c) The velocity and direction of the wind outside the building, particularly in conjunction with the item immediately above and as it may affect the flow of air from the outlet of the ventilator. d) The use of interior doors that my restrict the flow of air from areas normally expected to be ventilated. Assuming that these difficulties can be resolved and the desired system is fixed for the duration of the test, determine the fan performance by using one of the following methods: 1) Make field test measurements sufficient for determining fan static pressure, fan power input, fan speed, and the density of the air at the fan inlet. In this method for testing a free inlet, free outlet fan, the fan static pressure is calculated as the static pressure on the outlet side of the fan less the static pressure on the inlet side of the fan: Ps = Ps2 - Ps1. The static pressure measurements involved must be referred to the same atmospheric pressure and made at locations sufficiently distant from the fan inlet and outlet so as to be unaffected by the velocity of the air entering and leaving the fan. Using the fan manufacturer’s certified performance ratings, draw a performance curve for the fan for operation at the test values of fan speed and entering air density. Determine the fan air flow rate by entering this curve at the test values of fan static pressure and fan power input (see Example 5C in Annex A).
17.1 Free inlet, free outlet fans It is difficult to achieve an accurate field test of a free inlet, free outlet fan. The most obvious problem is the lack of a suitable location for the velocity pressure measurement plane. In addition, in the case of ventilators that supply or exhaust air from a buildingthe most commonly encountered applications of free inlet, free outlet fans-it is extremely difficult to define, set, and maintain for the duration of the test the “normal” system condition. Items affecting the system include:
18
2) Use the method as described above with the exception that the performance curve is established by a laboratory test of the fan, conducted in accordance with AMCA Standard 210. For the laboratory test, the fan must be set up in a manner that duplicates the field installation conditions. That is, all appurtenances must be in place and any restrictions or obstructions to the free flow of air into the fan inlet and away from the fan outlet must be accurately duplicated in the laboratory test setup.
AMCA 203-90 (R2007) 3) Install a duct on the inlet side of the fan for the purpose of providing a location for the velocity pressure measurement plane. All of the test measurements and calculations in this method for testing a free inlet, free outlet fan are the same as those required for a fan with a ducted inlet and a free outlet. The cross-sectional shape and area of the duct, which is temporarily installed for purposes of the test, should be selected on the basis of minimizing its interference with the flow of air into the fan inlet while providing velocity pressure of magnitudes that can be accurately measured. The length of the duct should be a minimum of twice its diameter or equivalent diameter, and the entrance to the duct should be flared in order to reduce the entrance loss. The velocity pressure measurement plane should be located a minimum of 1.5 diameters or equivalent diameters downstream from the duct inlet. The effect of this duct on the system is negligible, but in changing the pattern of the flow of air into the fan inlet, it may affect the performance of the fan slightly. Applications of this method of test are presented in Examples 5A and 5B in Annex A. The equation for calculating fan static pressure for this configuration is: Ps = Ps2 - (Ps1 + Pv1)
17.2 Free inlet, ducted outlet fans In the calculation of fan static pressure for this type of fan-system configuration, the sum of the static pressure at the fan inlet, Ps1, and the velocity pressure at the fan inlet, Pv1, is considered to be equal to the sum of the static pressure, Psx, and the velocity pressure, Pvx, at a point sufficiently distant from the fan inlet as to be in still air. At this point, the static pressure is zero, and the velocity pressure in still air is zero. Ps1 + Pv1 = Psx + Pvx = 0 This consideration, which is the same as that used in the methods for testing fans for performance rating purposes, charges to the fan the losses incurred in accelerating the air into the fan inlet and eliminates inaccuracies which may occur in any attempt to measure velocity pressure and static pressure at the fan inlet. Since Ps1 + Pv1 = 0, the equation for calculating fan static pressure for this configuration is:
Ps = Ps2 + SEF 1 +SEF 2 + ... + SEF n
17.3 Ducted inlet, ducted outlet fans In this type of fan-system configuration, there is no special consideration in the calculation of fan static pressure. The equation for this calculation is: Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2 + ... + SEF n In this configuration, the flow conditions on the inlet side of the fan are usually more favorable for the location of the velocity pressure measurement plane.
17.4 Ducted inlet, free outlet fans In this type of fan-system configuration, the static pressure at the fan outlet, Ps2, is zero gauge pressure, referred to the atmospheric pressure in the region of the fan outlet. However, the gas stream may be discharging from the fan into a region in which the atmospheric pressure is somewhat different from that to which all other pressure measurements are referred. When this possibility exists, it is essential that the static pressure measurements in the region of the fan outlet be referred to the same atmospheric pressure as used in all other pressure measurements. Ps = -Ps1 - Pv1 + SEF 1 + SEF 2 + ... + SEF n
17.5 Air handling units This category consists of draw-through and blowthrough types of equipment assemblies used in heating, ventilating, and air-conditioning applications. In addition to fans, these equipment assemblies may include any number of combinations of coils, filters, access sections, humidifiers, mixing boxes, dampers, etc. Air handling units include packaged units, factory assembled central station units, and built-up units. The basis used in establishing the air performance ratings for each of these unit types is described below. It is important that the field test method correspond to the rating method in each case. 17.5.1 Packaged units. This type of unit is supplied and rated by the manufacturer as an assembly. The static pressures at the inlet and outlet to the assembly and the velocity pressure at the inlet to the assembly are used in calculating the static pressure generated by this type of air handling unit. See Examples 4C and 4D in Annex A. 17.5.2 Factory assembled central station units. The air performance ratings for this type of unit are based on the operation of the fan section assembly only, but include the effects of the air flow conditions 19
AMCA 203-90 (R2007) entering and leaving the fan section which are created by accessory equipment such as plenums, coils, filters, mixing boxes, etc. The fan section assembly includes the fan and the cabinet in which the fan has been installed. The accessory items are considered to be included in the system in which the fan section operates. The static pressure and the velocity pressure at the inlet of the fan section and the static pressure at the fan section outlet, which coincides with the fan outlet, are used in calculating the static pressure generated by the fan section assembly. See examples 4B and 4E in Annex A. 17.5.3 Built-up units. Built-up units are similar to factory assembled central station units, except that in built-up units, the components are normally obtained from a number of equipment suppliers and the unit is assembled at the installation site. The fans which are used in built-up units are rated as free-standing, unencumbered by the cabinets in which they are installed. In the field test determination of the performance of the fan, the static pressure and velocity pressure at the fan inlet and the static pressure at the fan outlet are used in calculating the fan static pressure. An SEF that accounts for the effect of the cabinet is normally included in this calculation, and it may be necessary to include an SEF to account for the conditions at the fan outlet. See Example 4A in Annex A.
20
AMCA 203-90 (R2007)
Annex A. Field Test Examples This annex contains examples of field tests. The examples are presented in detail and cover several types of fansystem combinations. Field test procedures are illustrated in a variety of situations. Portions of the procedures are typical for all fan-system installations. Other portions of the procedures demonstrate methods for dealing with the more difficult features encountered in some installations. Not all of the possible fan-system combinations are included in the examples, but it is expected that the examples will provide sufficient guidance for dealing with those cases not covered.
EXAMPLES OF FANS, INSTALLATION TYPE B: FREE INLET, DUCTED OUTLET 1A: 1B: 1C: 1D:
Centrifugal Forced Draft Fan Centrifugal Forced Draft Fan with Inlet Silencers Axial Forced Draft Fan with Inlet Silencers Centrifugal Fans in Parallel
EXAMPLE OF FANS, INSTALLATION TYPE D: DUCTED INLET, DUCTED OUTLET 2A: 2B: 2C: 2D: 2E: 2F: 2G:
Utility Fan in a Ventilating System Centrifugal Fan in a Sawdust Conveying System Axial Fan in a Dryer System Centrifugal Fan in a Scrubber System Centrifugal Fan in a Process System Axial Fan in a Ventilation System High Pressure Centrifugal Fans in Series
EXAMPLES OF FANS, INSTALLATION TYPE C: DUCTED INLET, FREE OUTLET 3A: 3B: 3C: 3D:
Centrifugal Fan in an Exhaust System Axial Fan in an Exhaust System Centrifugal Fan in a Scrubber System Centrifugal Roof Ventilator with Ducted Inlet
EXAMPLES OF AIR HANDLING UNITS 4A: 4B: 4C: 4D: 4E:
Centrifugal Fan in a Built-up Air conditioning Unit Central Station Air Conditioning Unit, Factory Assembled Draw-Through Type Packaged Air Conditioning Unit Packaged Air Conditioning Unit Central Station Air Conditioning Unit, Factory Assembled Blow-Through Type
EXAMPLES OF FANS, INSTALLATION TYPE A: FREE INLET, FREE OUTLET 5A: 5B: 5C:
Free Inlet, Free Outlet Roof Ventilator with temporary duct Free Inlet, Free Outlet Propeller Fan with temporary duct Free Inlet, Free Outlet Roof Ventilator as installed
21
AMCA 203-90 (R2007)
EXAMPLE 1A: CENTRIFUGAL FORCED DRAFT FAN
SEF 1 DIFFUSER
3
2
L
A2
VARIABLE INLET VANES SIDE VIEW
A3 OUTLET SIDE VIEW
LOCATIONS OF PLANES 2 AND 3
ORIENTATION OF PITOT TUBE COMMENTS 1. The variable inlet vanes are considered part of the fan. Performance ratings for fans with inlet vanes cover operation with the inlet vanes in their full open position. In order to be able to compare the test results to the fan performance ratings, it is essential that the inlet vanes be fixed in their full open positions for the duration of the test. 2. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located near the end of the fan diffuser (evasé). Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for the traverse are described in Section 9.4. These velocity pressure and static pressure measurements are susceptible to error due to the turbulence existing in the region of the fan outlet. In addition, it is undesirable to have Plane 3 located in a diverging airway. However, no other more suitable location for Plane 3 exists in this example. It is recommended that the Pitot-static tube be oriented so that its nose is aligned with the anticipated flow streams, particularly near the walls of the diffuser, as shown in the diagram. Determine the area of the traverse plane, A3, which is located at the tip of the Pitot-static tube, as shown in the diagram, not at the location of the Pitot-static tube access holes in the diffuser.
22
3. Measure td1 and tw1 in the path of the air flowing into the fan inlets. Determine pb for the general vicinity of the fan. Measure td3 in Plane 3. All of these measurements are used in the determination of densities at the various planes of interest. 4. Measure the fan speed and the motor amps, volts, and, if possible, watts. Record all pertinent motor nameplate data, including volts (NPV) and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point (refer to Annex K). 5. SEF 1 is due to the effect of insufficient length of duct at the fan outlet. In order to calculate the value of SEF 1, it is necessary to measure the length of the outlet duct, L; the outlet area of the fan, A2; and the blast area of the fan. 6. The sum of the static pressure, Ps1, and velocity pressure, Pv1, at the inlets of a fan with unrestricted inlets is considered to be equal to the sum of the static pressure, Psx, and the velocity pressure, Pvx, at a point sufficiently distant from the fan inlets as to be in still air. At this point, the static pressure is zero, and
AMCA 203-90 (R2007) the velocity pressure in still air is zero.
GENERAL
Ps1 + Pv1 = Psx + Pvx = 0
VIVs in full open positions. Fan direct connected to motor.
This consideration, which is the same as that used in the methods for testing fans for performance rating purposes, charges to the fan losses incurred in accelerating the air into the fan inlets and eliminates the inaccuracies which arise in any attempt to measure the velocity pressure and static pressure at the fan inlets. To calculate the fan static pressure:
DENSITIES
Ps = Ps2 - Ps1 - Pv1 + SEF 1 = Ps2 - (Ps1 + Pv1) + SEF1
td1 = tw1 = p1 = =
CALCULATIONS
For fan inlet conditions of: 85°F 63°F pb 28.91 in. Hg
Since: Ps1 + Pv1 = 0 Ps = Ps2 + SEF 1
Use Figure N.1 in Annex N to obtain ρ1 = 0.0701 lbm/ft3 The density at Plane 3:
7. In order to compare the test results to the quoted fan curve drawn for operation at 1780 rpm and 0.0701 lbm/ft3 density, it is necessary to convert the results to the specified conditions. In this case, the test conditions are identical to the specified conditions and no calculations are required.
OBSERVATIONS
⎛ P + 13.6 pb ⎞ ⎛ t d1 + 460 ⎞ ρ3 = ρ1 ⎜ s3 ⎟ ⎟⎜ ⎝ 13.6 p1 ⎠ ⎝ t d3 + 460 ⎠ ⎛ 14.4 + 13.6 × 28.91 ⎞ ⎛ 545 ⎞ = 0.0701⎜ 13.6 × 28.91 ⎟⎠ ⎜⎝ 556 ⎟⎠ ⎝ = 0.0712 lbm/ft 3
SITE MEASUREMENTS
In this case, ρ2 is considered to be equal to ρ3.
pb = 28.91 in. Hg td1 = 85°F tw1 = 63°F td3 = 96°F Ps3 = 14.4 in. wg Pv3 = 1.52 in. wg N = 1780 rpm A2 = 11.94 ft2 A3 = 11.3 ft2 Blast Area = 7.76 ft2 L = 3 ft.
FLOW RATES
MEASURED MOTOR DATA Volts = = Amps = =
570, 560, 572 567 av. 160, 166, 163 163 av.
MOTOR NAMEPLATE DATA 200 hp, 3 phase, 60 hertz 575 volts, 1800 rpm, 181 FLA
V3 = 1096 (Pv3/ρ3)0.5 = 1096 (1.52/0.0712)0.5 = 5064 fpm Q3 = V3A3 = 5064 × 11.3 = 57223 cfm Q = = = =
Q1 Q3 (ρ3/ρ1) 57223 (0.0712/0.0701) 58121 cfm
FAN POWER INPUT Measured amps/FLA = (163/181) = 0.90 = 90% Annex K indicates that Equation A will provide a reasonably accurate estimate of motor power output for a 200 hp motor operating at 90% FLA.
23
AMCA 203-90 (R2007) Hmo = 200 (163/181) (567/575) = 178 hp Since the fan is direct connected to the motor: H
= Hmo = 178 hp
SYSTEM EFFECT FACTOR AMCA Publication 201-90, Figures 7.1 and 8.3 indicate the following calculations: Q2 = Q3 (ρ3/ρ2) = 57223 (0.0712/0.0712) = 57223 cfm V2 = (Q2/A2) = (57223/11.94) = 4793 fpm Duct diameter equivalent to the diffuser outlet area: De2 = 4 A2 / π =
( 4 × 11.94 ) / π
= 3.9 ft. Figure 8.3 shows that for velocities over 2500 fpm, 100% effective duct length is one duct diameter per 1000 fpm, = De2 (V2/1000) = 3.9 (4793/1000) = 18.7 ft L in % effective duct length = (L/18.7) 100 = (3/18.7) 100 = 16% Blast area ratio = Blast Area/A2 = 7.76/11.94 = 0.65 For blast area ratio of 0.65, and 16% effective duct length, Figure 8.3 shows System Effect Curve U applies. For 4793 fpm velocity and curve U, Figure 7.1 shows SEF 1 = 0.6 in. wg at 0.075 lbm/ft3. At 0.0712 lbm/ft3. SEF 1 = 0.6 (0.0712/0.075) = 0.57 in. wg
24
FAN STATIC PRESSURE Since A2 is greater than A3, there may be some conversion of velocity pressure to static pressure between Planes 3 and 2. However, the amount of conversion will be very small relative to the static pressure measured at Plane 3 and ignoring any change in static pressure from Plane 3 to Plane 2 will have no appreciable effect on the test results. Therefore, Ps2 is considered equal to Ps3. Ps = Ps2 + SEF 1 = 14.4 + 0.57 = 14.97 in. wg CONVERSION TO SPECIFIED CONDITIONS Qc = = Psc = = Hc = =
Q 58121 cfm Ps 14.97 in. wg H 178 hp
AMCA 203-90 (R2007)
EXAMPLE 1B: CENTRIFUGAL FORCED DRAFT FAN WITH INLET SILENCERS
TEMPORARY DUCT
DIFFUSER STATIC PRESSURE TAPS
3a
0.5 De
3b SILENCERS
1 A2
SEF 1
SIDE VIEW
VARIABLE INLET VANES
OUTLET SIDE VIEW
2 COMMENTS
1. This fan, as supplied and rated by the manufacturer, includes the variable inlet vanes and inlet boxes, but does not include the silencers. Performance ratings for fans with inlet vanes cover operation with the inlet vanes in the full open positions. In order to be able to compare the test results to the fan performance ratings, it is essential that the inlet vanes be fixed in their full open positions for the duration of the test. 2. Determine Pv3a and Pv3b by using the root mean square of the velocity pressure measurements made in traverses of Planes 3a and 3b. A3a and A3b are the areas traversed. Determine Ps3a and Ps3b by averaging each of the two sets of static pressure measurements made in the same traverses. Procedures for traverses are described in Section 9.4. Ps3a and Ps3b are used in determining the density at the traverse plane. A location for Plane 3 measurements may be obtained by installing ducts on each silencer inlet, as shown in the diagram. The ducts should be a minimum of one equivalent diameter in length, and have flared inlets to reduce entrance losses and provide more uniform velocity profiles at the pressure measurement planes. 3. Measure Ps1a and Ps1b at locations close to the entrances to the inlet boxes and in planes which are substantially equal in area to the planes of the
entrances to the inlet boxes (Plane 1). Determine Ps2 by averaging the pressure measurements at each of four static pressure taps located near the end of the fan diffuser (evasé). See Annex E for details of static pressure taps. 4. Measure td3 and tw3 near the inlet ducts. Determine pb for the general vicinity of the fan. Measure td2 in Plane 2. All of these measurements are used in the determination of densities at the various planes of interest. 5. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV) and amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K. 6. SEF 1 is due to the effect of there being no duct at the fan outlet. In order to calculate the value of SEF 1, it is necessary to measure the fan outlet area, A2, and the blast area of the fan. 7. To calculate the fan static pressure: 25
AMCA 203-90 (R2007) Ps = Ps2 - Ps1 - Pv1 + SEF 1
CALCULATIONS
Where:
DENSITIES
Pv1 = (Q/1096A1)2 ρ1
For Plane 3 conditions of:
8. In order to compare the test results to the quoted fan curve drawn for operation at 1180 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14.
td3 = 85°F tw3 = 58°F
OBSERVATIONS SITE MEASUREMENTS pb td2 td3 tw3 Ps1a Ps1b Ps2 Ps3a Ps3b Pv3a Pv3b N A1a
= 29.31 in. Hg = 93°F = 85°F = 58°F = -1.20 in. wg = -1.30 in. wg = 10.1 in. wg = -0.65 in. wg = -0.70 in. wg = 0.61 in. wg = 0.62 in. wg = 1180 rpm = A1b = 12.5 ft2 A2 = 18 ft2 A3a = A3b = 12.5 ft2 Blast Area = 13.5 ft2 MEASURED MOTOR DATA Volts = = Amps = =
460, 455, 465 460 av 257, 256, 258 257 av
MOTOR NAMEPLATE DATA 200 HP, 3 phase 60 hertz 460 volts, 1180 rpm, 285 FLA GENERAL VIVs in full open positions. Fan direct connected to motor. The motor manufacturer advises that this motor type has a peak efficiency of 91% at a power factor of approximately 0.89.
Ps3 = = = p3 = = =
(Ps3a + Ps3b)/2 (-0.65 - 0.70)/2 -0.675 in. wg pb + (Ps3/13.6) 29.31 + (-0.675/13.6) 29.26 in. Hg
Use Figure N.1 in Annex N to obtain ρ3 = 0.0712 lbm/ft3 It is assumed that the temperature at Plane 1 are the same as those at Plane 3. The density at Plane 1: ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ1 = ρ3 ⎜ s1 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d1 + 460 ⎠ ⎛ −1.25 + 13.6 × 29.31 ⎞ ⎛ 545 ⎞ = 0.0712 ⎜ ⎟ ⎜ 545 ⎟ 13.6 × 29.26 ⎝ ⎠⎝ ⎠ = 0.0711 lbm/ft 3 The density at Plane 2: ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ2 = ρ3 ⎜ s2 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d2 + 460 ⎠ ⎛ 10.1 + 13.6 × 29.31 ⎞ ⎛ 545 ⎞ = 0.0712 ⎜ ⎟⎜ ⎟ ⎝ 13.6 × 29.26 ⎠ ⎝ 553 ⎠ = 0.0721 lbm/ft 3 FLOW RATES V3a = 1096 (Pv3a/ρ3)0.5 = 1096 (0.61/0.0712)0.5 = 3208 fpm Q3a = V3aA3a = 3208 × 12.5 = 40100 cfm V3b = 1096 (Pv3b/ρ3)0.5 = 1096 (0.62/0.0712)0.5 = 3234 cfm Q3b = V3bA3b = 3234 × 12.5 = 40425 cfm
26
AMCA 203-90 (R2007) Q3 = Q3a + Q3b = 40100 + 40425 = 80525 cfm Q = = = =
Q1 Q3 (ρ3/ρ1) 80525 (0.0712/0.0711) 80638 cfm
SYSTEM EFFECT FACTOR AMCA Publication 201-90, Figures 7.1 and 8.3 indicate the following calculations: Q3 (ρ3/ρ2) = 80525 (0.0712/0.0721) = 79520 cfm (Q2/A2)
FAN POWER INPUT
= (79520/18) = 4418 fpm
Measured amps/FLA = (257/285) = 0.90 = 90%
Blast area ratio = Blast Area/A2 = 13.5/18 = 0.75
Annex K indicates that Equation A will provide a reasonably accurate estimate of motor power output for a 250 hp motor operating at 90% FLA.
For a blast area ratio of 0.75, and no duct, Figure 8.3 shows System Effect Curve T applies. For 4418 fpm velocity and curve T, Figure 7.1 shows SEF 1 = 0.65 in. wg at 0.075 lbm/ft3. At 0.0720 lbm/ft3:
Hmo = 250 (257/285) (460/460) = 225 hp As a check of this value, using the motor efficiency data and the appropriate equation in Section 11.2.2: 3 × 257 × 460 × 0.89 × 0.91 746 = 222 hp
Hmo =
Since the motor is not fully loaded, the power factor and efficiency may be less, which would reduce Hmo as calculated using the second method. However, this is a reasonable check. The value of Hmo is selected to be the average of the two results:
SEF 1 = 0.65 (0.0721/0.075) = 0.62 in. wg FAN STATIC PRESSURE Pv1 = (Q1/1096 A1)2 = (80638/1096 × 25)2 0.0711 = 0.62 in. wg Ps = Ps2 - Ps1 - Pv1 + SEF 1 = 10.1 - (-1.25) - 0.62 + 0.62 = 11.33 in. wg CONVERSION TO SPECIFIED CONDITIONS
Hmo = 224 hp
Qc = Q = 80638 cfm
Since the fan is direct-connected to the motor, there is no drive loss, and:
Psc = 11.33 (0.075/0.0711) = 11.95 in. wg
H = Hmo = 224 hp
Hc = 224 (0.075/0.0711) = 236 hp
27
AMCA 203-90 (R2007)
EXAMPLE 1C: AXIAL FORCED DRAFT FAN WITH INLET SILENCER
PLANE 3 LOCATION 3
TEMPORARY SHORT DUCT STATIC PRESSURE TAPS SILENCER
TRANSITION
0.5 De INLET BOX 1
DIFFUSER SECTION INNER CYLINDER
5 2
L
SIDE VIEW
GUIDE VANES COMMENTS 1. This is a variable pitch axial flow fan. The fan assembly, as supplied and rated by the manufacturer, includes the inlet box and diffuser section, but does not include the silencer. It is essential that the blade pitch angle be fixed for the duration of the test. This blade angle should be agreed upon by all interested parties. 2. A temporary short duct is installed upstream of the silencer to establish Plane 3 in which more uniform pressures can be obtained. The duct should be a minimum of one equivalent diameter in length, and have a flared inlet to reduce entrance losses and provide a more uniform velocity profile at the pressure measurement plane. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3. Ps3 is determined by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. Ps3 is used in determining the density at the traverse plane. 3. Measure Ps1 at a location close to the entrance to the inlet box and in a plane which is substantially equal in area to the plane of the entrance to the inlet box (Plane 1). Determine Ps5 by averaging the pressure measurements at each of four static pressure taps located near the end of the fan diffuser. 28
See Annex E for details of static pressure taps. In this example, Ps2 is considered to be equal to Ps5. 4. Measure td3 and tw3 near the entrance to the short inlet duct. Determine pb for the general vicinity of the fan. Measure td5 in Plane 5. All of these measurements are used in the determination of densities at the various planes of interest. 5. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV) and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Motor performance data, supplied by the motor manufacturer, are used in the determination of motor power output for this example. 6. SEF 1 is due to the effect of insufficient length of duct between the diffuser outlet and the elbow downstream of the diffuser. In order to calculate the value of SEF 1, it is necessary to measure the length of the transition, L, and the outlet area of the diffuser, A2.
AMCA 203-90 (R2007) CALCULATIONS
7. To calculate the Fan Static Pressure: Ps = Ps2 - Ps1 - Pv1 + SEF 1
DENSITIES
Where:
For Plane 3 conditions of:
Pv1 = Pv3 (A3/A1)2 (ρ3/ρ1)
td3 = 68°F tw3 = 62°F
8. Axial fans are often rated in Fan Total Pressure. Computation of Fan Total Pressure is illustrated in the CALCULATIONS section of this example. 9. In order to compare the test results to the quoted fan curve drawn for operation at 880 rpm and 0.0740 lbm/ft3 density, it is necessary to convert the results to the specified conditions. In this case, the test conditions are identical to the specified conditions and no calculations are required. OBSERVATIONS SITE MEASUREMENTS pb td3 tw3 td5 Ps1 Ps3 Ps5 Pv3 N A1 A2 A3 A5 L
= 29.8 in. Hg = 68°F = 62°F = 88°F = -1.80 in. wg = -1.40 in. wg = 20.8 in. wg = 1.30 in. wg = 880 rpm = 170.3 ft2 = 176 ft2 = 170.3 ft2 = A2 = 15 ft
MEASURED MOTOR DATA Volts = = Amps = =
4000, 4000, 4100 4033 av 450, 445, 448 448 av
MOTOR NAMEPLATE DATA 4000 hp, 3 phase 60 hertz 4000 volts, 900 rpm, 520 FLA GENERAL
p3 = pb + (Ps3/13.6) = 29.8 + (-1.40/13.6) = 29.70 in. Hg Use Figure 20 in Annex N to obtain ρ3 = 0.0744 lbm/ft3 It is assumed that td1 = td3. The density at Plane 1: ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ1 = ρ3 ⎜ s1 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d1 + 460 ⎠ ⎛ −1.8 + 13.6 × 29.8 ⎞ ⎛ 528 ⎞ = 0.0744 ⎜ ⎟⎜ ⎟ ⎝ 13.6 × 29.70 ⎠ ⎝ 528 ⎠ = 0.0743 lbm/ft 3 The density at Plane 2:
ρ 2 = ρ5 ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ = ρ3 ⎜ s5 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d5 + 460 ⎠ ⎛ 20.8 + 13.6 × 29.8 ⎞ ⎛ 528 ⎞ = 0.0744 ⎜ ⎟⎜ ⎟ ⎝ 13.6 × 29.70 ⎠ ⎝ 548 ⎠ = 0.0756 lbm/ft 3 FLOW RATE V3 = 1096 (Pv3/ρ3)0.5 = 1096 (1.3/0.0744)0.5 = 4581 fpm Q3 = V3A3 = 4581 × 170.3 = 780144 cfm Q = = = =
Q1 Q3 (ρ3/ρ1) 780144 (0.0744/0.0743) 781194 cfm
Fan direct connected to motor. Motor performance data at operating load, as supplied by motor manufacturer: 0.88 power factor, 95% efficiency.
29
AMCA 203-90 (R2007) FAN POWER INPUT Hmo =
3 × volts × amps × power factor × efficiency 746
3 × 4033 × 448 × 0.88 × 0.95 746 = 3507 hp =
SEF 1 = 0.32 (0.0756/0.075) = 0.32 in. wg FAN STATIC PRESSURE Pv1 = Pv3 (A3/A1)2 (ρ3/ρ1) = 1.3 (170.3/170.3)2 (0.0744/0.0743) = 1.30 in. wg
Since the fan is direct connected to the motor, there is no drive loss, and:
Ps2 = Ps5 = 20.8 in. wg
H = Hmo = 3507 hp
Ps = Ps2 - Ps1 - Pv1 + SEF 1 = 20.8 - (-1.80) - 1.30 + 0.32 = 21.62 in. wg
SYSTEM EFFECT FACTOR FAN TOTAL PRESSURE AMCA Publication 201-90, Figures 7.1, 8.1, and 8.4 indicate the following calculations: Q2 = Q3 (ρ3/ρ2) = 780144 (0.0744/0.0756) = 767761 cfm V2 = (Q2/A2) = (767761/176) = 4362 Duct diameter equivalent to the diffuser outlet area: De2 = 4 A2 / π =
( 4 × 176 ) / π
= 15 ft. Figure 8.1 shows that for velocities over 2500 fpm, 100% effective duct length is one duct diameter for every 1000 fpm: = De2 (V2/1000) = 15 (4362/1000) = 65.43 ft. L in % effective duct length = (L/65.43) 100 = (15/65.43) 100 = 23% For 23% effective duct length and a vaneaxial fan with a 2 piece elbow, Figure 8.4 shows System Effect Curve V applies. For 4362 fpm velocity and curve V, Figure 7.1 shows SEF 1 = 0.32 in. wg at 0.075 lbm/ft3. At 0.0756 lbm/ft3.
30
Pt1 = Ps1 +Pv1 = -1.8 + 1.30 = -0.50 in. wg Pv2 = Pv3 (A3/A2)2 (ρ3/ρ2) = 1.3 (170.3/176)2 (0.0744/0.0756) = 1.20 in. wg Pt2 = Ps2 + Pv2 = 20.8 + 1.20 = 22.00 in. wg Pt = Pt2 - Pt1 + SEF 1 = 22.00 - (-0.50) + 0.32 = 22.82 in. wg Also: Pt = Pv = = Pt = =
Ps + Pv Pv2 1.20 in. wg 21.62 + 1.20 22.82 in. wg
CONVERSION TO SPECIFIED CONDITIONS Qc = = Psc = = Ptc = = Hc = =
Q 781194 cfm Ps 21.62 in. wg Pt 22.82 in. wg H 3507 hp
AMCA 203-90 (R2007)
EXAMPLE 1D: CENTRIFUGAL FANS IN PARALLEL
3
STATIC PRESSURE TAPS OUTLET DAMPER
2
SEF 1 PLENUM 1
PLAN VIEW
1
SIDE VIEW
COMMENTS 1. Each of the fans, as supplied and rated by the manufacturer, includes an outlet damper. Performance ratings for fans with outlet dampers cover operation with the outlet damper in the full open position. In order to be able to compare the test results to the fan performance ratings it is essential that the outlet dampers be fixed in the full open positions for the duration of the test. 2. In this example, there are no suitable locations for traverse planes for use in determining directly the flow rate for each fan. The alternative is to determine the total flow rate and since the fans and their operating speeds are alike, assume that each fan delivers a flow rate proportional to its actual speed. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located near the end of a straight run of duct, such as shown in the diagram. Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. Ps3 is used in determining the density at the traverse plane. Measure the area of traverse plane, A3, which is located at the tip of the Pitot-static tube. 3. Determine Ps2 for each fan by averaging the pressure measurements at each of four static pressure taps located in the short length of duct
between the outlet damper and the plenum. See Annex E for details of static pressure taps. Measure td2 in Plane 2 for each fan. 4. For each fan, measure td1 and tw1 in the path of the air flowing into the fan inlet. Determine pb for the general vicinity of the fans. Measure td3 in Plane 3. All of these measurements are used in the determination of densities at the various planes of interest. 5. Measure the fan speed and the motor amps, volts, and if possible, watts for each fan. Record all pertinent motor nameplate data, including volts (NPV) and full load amps (FLA). If the motor power outputs are to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drives and measure the no load amps (NLA) if the motors are not operating at or near their full load points. Refer to Annex K. 6. SEF 1 is due to the effect of insufficient length of duct between the outlet of each fan and the plenum. In this case, the duct length is so short as to be judged equivalent to there being no duct at all. In order to calculate the value of SEF 1, it is necessary to measure the outlet areas of the fans, A2, and their blast areas. 31
AMCA 203-90 (R2007)
7. The sum of the static pressure, Ps1, and the velocity pressure, Pv1, at the inlet of a fan with an unrestricted inlet is considered to be equal to the sum of the static pressure, Psx, and the velocity pressure, Pvx, at a point sufficiently distant from the inlet as to be in still air. At this point, the static pressure is zero, and the velocity pressure in still air is zero. Ps1 + Pv1 = Psx + Pvx =0 This consideration, which is the same as that used in the methods for testing fans for performance rating purposes, charges to the fan losses incurred in accelerating the air into the fan inlet and eliminates the inaccuracies which arise in any attempt to measure the velocity pressure and static pressure at the fan inlet. To calculate the Fan Static Pressure: Ps = Ps2 - Ps1 - Pv1 + SEF 1 = Ps2 - (Ps1 + Pv1) + SEF 1 Since Ps1 + Pv1 = 0: Ps = Ps2 + SEF 1 8. In order to compare the test results to the quoted fan curve drawn for operation at 1900 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS
Ps2 = 6.4 in. wg N = 1890 rpm, RH fan speed A2 = 3.2 ft2 Blast Area = 2.25 ft2 MEASURED MOTOR DATA LH Fan Volts = = Amps = = NLA =
575, 572, 578 575 av 16, 17, 17 16.7 av 7.0
RH Fan Volts = = Amps = = NLA =
575, 574, 573 574 av 15, 16, 16 15.7 av 7.0
MOTOR NAMEPLATE DATA LH Fan 25 hp, 3 phase, 60 hertz 575 volts, 1780 rpm, 23 FLA RH Fan 25 hp, 3 phase, 60 hertz 575 volts, 1780 rpm, 23 FLA GENERAL Outlet dampers in full open positions. Fans connected to motors through belt drives.
SITE MEASUREMENTS pb = td3 = Ps3 = Pv3 = A3 =
29.05 in. Hg 78°F 5.6 in. wg 0.47 in. wg 7.4 ft2
LH Fan td1 = 75°F tw1 = 57°F td2 = 79°F Ps2 = 6.4 in. wg N = 1910 rpm, LH fan speed A2 = 3.2 ft2 Blast Area = 2.25 ft2 RH Fan td1 = 75°F tw1 = 57°F td2 = 79°F 32
CALCULATIONS DENSITIES For inlet conditions for both fans of: td1 = 75°F tw1 = 57°F p1 = pb = 29.05 in. Hg Use Figure N.1 in Annex N to obtain ρ1 = 0.0718 lbm/ft3 The density at Plane 2:
AMCA 203-90 (R2007) ⎛ P + 13.6 pb ⎞ ⎛ t d1 + 460 ⎞ ρ2 = ρ1 ⎜ s2 ⎟ ⎟⎜ ⎝ 13.6 p1 ⎠ ⎝ t d2 + 460 ⎠ ⎛ 6.4 + 13.6 × 29.05 ⎞ ⎛ 535 ⎞ = 0.0718 ⎜ ⎟⎜ ⎟ ⎝ 13.6 × 29.05 ⎠ ⎝ 539 ⎠ = 0.0724 lbm/ft 3
Annex K indicates that the average of the results of Equation A and Equation B will provide a reasonably accurate estimate of motor power output for a 25 hp motor operating at approximately 70% FLA. LH Fan Eqn A = 25 (16.7/23) (575/575) = 18.15 hp
The density at Plane 3: ⎛ P + 13.6 pb ⎞ ⎛ t d1 + 460 ⎞ ρ3 = ρ1 ⎜ s3 ⎟ ⎟⎜ ⎝ 13.6 p1 ⎠ ⎝ t d3 + 460 ⎠ ⎛ 5.6 + 13.6 × 29.05 ⎞ ⎛ 535 ⎞ = 0.0718 ⎜ ⎟⎜ ⎟ ⎝ 13.6 × 29.05 ⎠ ⎝ 538 ⎠ = 0.0724 lbm/ft 3
Eqn B = 25 [(16.7 - 7)/(23 - 7)] (575/575) = 15.16 hp Hmo
= (18.15 + 15.16)/2 = 16.66 hp
RH Fan Eqn A = 25 (15.7/23) (574/575) = 17.04 hp
FLOW RATES V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.47/0.0724)0.5 = 2792 fpm
Eqn B = 25 [(15.7 - 7)/(23 - 7)] (574/575) = 13.57 hp Hmo
= (17.04 + 13.57)/2 = 15.31 hp
Q3 = V3A3 = 2792 × 7.4 = 20661 cfm
Figure L.1 in Annex L indicates estimated belt drive loss of 5% for each fan.
Q = Q1 = Q3 (ρ3/ρ1) = 20661 (0.0724/0.0718) = 20834 cfm Assume that the air flow rate for each fan is proportional to its speed.
LH Motor HL = 0.05 Hmo = 0.05 × 16.66 = 0.83 hp H = Hmo - HL = 16.66 - 0.83 = 15.83 hp
LH Fan Q = Q1 = 20834 [1910/(1910 + 1890)] = 10472 cfm RH Fan Q = Q1 = 20834 [1890/(1910 + 1890)] = 10362 cfm FAN POWER INPUT LH Fan Measured amps/FLA = (16.7/23) = 0.73 = 73% RH Fan Measured amps/FLA = (15.7/23) = 0.68 = 68%
RH Motor HL = 0.05 Hmo = 0.05 × 15.31 = 0.77 hp H = Hmo - HL = 15.31 - 0.77 = 14.54 hp SYSTEM EFFECT FACTOR AMCA Publication 201-90, Figures 7.1 and 8.3 indicate the following calculations: LH Fan Q2 = Q1 (ρ1/ρ2) = 10472 (0.0718/0.0724) = 10385 cfm V2 = (Q2/A2) = (10385/3.2) = 3245 fpm 33
AMCA 203-90 (R2007) Blast area ratio = Blast Area/A2 = 2.25/3.2 = 0.70 RH Fan Q2 = Q1 (ρ1/ρ2) = 10362 (0.0718/0.0724) = 10276 cfm V2 = (Q2/A2) = (10276/3.2) = 3211 fpm Blast area ratio = Blast Area/A2 = 2.25/3.2 = 0.70 For a blast area ratio of 0.7 and no duct, Figure 8.3 shows System Effect Curve S applies. For each fan with velocities of 3245 fpm and 3211 fpm and curve S, Figure 7.1 shows SEF 1 = 0.5 in. wg at 0.075 lbm/ft3. At 0.0724 lbm/ft3: SEF 1 = 0.5 (0.0724/0.075) = 0.48 in. wg FAN STATIC PRESSURE Ps = Ps2 + SEF 1 LH Fan Ps = 6.4 + 0.48 = 6.88 in. wg RH Fan Ps = 6.4 + 0.48 = 6.88 in. wg
34
CONVERSION TO SPECIFIED CONDITIONS LH Fan Qc = 10472 (1900/1910) = 10417 cfm Psc = 6.88 (1900/1910)2 (0.075/0.0718) = 7.11 in. wg Hc = 15.83 (1900/1910)3 (0.075/0.0718) = 16.28 hp RH Fan Qc = 10362 (1900/1890) = 10417 cfm Psc = 6.88 (1900/1890)2 (0.075/0.0718) = 7.26 in. wg Hc = 14.54 (1900/1890)3 (0.075/0.0718) = 15.43 hp
AMCA 203-90 (R2007)
EXAMPLE 2A: UTILITY FAN IN A VENTILATION SYSTEM
3 STATIC PRESSURE TAPS 1
2
PLAN VIEW
L
3-PIECE ELBOW R/D = 1 SEF 1
SEF 2
SIDE VIEW
OUTLET SIDE VIEW
COMMENTS 1. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located near the end of a straight run of duct, such as shown in the diagram. Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. Ps3 is used in determining the density at the traverse plane. Measure the area of the traverse plane, A3, which is located at the tip of the Pitot-static tube. 2. Determine Ps1 by averaging the pressure measurements at each of four static pressure taps in the collar connection at the fan inlet. Determine Ps2 by averaging the pressure measurements at each of four static pressure taps located near the fan outlet. 3. Measure td3 and tw3 in the traverse plane. Assume td1 is equal to td3. Determine pb for the general vicinity of the fan. Measure td2 in Plane 2. All of these measurements are used in determining densities at the various planes of interest. 4. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV) and full load amps (FLA). If the motor power output is to be estimated by using the phase current method
described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K. 5. SEF 1 is due to the effect of the elbow located at the fan inlet. SEF 2 is due to the effect of insufficient length of duct between the fan outlet and the elbow downstream of the fan. In order to calculate the values of the SEFs, it is necessary to measure the inlet area and the outlet area of the fan, A1 and A2; the length of the outlet duct, L; and the blast area of the fan. 6. To calculate the Fan Static Pressure: Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2 Where: Pv1 = Pv3 (A3/A1)2 (ρ3/ρ1) 7. In order to compare the test results to the quoted fan curve drawn for operation at 1880 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14.
35
AMCA 203-90 (R2007) OBSERVATIONS SITE MEASUREMENTS pb = 29.20 in. Hg td2 = 72°F td3 = 72°F tw3 = 66°F Ps1 = -2.18 in. wg Ps2 = 0.35 in. wg Ps3 = -1.95 in. wg Pv3 = 0.45 in. wg N = 1730 rpm A1 = 1.07 ft2 A2 = 1.17 ft2 A3 = 1.07 ft2 Blast Area = 0.7 ft2 L = 0.83 ft MEASURED MOTOR DATA Volts = = Amps = = NLA =
227, 229, 228 228 av 10.2, 10.3, 10.4 10.3 av 7.1
MOTOR NAMEPLATE DATA 5 hp, 3 phase, 60 hertz 230 volts, 1750 rpm, 14 FLA
The density at Plane 1: ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ1 = ρ3 ⎜ s1 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d1 + 460 ⎠ ⎛ −2.18 + 13.6 × 29.20 ⎞ ⎛ 532 ⎞ = 0.0719 ⎜ ⎟ ⎜ 532 ⎟ 13.6 × 29.06 ⎝ ⎠⎝ ⎠ = 0.0718 lbm/ft 3 The density at Plane 2: ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ2 = ρ3 ⎜ s2 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d2 + 460 ⎠ ⎛ 0.35 + 13.6 × 29.20 ⎞ ⎛ 532 ⎞ = 0.0719 ⎜ ⎟ ⎜ 532 ⎟ 13.6 × 29.06 ⎝ ⎠⎝ ⎠ 3 = 0.0723 lbm/ft FLOW RATES V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.45/0.0719)0.5 = 2742 fpm Q3 = V3A3 = 2742 × 1.07 = 2934 cfm
GENERAL
Q = = = =
Fan connected to motor through belt drive.
FAN POWER INPUT
CALCULATIONS DENSITIES For Plane 3 conditions of: td3 = 72°F tw3 = 66°F p3 = pb + (Ps3/13.6) = 29.20 + (-1.95/13.6) = 29.06 in. Hg Use Figure N.1 in Annex N to obtain ρ3 = 0.0719 lbm/ft3
Measured amps/FLA = 10.3/14 = 0.74 = 74% Annex K indicates that the average of the results of Equation A and Equation B will provide a reasonably accurate estimate of motor power output for a 5 hp motor operating at 74% FLA. Eqn A = 5 (10.3/14) (228/230) = 3.65 hp Eqn B = 5 [(10.3 - 7.1)/(14 - 7.1)] (228/230) = 2.30 hp Hmo
It is assumed that td1 = td3
36
Q1 Q3 (ρ3/ρ1) 2934 (0.0719/0.0718) 2938 cfm
= (3.65 + 2.30)/2 = 2.98 hp
AMCA 203-90 (R2007) Figure L.1 in Annex L indicates estimated belt drive loss of 6.5%. HL = 0.065 Hmo = 0.065 × 2.98 = 0.19 hp H = Hmo - HL = 2.98 - 0.19 = 2.79 hp SYSTEM EFFECT FACTORS To determine the value of SEF 1, calculate the velocity at the fan inlet: V1 = Q1/A1 = 2938/1.07 = 2746 fpm
L in % effective duct length = (L/3.05) 100 = (0.83/3.05) 100 = 27% Blast area ratio = Blast Area/A2 = 0.7/1.17 = 0.6 For blast area ratio of 0.6, 27% effective duct length and elbow position C, Figure 8.5 shows System Effect Curve P - Q applies. For 2494 fpm velocity and curve P - Q, Figure 7.1 shows SEF 2 = 0.7 in. wg at 0.075 lbm/ft3. At 0.0723 lbm/ft3: SEF 2 = 0.7 (0.0723/0.075) = 0.67 in. wg FAN STATIC PRESSURE
AMCA Publication 201-90, Figure 9.5 indicates that for a three piece elbow with radius to diameter ratio of 1, and with no duct between the elbow and the fan inlet, System Effect Curve R applies. For 2746 fpm velocity and curve R, Figure 7.1 shows SEF 1 = 0.55 in. wg at 0.075 lbm/ft3. At 0.0718 lbm/ft3: SEF 1 = 0.55 (0.0718/0.075) = 0.53 in. wg
Pv1 = Pv3 (A3/A1)2 (ρ3/ρ1) = 0.45 (1.07/1.07)2 (0.0719/0.0718) = 0.45 in. wg Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2 = 0.35 - (-2.18) - 0.45 + 0.53 + 0.67 = 3.28 in. wg CONVERSION TO SPECIFIED CONDITIONS
For SEF 2, AMCA Publication 201-90, Figures 7.1, 8.1, and 8.5 indicate the following calculations: Q2 = Q3 (ρ3/ρ2) = 2934 (0.0719/0.0723) = 2918 cfm V2 = (Q2/A2) = 2918/1.17 = 2494 fpm
Qc = 2938 (1880/1730) = 3193 cfm Psc = 3.28 (1880/1730)2 (0.075/0.0718) = 4.05 in. wg Hc = 2.79 (1880/1730)3 (0.075/0.0718) = 3.74 hp
Duct diameter equivalent to the fan outlet area: De2 = (4A2/π)0.5 = (4 × 1.17/π)0.5 = 1.22 ft Figure 8.1 shows that for velocities of 2500 fpm or less, the 100% effective outlet duct length is 2.5 duct diameters, = 2.5 × 1.22 = 3.05 ft
37
AMCA 203-90 (R2007)
EXAMPLE 2B: CENTRIFUGAL FAN IN A SAWDUST CONVEYING SYSTEM
SEF 2 2
1 SEF 1 4-PIECE ELBOW R/D = 1
L2 L1
3 SIDE VIEW
OUTLET SIDE VIEW
COMMENTS 1. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located near the end of a straight run of duct, such as shown in the diagram. Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. Ps3 is used in determining the density at the traverse plane. Measure the area of the traverse plane, A3, which is located at the tip of the Pitot-static tube. 2. Determine Ps1 by using a Pitot-static tube or static pressure taps in the duct connection at the fan inlet. If a Pitot-static tube is used, it should not project into the upstream elbow but be located well within the length of the duct connection as shown in the diagram. The friction loss in the short length of outlet duct is assumed to be negligible, and Ps2 is considered to be equal to the static pressure at the duct outlet. The static pressure at the outlet of the duct is zero gauge pressure, referred to the atmospheric pressure in the region of the duct outlet. In situations such as this example, the air may be discharging from the duct into a region in which the atmospheric pressure is somewhat different from that to which all other pressure measurements are referred. When this possibility exists, it is essential that the static pressure in the region of the discharging air be measured, referred to the same 38
atmospheric pressure as used in all other pressure measurements. In this case, the pressure was measured as 0.1 in. wg. 3. Measure td3 and tw3 in the traverse plane. Determine pb for the general vicinity of the fan. Measure td1 and td2. All of these measurements are used in determining densities at the various planes of interest. 4. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV) and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K. 5. SEF 1 is due to the effect of insufficient length of duct between the fan inlet and the elbow upstream of the fan. SEF 2 is due to the effect of insufficient length of duct at the fan outlet. In order to calculate the values of the SEFs, it is necessary to measure the inlet area and the outlet area of the fan, A1 and A2; the lengths of the inlet connection and the outlet duct, L1 and L2; and the blast area of the fan.
AMCA 203-90 (R2007) CALCULATION
6. To calculate the Fan Static Pressure: Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2
DENSITIES
Where:
For Plane 2 conditions of:
Pv1 = Pv3 (A3/A1)2 (ρ3/ρ1)
td2 = 91.3°F tw2 = 70.4°F
7. In order to compare the test results to the quoted fan curve drawn for operation at 2075 rpm and 0.075 lbm/ft3, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS pb td1 td2 tw2 td3 Ps1 Ps2 Ps3 Pv3 N A1 A2 A3
= = = = = = = = = = = = =
29.82 in. Hg 86.6°F 91.3°F 70.4°F 86°F -11.4 in. wg 0.1 in. wg -8.9 in. wg 1.24 in. wg 2120 rpm, fan speed 1.40 ft2 1.40 ft2 1.57 ft2
Blast Area = 1.26 ft2 L1 = 1.33 ft L2 = 3.0 ft MEASURED MOTOR DATA Volts = = Amps = = NLA =
460, 460, 459 460 av 26.5, 25.5, 26 26 av 11.3
p2 = pb + (Ps2/13.6) = 29.82 + (0.1/13.6) = 29.83 in. Hg Use Figure N.1 in Annex N to obtain ρ2 = 0.0714 lbm/ft3 The density at Plane 1: ⎛ P + 13.6 pb ⎞ ⎛ t d2 + 460 ⎞ ρ1 = ρ2 ⎜ s1 ⎟ ⎟⎜ ⎝ 13.6 p2 ⎠ ⎝ t d1 + 460 ⎠ 4 + 13.6 × 29.82 ⎞ ⎛ 551.3 ⎞ ⎛ −11.4 = 0.0714 ⎜ ⎟ ⎜ 546.6 ⎟ 13.6 × 29.83 ⎝ ⎠⎝ ⎠ = 0.0700 lbm/ft 3 The density at Plane 3: ⎛ P + 13.6 pb ⎞ ⎛ t d2 + 460 ⎞ ρ3 = ρ2 ⎜ s3 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d3 + 460 ⎠ ⎛ −8.9 + 13.6 × 29.82 ⎞ ⎛ 551.3 ⎞ = 0.0714 ⎜ ⎟ ⎜ 546 ⎟ 13.6 × 29.83 ⎝ ⎠⎝ ⎠ = 0.0705 lbm/ft 3 FLOW RATES V3 = 1096 (Pv3/ρ3)0.5 = 1096 (1.24/0.0705)0.5 = 4596 fpm Q3 = V3A3 = 4596 × 1.57 = 7216 cfm
MOTOR NAMEPLATE DATA 30 hp, 3 phase, 60 hertz 460 volts, 1750 rpm, 36 FLA GENERAL Fan connected to motor through belt drive.
Q = = = =
Q1 Q3 (ρ3/ρ1) 7216 (0.0705/0.0700) 7268 cfm
FAN POWER INPUT Measured amps/FLA = (26/36) = 0.72 = 72%
39
AMCA 203-90 (R2007) Annex K indicates that the average of the results of Equation A and Equation B will provide a reasonably accurate estimate of motor power output for a 30 hp motor operating at 72% FLA. Eqn A = 30 (26/36) (460/460) = 21.67 hp Eqn B = 30 [(26 - 11.3)/(36 - 11.3)] (460/460) = 17.85 hp Hmo
= (21.67 + 17.85)/2 = 19.76 hp
Q2 = Q3 (ρ3/ρ2) = 7216 (0.0705/0.0714) = 7125 cfm V2 = (Q2/A2) = (7125/1.40) = 5089 fpm Duct diameter equivalent to the fan outlet area:
Figure L.1 in Annex L indicates estimated belt drive loss of 4.8%.
De2 = (4A2/π)0.5 = (4 × 1.40/π)0.5 = 1.34 ft
HL = 0.048 Hmo = 0.048 × 19.76 = 0.95 hp
Figure 8.3 shows that for velocities over 2500 fpm, 100% effective duct length is one duct diameter per 1000 fpm:
H = Hmo - HL = 19.76 - 0.95 = 18.81 hp SYSTEM EFFECT FACTORS To determine the value of SEF 1, calculate the velocity at the fan inlet: V1 = (Q1/A1) = (7268/1.40) = 5191 fpm
= D2 (V2/1000) = 1.34 (5089/1000) = 6.82 ft The length of the outlet duct in % effective duct length: = (L2/6.82) 100 = (3.0/6.82) 100 = 44%
The diameter of the fan inlet:
Blast ratio area = Blast Area/A2 = 1.26/1.40 = 0.9
D1 = (4A1/π)0.5 = (4 × 1.40/π)0.5 = 1.34 ft.
For blast area ratio of 0.9 and 44% effective duct length, Figure 8.3 shows no System Effect Curve applies and SEF 2 = 0.
The length of the duct between the elbow and the fan inlet in terms of D1:
FAN STATIC PRESSURE
= (L1/D1) = (1.33/1.34) = 1.0 AMCA Publication 201-90, Figure 9.5 indicates that for a four piece elbow with a radius to diameter ratio of 1, and with a length of duct between the elbow and the fan inlet equal to 1 equivalent diameter, System Effect Curve S applies. For 5191 fpm velocity and curve S, Figure 7.1 shows SEF 1 = 1.3 in. wg at 0.075 lbm/ft3. At 0.0700 lbm/ft3: SEF 1 = 1.3 (0.0700/0.075) = 1.2 in. wg 40
For SEF 2, AMCA Publication 201-90, Figure 8.3 indicates the following calculations:
Pv1 = Pv3 (A3/A1)2 (ρ3/ρ1) = 1.24 (1.57/1.40)2 (0.0705/0.0700) = 1.57 in. wg Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2 = 0.1 - (-11.4) - 1.57 + 1.2 + 0 = 11.13 in. wg CONVERSIONS TO SPECIFIED CONDITIONS Qc = 7268 (2075/2120) = 7114 cfm Psc = 11.13 (2075/2120)2 (0.075/0.0700) = 11.42 in. wg
AMCA 203-90 (R2007) Hc = 18.81 (2075/2120)2 (0.075/0.0700) = 18.90 hp
41
AMCA 203-90 (R2007)
EXAMPLE 2C: AXIAL FAN IN A DRYER SYSTEM
5
4 1
2
STRAIGHTENING VANES 3
SEF 2
STATIC PRESSURE TAPS
A3
SEF 1
PLAN VIEW
INNER CYLINDER LOCATION OF PLANE 3
SIDE VIEW
COMMENTS 1. This type of installation is normally classified as one in which a satisfactory test cannot be conducted. Due to the configurations of the airways, there are no locations at which reasonably accurate pressure measurements can be made. In addition, the judgments required in determining the values of the SEFs are susceptible to error. The purpose of presenting this example is to illustrate the not uncommon instance in which a test must be conducted in order to provide performance information, even though the results will be innaccurate to a degree which is not normally acceptable. 2. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located as shown in the diagram. Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. These velocity pressure and static pressure measurements are susceptible to error due to the turbulence existing in the region of the fan outlet. In addition, it is undesirable to have Plane 3 located in a diverging airway. However, no other more suitable location for Plane 3 exists in this example. It is recommended that the Pitot-static tube be oriented so that its nose is aligned with the anticipated flow streams, particularly near the walls of the diffuser. Determine the area of the traverse plane, A3, which is 42
located at the tip of the Pitot-static tube, as shown in the diagram, not at the location of the Pitot-static tube access holes. 3. Determine Ps4 by averaging the pressure measurements at each of four static pressure taps located near the fan inlet. In the same manner, determine Ps5 at a location near the fan outlet. It is undesirable to have pressure measurement planes located in converging and diverging airways, but there are no other more suitable locations for these planes in this installation. Measure A4 and A5, the cross-sectional areas of the airways at Planes 4 and 5. 4. Measure td3, tw3, and td4. Determine pb for the general vicinity of the fan. These measurements are used in the determination of densities at the various planes of interest. 5. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV) and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K.
AMCA 203-90 (R2007) 6. Although an elbow is located a short distance upstream of the fan, it is considered to produce no system effect since it is equipped with turning vanes and the average velocity through the elbow will be relatively low due to its large cross-sectional area. Therefore, SEF 1 = 0. In judging SEF 2, the rapidly diverging transition fitting downstream of the fan is considered equivalent to no duct at the fan outlet. In order to calculate the value of SEF2, it is necessary to measure the outlet area of the fan, A2.
MOTOR NAMEPLATE DATA
7. To calculate the Fan Static Pressure,
DENSITIES
Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2
For Plane 3 conditions of:
Where:
td3 = 86.5°F tw3 = 75.5°F
25 hp, 3 phase, 60 hertz 460 volts, 1750 rpm, 31 FLA GENERAL Fan connected to motor through belt drive CALCULATIONS
Pv1 = Pv3 (A3/A1)2 (ρ3/ρ1) Ps1 and Ps2 are calculated on the basis of total pressure considerations, using Ps4, Ps5, and the calculated velocity pressures at Planes 1, 2, 4, and 5. 8. In order to compare the test results to the quoted fan curve drawn for operation at 1580 rpm and 0.0690 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS pb td3 tw3 td4 Ps3 Pv3 Ps4 Ps5 N A1 A3 A4 A5
= 28.90 in. Hg = 86.5°F = 75.5°F = 85°F = 1.5 in. wg = 0.044 in. wg = -1.57 in. wg = 1.22 in. wg = 1590 rpm = A2 = 8.0 ft2 = 29.8 ft2 = 12.4 ft2 = 9.6 ft2
MEASURED MOTOR DATA Volts = = Amps = = NLA =
450, 449, 448 449 av 25.0, 24.5, 25.0 24.8 av 9.4
p3 = pb + (Ps3/13.6) = 28.90 + (1.5/13.6) = 29.01 in. Hg Use Figure N.1 from Annex N to obtain ρ3 = 0.0694 lbm/ft3 The density at Plane 4: ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ 4 = ρ3 ⎜ s4 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d4 + 460 ⎠ ⎛ −1.57 + 13.6 × 28.90 ⎞ ⎛ 546.5 ⎞ = 0.0694 ⎜ ⎟ ⎜ 545 ⎟ 13.6 × 29.01 ⎝ ⎠⎝ ⎠ = 0.0691 lbm/ft 3 It is assumed that td1 = td4 and at the low pressure levels which exist at Planes 1 and 4, the difference between these pressures will be small, and assuming ρ1 = ρ4, will result in an error which is considered negligible. By similar reasoning, it is assumed that ρ5 = ρ2 = ρ3. FLOW RATES V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.044/0.0694)0.5 = 873 fpm Q3 = V3A3 = 873 × 29.8 = 26015 cfm Q = = = =
Q1 Q3 (ρ3/ρ1) 26015 (0.0694/0.0691) 26128 cfm 43
AMCA 203-90 (R2007) FAN POWER INPUT
FAN STATIC PRESSURE
Measured amps/FLA = (24.8/31) = 0.80 = 80%
Pv4 = Pv3 (A3/A4)2 (ρ3/ρ4) = 0.044 (29.8/12.4)2 (0.0694/0.0691) = 0.26 in. wg
Annex K indicates that the average of the results of Equation A and Equation B will provide a reasonably accurate estimate of motor power output for a 25 hp motor operating at 80% FLA.
Pv1 = Pv3 (A3/A1)2 (ρ3/ρ1) = 0.044 (29.8/8.0)2 (0.0694/0.0691) = 0.61 in. wg
Eqn A = 25 (24.8/31) (449/460) = 19.52 hp
Ps1 + Pv1 = Ps4 + Pv4 Ps1 = Ps4 + Pv4 - Pv1 = -1.57 + 0.26 - 0.61 = -1.92 in. wg
Eqn B = 25 [(24.8 - 9.4)/(31 - 9.4)] (449/460) = 17.40 hp Hmo
= (19.52 + 17.40)/2 = 18.46 hp
Figure L.1 in Annex L indicates estimated belt drive loss of 4.9%. HL = 0.049 Hmo = 0.049 × 18.46 = 0.90 hp H = Hmo - HL = 18.46 - 0.90 = 17.56 hp SYSTEM EFFECT FACTORS SEF 1 = 0 See item 6 under COMMENTS. To determine the value of SEF 2, AMCA Publication 201-90, Figure 8.2 indicates that a vaneaxial fan with no outlet duct will use System Effect Curve U. Q2 = Q3 (ρ3/ρ2) = 26015 (0.0694/0.0694) = 26015 cfm V2 = (Q2/A2) = (26015/8.0) = 3252 fpm From Figure 7.1, using 3252 fpm and curve U, SEF 2 = 0.26 in. wg at 0.075 lbm/ft3. At 0.0694 lbm/ft3: SEF 2 = 0.26 (0.0694/0.075) = 0.24 in. wg
44
Pv5 = Pv3 (A3/A5)2 (ρ3/ρ5) = 0.044 (29.8/9.6)2 (0.0694/0.0694) = 0.42 in. wg Pv2 = Pv3 (A3/A2)2 (ρ3/ρ2) = 0.044 (29.8/8.0)2 (0.0694/0.0694) = 0.61 in. wg Ps2 + Pv2 = Ps5 + Pv5 Ps2 = Ps5 + Pv5 - Pv2 = 1.22 + 0.42 - 0.61 = 1.03 in. wg Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2 = 1.03 - (-1.92) - 0.61 + 0 + 0.24 = 2.58 in. wg Losses between Planes 1 and 4 and between Planes 2 and 5 have been ignored. CONVERSION TO SPECIFIED CONDITIONS Qc = 26128 (1580/1590) = 25964 cfm Psc = 2.58 (1580/1590)2 (0.0690/0.0691) = 2.54 in. wg Hc = 17.56 (1580/1590)3 (0.0690/0.0691) = 17.21 hp
AMCA 203-90 (R2007)
EXAMPLE 2D: CENTRIFUGAL FAN IN A SCRUBBER SYSTEM
INLET BOX DAMPER STATIC PRESSURE TAPS SEF 1
3 1 L INLET BOX 2 DIFFUSER SIDE VIEW
OUTLET SIDE VIEW
COMMENTS 1. This fan, as supplied and rated by the manufacturer, includes the inlet box damper and the inlet box. Performance ratings for fans with inlet box dampers cover operation with the dampers in the full open positions. In order to be able to compare the test results to the fan performance ratings, it is essential that the damper be fixed in the full open position for the duration of the test. 2. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located shortly upstream of the inlet damper. Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. Measure A3, the area of the traverse plane, located at the tip of the Pitot-static tube and A1, the area of the inlet to the damper. 3. Determine Ps2 by averaging the pressure measurements at each of four static pressure taps located near the end of the fan outlet. See Annex E for details of static pressure taps. 4. Measure td3 and tw3 in the traverse plane. Determine pb for the general vicinity of the fan. Measure td2 in Plane 2. These measurements are used in the determination of densities at the various planes of interest.
5. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV), and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K. 6. SEF 1 is due to the effect of insufficient length of duct at the fan outlet. In order to calculate the value of SEF 1, it is necessary to measure the length of the outlet duct, L; the fan outlet area, A2; and the blast area of the fan. 7. To calculate the Fan Static Pressure: Ps = Ps2 - Ps1 - Pv1 + SEF 1 Since Plane 1 is located shortly downstream of Plane 3 in an airway of uniform cross-section (A1 = A3), the conditions which exist at Plane 3 are assumed to exist at Plane 1. Therefore, Ps1 = Ps3 and Pv1 = Pv3. 8. In order to compare the test results to the quoted fan curve drawn for operation at 1780 rpm and 0.059 lbm/ft3 density, it is necessary to convert the results 45
AMCA 203-90 (R2007) to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS
pe = 0.5603 in. Hg pp = pe - [p3 (td3 - tw3)/2700] = 0.5603 - [24.28 (63 - 62)/2700] = 0.5513 in. Hg
SITE MEASUREMENTS pb = 29.44 in. Hg td2 = 97°F td3 = 63°F tw3 = 62°F Ps2 = 1.1 in. wg Ps3 = -70.2 in. wg Pv3 = 0.64 in. wg N = 1790 rpm A1 = 6.5 ft2 A2 = 5.32 ft2 A3 = 6.5 ft2 Blast Area = 1.89 ft2 L = 2.50 ft MEASURED MOTOR DATA Volts = 4160, 4150, 4150 = 4153 av Amps = 50, 51, 52 = 51 av NLA = 14 MOTOR NAMEPLATE DATA 500 hp, 3 phase, 60 hertz 4160 volts, 1785 rpm, 61 FLA GENERAL Inlet box damper in full open position. Fan direct connected to motor. CALCULATIONS DENSITIES For Plane 3 conditions of: td3 = 63°F tw3 = 62°F p3 = pb + (Ps3/13.6) = 29.44 + (-70.2/13.6) = 24.28 in. Hg Use the modified Apjohn equation, described in Section M.2.3 in Annex M, and the table in Figure N.2 in Annex N to calculate the density at Plane 3.
46
ρ3 = =
1.3257( p3 − 0.378 pp ) t d3 + 460 1.3257 ( 24.28 − 0.378 × 0.5513 )
63 + 460 = 0.0610 lbm/ft 3
Consider ρ1 to be equal to ρ3. The density at Plane 2: ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ2 = ρ3 ⎜ s2 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d2 + 460 ⎠ ⎛ 1.1 + 13.6 × 29.44 ⎞ ⎛ 523 ⎞ = 0.0610 ⎜ ⎟⎜ ⎟ ⎝ 13.6 × 24.28 ⎠ ⎝ 557 ⎠ = 0.0696 lbm/ft 3 FLOW RATES V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.64/0.0610)0.5 = 3550 fpm Q3 = = = Q = =
V3A3 3550 × 6.5 23075 cfm Q1 = Q3 23075 cfm
FAN POWER INPUT Measured amps/FLA = 51/61 = 0.84 = 84% Annex K indicates that the average of the results of Equation A and Equation B will provide a reasonably accurate estimate of motor power output for a 500 hp motor operating at 84% FLA. Eqn A = 500 (51/61) (4153/4160) = 417 hp Eqn B = 500 [(51 - 14)/(61 - 14)] (4153/4160) = 393 hp Hmo
= (417 + 393)/2 = 405 hp
AMCA 203-90 (R2007) Since the fan is direct-connected to the motor, there is no drive loss, and: H = Hmo = 405 hp SYSTEM EFFECT FACTOR AMCA Publication 201-90, Figures 7.1 and 8.3 indicate the following calculations. Q2 = Q3 (ρ3/ρ2) = 23075 (0.0610/0.0696) = 20224 cfm V2 = Q2/A2 = 20224/5.32 = 3802 fpm Duct diameter equivalent to the diffuser outlet area: De2 = (4A2/π)0.5 = (4 × 5.32/π)0.5 = 2.60 ft Figure 8.3 shows that for velocities over 2500 fpm 100% effective duct length is one duct diameter for every 1000 fpm: = De2 (V2/1000) = 2.60 (3802/1000) = 9.89 ft. L in % effective duct length:
Blast area ratio = Blast Area/A2 = 1.89/5.32 = 0.36 For a blast area ratio of 0.36, and 25% effective duct length, Figure 8.3 shows System Effect Curve U applies. For 3802 fpm velocity and curve U, Figure 7.1 shows SEF 1 = 0.36 in. wg at 0.075 lbm/ft3. At 0.0696 lbm/ft3: SEF 1 = 0.36 (0.0696/0.075) = 0.33 in. wg FAN STATIC PRESSURE Ps1 = = Pv1 = = Ps = = =
Ps3 - 70.2 in. wg Pv3 0.64 in. wg Ps2 - Ps1 - Pv1 + SEF 1 1.1 - (-70.2) - 0.64 + 0.33 71.0 in. wg
CONVERSION TO SPECIFIED CONDITIONS Qc = 23075 (1780/1790) = 22946 cfm Psc = 71.0 (1780/1790)2 (0.059/0.0610) = 67.9 in. wg Hc = 405 (1780/1790)3 (0.059/0.0610) = 385 hp
= (L/9.89) 100 = (2.50/9.89) 100 = 25%
47
AMCA 203-90 (R2007)
EXAMPLE 2E: CENTRIFUGAL FAN IN A PROCESS SYSTEM
STATIC OUTLET DAMPER PRESSURE TAPS 5 2
INLET BOXES
INLET BOX DAMPERS
1a
1b
3a
3b
OPPOSITE OUTLET SIDE VIEW
SIDE VIEW
COMMENTS 1. This fan, as supplied and rated by the manufacturer, includes the inlet box dampers and the inlet boxes, but does not include the outlet damper. Performance ratings for fans with inlet box dampers cover operation with the dampers in the full open positions. Also, performance ratings for items such as the outlet damper are for operation in the full open position. In order to be able to compare the test results to the fan performance ratings, it is essential that the outlet damper and the inlet dampers be fixed in their full open positions. 2. Determine Pv3a and Pv3b by using the root mean square of the velocity pressure measurements made in Planes 3a and 3b. Determine Ps3a and Ps3b by averaging each of the two sets of static pressure measurements made in the same traverses. Procedures for traverses are described in Section 9.4. Measure A3a and A3b, the areas of the traverse planes and A1a and A1b, the areas of the inlets to the inlet dampers. 3. Determine Ps5 by averaging the pressure measurements of each of four static pressure taps located downstream of the outlet damper.
48
4. Measure td3a, td3b, and td5. Since flue gas is being handled by the fan, the Orsat apparatus is used by process personnel to determine the density of the gas. Determine pb for the general vicinity of the fan. These data are used in the determination of densities at the various planes of interest. 5. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV) and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Motor performance data, supplied by the motor manufacturer, are used in the determination of motor power output for this example. 6. In this example, the duct downstream of the outlet damper is of sufficient length, and no SEF applies. 7. To calculate the Fan Static Pressure: Ps = Ps2 - Ps1 - Pv1
AMCA 203-90 (R2007) Ps2 is calculated on the basis of total pressure considerations using Ps5, the outlet damper pressure loss, and the calculated velocity pressures at Planes 2 and 5. Since the inlets to the inlet dampers (Planes 1a and 1b) are located shortly downstream of the traverse planes (Planes 3a and 3b) in an airway of uniform cross-section, the conditions which exist at the traverse planes are assumed to exist at the inlets to the inlet dampers. Ps1 = Ps3 = (Ps3a + Ps3b)/2 Pv1 is calculated using the total flow rate and the total area at the inlets to the inlet dampers. Pv1 = (Q1/1096A1)2 ρ1 8. In order to compare the test results to the quoted fan curve drawn for operation at 880 rpm and 0.049 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS pb td3a td3b td5 Ps3a Ps3b Pv3a Pv3b Ps5 N A1a
= 30.12 in. Hg = 345°F = 359°F = 363°F = -18.8 in. wg = -18.3 in. wg = 2.053 in. wg = 2.028 in. wg = -1.6 in. wg = 892 rpm = A1b = 60.7 ft2 A2 = 115 ft2 A3a = A3b = 60.7 ft2 A5 = 140 ft2 Blast Area = 80 ft2 The density of the gas, as determined by Orsat analysis, is 0.0725 lbm/ft3 at 29.92 in. Hg and 70°F. MEASURED MOTOR DATA Volts = = Amps = = kW =
4300, 4250, 4200 4250 av 378, 376, 380 378 av 2519
MOTOR NAMEPLATE DATA 3000 hp, 3 phase, 60 hertz 4000 volts, 880 rpm, 385 FLA GENERAL Inlet box dampers and outlet damper in full open positions. Fan direct connected to motor. Motor efficiency data supplied by motor manufacturer. Pressure loss data supplied by manufacturer of outlet damper. CALCULATIONS DENSITIES The densities at Planes 3a and 3b are: ⎛ P + 13.6 pb ⎞ ⎛ 70 + 460 ⎞ ρ3a = 0.0725 ⎜ s3a ⎟ ⎟⎜ ⎝ 13.6 × 29.92 ⎠ ⎝ t d3a + 460 ⎠ ⎛ −18.8 + 13.6 × 30.12 ⎞ ⎛ 530 ⎞ = 0.0725 ⎜ ⎟ ⎜ 805 ⎟ 13.6 × 29.92 ⎝ ⎠⎝ ⎠ = 0.0458 lbm/ft 3 ⎛ P + 13.6 pb ⎞ ⎛ 70 + 460 ⎞ ρ3b = 0.0725 ⎜ s3b ⎟ ⎟⎜ ⎝ 13.6 × 29.92 ⎠ ⎝ t d3b + 460 ⎠ ⎛ −18.3 + 13.6 × 30.12 ⎞ ⎛ 530 ⎞ = 0.0725 ⎜ ⎟ ⎜ 819 ⎟ 13.6 × 29.92 ⎝ ⎠⎝ ⎠ = 0.0451 lbm/ft 3 It is assumed that ρ1a = ρ3a and ρ1b = ρ3b. The density at Plane 5: ⎛ P + 13.6 pb ⎞ ⎛ 70 + 460 ⎞ ρ5 = 0.0725 ⎜ s5 ⎟ ⎟⎜ ⎝ 13.6 × 29.92 ⎠ ⎝ t d5 + 460 ⎠ ⎛ −1.6 + 13.6 × 30.12 ⎞ ⎛ 530 ⎞ = 0.0725 ⎜ ⎟ ⎜ 823 ⎟ 13.6 × 29.92 ⎝ ⎠⎝ ⎠ 3 = 0.0468 lbm/fft It is assumed that ρ2 = ρ5. FLOW RATES V3a = 1096 (Pv3a/ρ3a)0.5 = 1096 (2.053/0.0458)0.5 = 7338 fpm Q3a = V3aA3a = 7338 × 60.7 = 445417 cfm 49
AMCA 203-90 (R2007) V3b = 1096 (Pv3b/ρ3b)0.5 = 1096 (2.028/0.0451)0.5 = 7349 fpm Q3b = V3bA3b = 7349 × 60.7 = 446084 cfm Q3 = Q3a + Q3b = 445417 + 446084 = 891501 cfm Since the air is divided evenly between the two inlet boxes:
Pv1 = (Q1/1096A1)2 ρ1 = (891501/1096 × 121.4)2 0.0455 = 2.04 in. wg Pv2 = Pv1 (A1/A2)2 (ρ1/ρ2) = 2.04 (121.4/115)2 (0.0455/0.0468) = 2.21 in. wg Pv5 = Pv1 (A1/A5)2 (ρ1/ρ5) = 2.04 (121.4/140)2 (0.0455/0.0468) = 1.49 in. wg Ps2 + Pv2 = Ps5 + Pv5 + Damper Loss
ρ1 = ρ3 = (ρ3a + ρ3b)/2 = (0.0458 + 0.0451)/2 = 0.0455 lbm/ft3
Ps2 = Ps5 + Pv5 + Damper Loss - Pv2 = -1.6 + 1.49 + 0.75 - 2.21 = -1.57 in. wg
Q = = = =
Ps1 = = = =
Q1 Q3 (ρ3/ρ1) 891501 (0.0455/0.0455) 891501 cfm
FAN POWER INPUT Measured amps/FLA = (378/385) = 0.98 = 98% Annex K indicates that Equation A will provide a reasonably accurate estimate of motor power output for a 3000 hp motor operating at 98% FLA. Hmo = 3000 (378/385) (4250/4000) = 3130 hp The data supplied by the motor manufacturer indicate motor efficiency of 94% at the measured power input of 2519 kW. Using this information: Hmo = (2519 × 0.94)/0.746 = 3174 hp The more accurate method of estimating the motor power output is assumed to be the latter. Since the fan is direct connected to the motor, there is no drive loss, and: H = Hmo = 3174 hp
50
FAN STATIC PRESSURE
Ps3 (Ps3a + Ps3b)/2 (-18.8 - 18.3)/2 -18.55 in. wg
Ps = Ps2 - Ps1 - Pv1 = -1.57 - (-18.55) - 2.04 = 14.94 in. wg CONVERSION TO SPECIFIED CONDITIONS Qc = 891501 (880/892) = 879508 cfm Psc = 14.94 (880/892)2 (0.049/0.0455) = 15.66 in. wg Hc = 3174 (880/892)3 (0.049/0.0455) = 3282 hp
AMCA 203-90 (R2007)
EXAMPLE 2F: AXIAL FAN IN A VENTILATION SYSTEM
3 GUIDE VANES
STATIC PRESSURE TAPS 5
4 SEF 1
2-PIECE ELBOW (TYPICAL) INNER CYLINDER
L1 1
SEF 2
L2 2
COMMENTS 1. The unusual duct arrangement in this example makes it very difficult to obtain accurate pressure measurements, and this fact should be understood before testing begins. Also, the use of a diverging inlet fitting and a converging outlet fitting with this fan can pose additional problems. Unless the degrees of divergence and convergence are moderate, as they are in this example, the fan performance will be adversely affected. 2. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located well downstream in a straight run of duct, such as shown in the diagram. Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. Ps3 is used in determining the density at the traverse plane. Measure the area of the traverse plane, A3. 3. Determine Ps5 by averaging the pressure measurements at each of four static pressure taps located near the end of the duct connection at the fan outlet. Determine Ps4 by using static pressure taps in the duct connection at the fan inlet. Measure A4 and A5, the cross-sectional areas of the duct connections at the static pressure taps.
4. Measure td3 and tw3 in the traverse plane. Determine pb for the general vicinity of the fan. Measure td4. These measurements are used in determining densities at the various planes of interest. 5. Measure the fan speed, motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV) and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Motor performance data, supplied by the motor manufacturer, are used in the determination of motor power output for this example. 6. SEF 1 is due to the effect of insufficient length of duct between the fan inlet and the elbow upstream of the fan. SEF 2 is due to the effect of insufficient length of duct between the fan outlet and the elbow downstream of the fan. In order to calculate the values of the SEFs, it is necessary to measure the inlet area and the outlet area of the fan, A1 and A2; and the lengths of the inlet and outlet duct connections, L1 and L2.
51
AMCA 203-90 (R2007) 7. To calculate the Fan Static Pressure:
460 volts, 1760 rpm, 24.6 FLA
Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2
GENERAL
Where:
Fan direct connected to motor. Motor efficiency data supplied by motor manufacturer. Fan speed measurement was not obtained due to the closed duct arrangements on both sides of the fan. The measured amps indicate that the motor is operating very close to the full load condition, so the rpm was assumed to be the motor nameplate value of 1760.
Pv1 = Pv3 (A3/A1)2 (ρ3/ρ1) Ps2 and Ps1 are calculated using measured static pressure values and constant total pressure considerations. Ps1 + Pv1 = Ps4 + Pv4 Ps2 + Pv2 = Ps5 + Pv5 Where each velocity pressure is calculated in a manner similar to the calculation of Pv1, shown above. 8. In order to compare the test results to the quoted fan curve drawn for operation at 1750 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS pb td3 tw3 td4 Ps3 Pv3 Ps4 Ps5 A1 A3 A4 L1 L2
= = = = = = = = = = = = = = =
29.76 in. Hg 82.8°F 57.2°F 80°F 0.5 in. wg 0.783 in. wg -1.1 in. wg 0.82 in. wg A2 7.1 ft2 A5 4.91 ft2 6.2 ft2 3.0 ft 3.5 ft
CALCULATIONS DENSITIES For Plane 3 conditions of: td3 = 82.8°F tw3 = 57.2°F p3 = pb + (Ps3/13.6) = 29.76 + (0.5/13.6) = 29.80 in. Hg Use Figure N.1 in Annex N to obtain ρ3 = 0.0728 lbm/ft3. It is assumed ρ2 = ρ5 = ρ3. The density at Planes 1 and 4:
ρ1 = ρ 4 ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ = ρ3 ⎜ s4 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d4 + 460 ⎠ ⎛ −1.1 + 13.6 × 29.76 ⎞ ⎛ 542.8 ⎞ = 0.0728 ⎜ ⎟ ⎜ 540 ⎟ 13.6 × 29.80 ⎝ ⎠⎝ ⎠ = 0.0729 lbm/ft 3 FLOW RATES V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.783/0.0728)0.5 = 3594 fpm
MEASURED MOTOR DATA Volts = = Amps = = kW =
460, 461, 459 460 av 25.0, 25.0, 24.8 24.9 av 18.0
MOTOR NAMEPLATE DATA 20 hp, 3 phase, 60 hertz 52
Q3 = V3A3 = 3594 × 4.91 = 17647 cfm Q = = = =
Q1 Q3 (ρ3/ρ1) 17647 (0.0728/0.0729) 17623 cfm
AMCA 203-90 (R2007) FAN POWER INPUT
Diameter of the fan outlet:
The data supplied by the motor manufacturer indicate motor efficiency of 87.5% at the measured power input of 18.0 kW. Using this information:
D2 = (4A2/π)0.5 = (4 × 7.1/π)0.5 = 3.01 ft
Hmo = (18.0 × 0.875)/0.746 = 21.1 hp
Figure 8.1 shows that for velocities of 2500 fpm or less, the 100% effective duct length is 2.5 diameters:
Since the fan is direct connected to the motor, there is no drive loss, and: H = Hmo = 21.1 hp SYSTEM EFFECT FACTORS To determine the value of SEF 1, calculate the velocity at the fan inlet: V1 = (Q1/A1) = (17623/7.1) = 2482 fpm
= 2.5 × 3.01 = 7.53 ft The length of the outlet duct in % effective duct length: = (L2/7.53) 100 = (3.5/7.53) 100 = 46% From Figure 8.4, for a vaneaxial fan with a 46% effective duct length between its discharge and a two piece elbow, System Effect Curve W applies. From Figure 7.1 for 2485 fpm velocity and curve W, SEF 2 is less than 0.1 in. and is considered negligible.
Diameter of the fan inlet: SEF 2 = 0.00 D1 = (4A1/π)0.5 = (4 × 7.1/π)0.5 = 3.01 ft The length of the duct between the elbow and the fan inlet in terms of the fan inlet diameter: = (L1/D1) = (3.0/3.01) = 1.00 AMCA Publication 201-90, Figure 9.2 indicates that for a two piece elbow with a length of duct between the elbow and the fan inlet equal to 1.00 diameter System Effect Curve S-T applies. For a velocity of 2482 fpm and curve S-T, Figure 7.1 shows SEF 1 = 0.25 in. wg at 0.075 lbm/ft3. At 0.0729 lbm/ft3: SEF 1 = 0.25 (0.0729/0.075) = 0.24 in. wg For SEF 2, AMCA Publication 201-90, Figures 7.1, 8.1, and 8.4 indicate the following calculations: Q2 = Q3 (ρ3/ρ2) = 17647 (0.0728/0.0728) = 17647 cfm V2 = Q2/A2 = 17647/7.1 = 2485 fpm
FAN STATIC PRESSURE Pv5 = Pv3 (A3/A5)2 (ρ3/ρ5) = 0.783 (4.91/4.91)2 (0.0728/0.0728) = 0.783 in. wg Pv2 = Pv3 (A3/A2)2 (ρ3/ρ2) = 0.783 (4.91/7.1)2 (0.0728/0.0728) = 0.37 in. wg Ps2 + Pv2 = Ps5 + Pv5 Ps2 = Ps5 + Pv5 - Pv2 = 0.82 + 0.783 - 0.37 = 1.23 in. wg Pv4 = Pv3 (A3/A4)2 (ρ3/ρ4) = 0.783 (4.91/6.2)2 (0.0728/0.0729) = 0.49 in. wg Pv1 = Pv3 (A3/A1)2 (ρ3/ρ1) = 0.783 (4.91/7.1)2 (0.0728/0.0729) = 0.37 in. wg Ps1 + Pv1 = Ps4 + Pv4 Ps1 = Ps4 + Pv4 - Pv1 = -1.1 + 0.49 - 0.37 = -0.98 in. wg
53
AMCA 203-90 (R2007) Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2 = 1.23 - (-0.98) - 0.37 + 0.24 + 0 = 2.08 in. wg CONVERSION TO SPECIFIED CONDITIONS Qc = 17623 (1750/1760) = 17523 cfm Psc = 2.08 (1750/1760)2 (0.075/0.0729) = 2.12 in. wg Hc = 21.1 (1750/1760)3 (0.075/0.0729) = 21.3 hp
54
AMCA 203-90 (R2007)
EXAMPLE 2G: HIGH PRESSURE CENTRIFUGAL FAN IN A SERIES
3 2b
STATIC PRESSURE TAPS 1b 2a
FAN B
1a
DAMPER
INLET BOX
INLET BOX
FAN A SIDE VIEW
COMMENTS 1. The two single inlet fans in this example have been rated by the manufacturer as a two stage assembly. Although rated as an assembly, sufficient measurements are made to provide performance data for each fan. The damper downstream of the second fan is not included as part of the rated assembly. In virtually all cases in which an air flow control damper, such as the one shown in the diagram, is included in the system, the point of operation of major interest and for which the fan has been selected is at the maximum air flow rate. This example is no exception. Therefore, it is essential that the damper be fixed in its full open position for the duration of the test. 2. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3. Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. Ps3 is used in determining the density at the traverse plane, A3, which is located at the tip of the Pitot-static tube. 3. Determine the static pressures at Planes 1a, 1b2a, and 2b. As shown in the diagram, these planes are located shortly downstream of the inlets and outlets of the fans, which are the planes of interest. In each case, the conditions which exist at the plane of
measurements are assumed to exist at the respective plane of interest because of the close proximity and the fact that the two planes are equal in area. The static pressure at each plane may be determined by averaging the static pressure measurements at each of four static pressure taps, or by averaging the static pressure measurements made in a Pitot-static tube traverse of the plane. However, due to the turbulence existing in the regions of the outlets of the fans, it is recommended that static pressure taps be used at Planes 1b-2a and 2b. 4. Measure td3, tw3, td1b, and td2b; td1a is assumed to be equal to td3. Determine pb for the general vicinity of the fan. These measurements are used in determining densities at the planes of interest. 5. Measure the fan speed and the motor amps, volts, and if possible, watts for each fan. Record all pertinent motor nameplate data, including volts (NPV), and full load amps (FLA). If the motor power outputs are to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drives and measure the no load amps (NLA) if the motors are not operating at or near their full load points. In this example, a watts input measurement is made for 55
AMCA 203-90 (R2007) each motor and motor performance data, supplied by the motor manufacturer, are used in determining motor power outputs. 6. The SEF which would normally be attributed to insufficient length of duct at the outlet of the first stage fan does not apply in this case because the fans have been rated as an assembly. 7. To calculate the static pressure for the two stage assembly:
Second Stage Volts = 4080, 4040, 4020 = 4047 av Amps = 44, 44.5, 45 = 44.5 av kW = 272 MOTOR NAMEPLATE DATA Data identical for each stage: 350 hp, 3 phase, 60 hertz 4000 volts, 1790 rpm, 44.5 FLA
Ps = Ps2b - Ps1a - Pv1a GENERAL Where: Pv1a = Pv3 (A3/A1a)2 (ρ3/ρ1a) 8. In order to compare the test results to the performance quoted for the two stage assembly for operation at 1780 rpm and 0.045 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS pb td3 tw3 td1b
= = = = = td2b = Pv3 = Ps3 = Ps1b = = Ps2b = Na = Nb = A1a = = A3 =
28.64 in. Hg 35°F 33°F td2a 95°F 147°F 0.745 in. wg -150 in. wg Ps2a -79.5 in. wg 0.5 in. wg 1790 rpm, first stage fan speed 1790 rpm, second stage fan speed A2a = A1b = A2b 5.6 ft2 4.92 ft2
MEASURED MOTOR DATA First Stage Volts = = Amps = = kW =
56
4000, 4040, 4080 4040 av 44.5, 45, 45.5 45 av 278
Fans direct connected to motors. Motor efficiency data supplied by motor manufacturer. CALCULATIONS DENSITIES For Plane 3 conditions of: td3 = 35°F tw3 = 33°F p3 = pb + (Ps3/13.6) = 28.64 + (-150/13.6) = 17.61 in. Hg Use the modified Apjohn equation for partial vapor pressure and the density equation based on perfect gas relationships, both of which are described in Annex M, and the data in Figure N.2 in Annex N to calculate the density at Plane 3. pe = 0.1879 in. Hg p3 (t d3 − t w3 ) 2700 17.61(35 − 33) = 0.1879 − 2700 = 0.1749 in. Hg
pp = pe −
ρ3 = =
1.3257( p3 − 0.378 pp ) t d3 + 460 1.3257 (17.61 − 0.378 × 0.1749 )
35 + 460 = 0.0470 lbm/ft 3
Any conversion of velocity pressure to static pressure which may occur between Planes 3 and 1a can be ignored with no significant effect on the accuracy of the test results. Therefore:
AMCA 203-90 (R2007) Ps1a = Ps3 = -150 in. wg
FAN POWER INPUT
Assuming no change in temperature between Planes 3a and 1a:
At the measured power input values of 278 kW and 272 kW, the data supplied by the motor manufacturer indicate efficiency of 95% for each motor.
ρ1a = ρ3 = 0.0470 lbm/ft3
Hmoa = (278 × 0.95)/0.746 = 354 hp
To provide information regarding the flow rates between stages and leaving the second stage, additional density values are calculated as follows:
Hmob = (272 × 0.95)/0.746 = 346 hp
ρ1b = ρ2a ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ = ρ3 ⎜ s1b ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d4 + 460 ⎠ ⎛ −79.5 + 13.6 × 28.64 ⎞ ⎛ 495 ⎞ = 0.0470 ⎜ ⎟ ⎜ 555 ⎟ 13.6 × 17.61 ⎝ ⎠⎝ ⎠
Since each fan is direct connected to its motor, there are no drive losses and: Ha
= Hmoa = 354 hp
Hb
= Hmob = 346 hp
= 0.0543 lbm/fft 3
FAN STATIC PRESSURE ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ2b = ρ3 ⎜ s2b ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d2b + 460 ⎠ ⎛ 0.5 + 13.6 × 28.64 ⎞ ⎛ 495 ⎞ = 0.0470 ⎜ ⎟⎜ ⎟ ⎝ 13.6 × 17.61 ⎠ ⎝ 607 ⎠ = 0.0624 lbm/ftt 3 FLOW RATES V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.745/0.0470)0.5 = 4364 fpm Q3 = V3A3 = 4364 × 4.92 = 21471 cfm Q
= = = =
Q1a Q3 (ρ3/ρ1a) 21471 (0.0470/0.0470) 21471 cfm
Q1b = = = =
Q2a Q3 (ρ3/ρ2a) 21471 (0.0470/0.0543) 18584 cfm
Pv1a = Pv3 (A3/A1a)2 (ρ3/ρ1a) = 0.745 (4.92/5.6)2 (0.0470/0.0470) = 0.575 in. wg The static pressure for the two stage assembly: Ps
= Ps2b - Ps1a - Pv1a = 0.5 - (-150) - 0.575 = 149.9 in. wg
CONVERSION TO SPECIFIED CONDITIONS Qc
= 21471 (1780/1790) = 21351 cfm
Psc
= 149.9 (1780/1790)2 (0.045/0.0470) = 141.9 in. wg
Hac = 354 (1780/1790)3 (0.045/0.0470) = 333 hp Hbc = 346 (1780/1790)3 (0.045/0.0470) = 326 hp
Q2b = Q3 (ρ3/ρ2b) = 21471 (0.0470/0.0624) = 16172 cfm
57
AMCA 203-90 (R2007)
EXAMPLE 3A: CENTRIFUGAL FAN IN AN EXHAUST SYSTEM
AIR INTAKE VENTS BACKDRAFT DAMPER SEF 1 3a
2 3c
3b
1 STATIC PRESSURE TAPS PLAN VIEW
COMMENTS 1. This fan, as supplied and rated by the manufacturer, does not include the backdraft damper. 2. Normally, velocity pressure measurements would be made in a single plane, located in a duct common to all branches. In this example, a measurement plane which provides a satisfactory velocity profile cannot be located within the short length of duct between the point of connection of the branch ducts and the fan inlet. The alternative, as indicated in the diagram, is to make a velocity pressure measurement traverse in the longest available duct run of each branch. The velocity pressure for each branch is determined by using the root mean square of the velocity pressure measurements made in the traverse. The static pressure at each traverse plane is determined by averaging the static pressure measurements made in the same traverse. These static pressure values are used in determining the densities at the traverse planes. Procedures for traverses are described in Section 9.4. In order to determine the air flow rates it is necessary to measure the area of each traverse point. 3. Ps1, the static pressure at the fan inlet may be determined by averaging the static pressure measurements at each of four static pressure taps or by averaging the static pressure measurements made in a Pitot-static tube traverse of Plane 1. If a 58
Pitot-static tube is used, it should be positioned well within the inlet collar in which Plane 1 is located. Measure the area of Plane 1 for use in calculating Pv1. The static pressure at the outlet of the backdraft damper is zero gauge pressure, referred to the atmospheric pressure in the region of the outlet of the backdraft damper. In situations such as this example, the air may be discharging from the damper into a region in which the atmospheric pressure is somewhat different from that to which all other pressure measurements are referred. When this possibility exists, it is essential that the static pressure in the region of the discharging air be measured, referred to the same atmospheric pressure as used in all other pressure measurements. 4. Measure the dry-bulb and wet-bulb temperatures at each velocity traverse plane and the dry-bulb temperature at Plane 1. In this example, td2 is assumed to be equal to td1. Determine pb for the general vicinity of the fan. These measurements are used in determining densities at the planes of interest. 5. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV) and full load amps (FLA). If the motor power output is to be
AMCA 203-90 (R2007) estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K. 6. SEF 1 is due to the effect of there being no duct at the fan outlet. In order to calculate the value of SEF 1, it is necessary to measure the outlet area of the fan, A2, and the blast area of the fan. 7. Determine the backdraft damper pressure loss by using the performance ratings supplied by the manufacturer and the pressure loss multiplier data in Figure 8.7 of AMCA Publication 201-90. The use of the multiplier is indicated because the damper is mounted directly to the fan outlet.
= 0.765 in. wg = 0.88 in. wg = 0.86 in. wg = 800 rpm = 16.8 ft2 = 13.8 ft2 = 5.4 ft2 = A3c = 3.0 ft2 Blast Area = 11.0 ft2
Pv3a Pv3b Pv3c N A1 A2 A3a A3b
MEASURED MOTOR Volts = = Amps = = NLA =
460, 458, 462 460 av 28, 27, 26 27 av 14.7
8. To calculate the Fan Static Pressure: MOTOR NAMEPLATE DATA Ps = Ps2 - Ps1 - Pv1 + SEF 1 Where:
25 hp, 3 phase, 60 hertz 460 volts, 1760 rpm, 32 FLA
Pv1 = (Q1/1096 A1)2 ρ1
GENERAL
Q1 = Q3a (ρ3a/ρ1) + Q3b (ρ3b/ρ1) + Q3c (ρ3c/ρ1)
Fan connected to motor through belt drive. Pressure loss data supplied by manufacturer of backdraft damper.
Ps2 is the sum of the static pressure in the region of the damper outlet, which was measured as zero, and the backdraft damper pressure loss. 9. In order to compare the test results to the quoted fan curve drawn for operation at 810 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS pb td1 tw1 td3a tw3a td3b tw3b td3c tw3c Ps1 Ps3a Ps3b Ps3c
= = = = = = = = = = = = =
29.8 in. Hg 72°F 62°F 77°F 67°F 65°F 56°F 70°F 62°F -1.00 in. wg -0.80 in. wg -0.45 in. wg -0.040 in. wg
CALCULATIONS DENSITIES Since the static pressure values at Planes 1, 3a, 3b, and 3c are very small, no appreciable error will occur by using the barometric pressure instead of the absolute pressure at each plane in the determination of the densities. The densities at these planes are obtained by using Figure N.1 in Annex N.
ρ1 ρ3a ρ3b ρ3c
= = = =
0.0739 0.0731 0.0750 0.0741
lbm/ft3 lbm/ft3 lbm/ft3 lbm/ft3
FLOW RATES V3a = 1096 (Pv3a/ρ3a)0.5 = 1096 (0.765/0.0731)0.5 = 3546 fpm V3b = 1096 (Pv3b/ρ3b)0.5 = 1096 (0.88/0.0750)0.5 = 3754 fpm 59
AMCA 203-90 (R2007) V3c = 1096 (Pv3c/ρ3c)0.5 = 1096 (0.86/0.0741)0.5 = 3734 fpm Q3a = V3aA3a = 3546 × 5.4 = 19148 cfm Q3b = V3bA3b = 3754 × 3.0 = 11262 cfm Q3c = V3cA3c = 3734 × 3.0 = 11202 cfm Q = Q1 = Q3a ( ρ3a / ρ1 ) + Q3 b ( ρ3 b / ρ1 ) + Q3c ( ρ3c / ρ1 ) ⎛ 0.0731 ⎞ ⎛ 0.0750 ⎞ ⎛ 0.0741 ⎞ = 19148 ⎜ ⎟ + 11262 ⎜ 0.0739 ⎟ + 11202 ⎜ 0.0739 ⎟ ⎝ 0.0739 ⎠ ⎝ ⎠ ⎠ ⎝ = 41603 cfm
AMCA Publication 201-90, Figures 7.1 and 8.3 indicate the following calculations: Q2 = Q1 = 41603 cfm It is assumed that ρ2 = ρ1. V2 = (Q2/A2) = (41603/13.8) = 3015 fpm Blast area ratio = Blast area/A2 = 11.0/13.8 = 0.80 For a blast area ratio of 0.8 and no duct, Figure 8.3 shows System Effect Curve T-U applies. For 3015 fpm velocity and curve T-U, Figure 7.1 shows SEF 1 = 0.27 in. wg at 0.075 lbm/ft3 density. At 0.0739 lbm/ft3:
FAN POWER INPUT
SEF 1 = 0.27 (0.0739/0.075) = 0.27 in. wg
Measured amps/FLA = (27/32) = 0.84 = 84%
BACKDRAFT DAMPER LOSS MULTIPLIER
Annex K indicates that the average of the results of Equation A and Equation B will provide a reasonably accurate estimate of motor power output for a 25 hp motor operating at 84% FLA. Eqn A = 25 (27/32) (460/460) = 21.1 hp
The data supplied by the manufacturer of the damper indicate that the pressure loss for the damper, ΔPs, is 0.4 in. wg at the flow rate of 41603 cfm at 0.075 lbm/ft3 density. AMCA Publication 201-90, Figure 8.7 indicates a ΔPs multiplier of 1.9 for a damper which is mounted directly to the outlet of a fan which has a blast area ratio of 0.8.
Eqn B = 25 [(27 - 14.7)/(32 - 14.7)] (460/460) = 17.8 hp
Backdraft damper loss = ΔPs × 1.9 × (ρ2/0.075) = 0.4 × 1.9 (0.0739/0.075) = 0.75 in. wg
Hmo
FAN STATIC PRESSURE
= (21.1 + 17.8)/2 = 19.45 hp
Figure L.1 in Annex L indicates estimated belt drive loss of 4.8%.
Pv1 = (Q1/1096 A1)2 ρ1 = [41603/(1096 × 16.8)]2 0.0739 = 0.38 in. wg
HL = 0.048 Hmo = 0.048 × 19.45 = 0.93 hp
Ps2 is equal to the static pressure at the outlet of the damper, which is zero, plus the damper loss.
H = Hmo - HL = 19.45 - 0.93 = 18.52 hp
60
SYSTEM EFFECT FACTOR
Ps2 = = = Ps = = =
0 + damper loss 0 + 0.75 0.75 in. wg Ps2 - Ps1 - Pv1 + SEF 1 0.75 - (-1.0) - 0.38 + 0.27 1.64 in. wg
AMCA 203-90 (R2007) CONVERSION TO SPECIFIED CONDITIONS Qc = 41603 (810/800) = 42123 cfm Psc = 1.64 (810/800)2 (0.075/0.0739) = 1.71 in. wg Hc = 18.52 (810/800)3 (0.075/0.0739) = 19.51 hp
61
AMCA 203-90 (R2007)
EXAMPLE 3B: AXIAL FAN IN AN EXHAUST SYSTEM 3 2-PIECE ELBOW
SEF 1 L1 1 STATIC PRESSURE TAPS
GUIDE VANES INNER CYLINDER 2 L2
SEF 2 5
PLAN VIEW
COMMENTS 1. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located near the end of a straight run of duct, as shown in the diagram. Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. Ps3 is used in determining the density at the traverse plane. Measure the area of the traverse plane, A3, which is located at the tip of the Pitot-static tube. 2. Determine Ps5 by averaging the pressure measurements at each of four static pressure taps located near the end of the duct connection at the fan outlet. Determine Ps1 by using a Pitot-static tube or static pressure taps in the duct connection at the fan inlet. If a Pitot-static tube is used, it should not project into the upstream elbow but be located well within the length of the duct connection. 3. Measure td3 and tw3 in the traverse plane; td1 is assumed to be equal to td3. Determine pb for the general vicinity of the fan. Measure td5. These measurements are used in determining densities at the planes of interest. 4. Measure the fan speed and the motors amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV), and full 62
load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K. 5. SEF 1 is due to the effect of insufficient length of duct between the fan inlet and the elbow upstream of the fan. SEF 2 is due to the effect of insufficient length of duct between the fan outlet and the elbow downstream of the fan. In order to calculate the values of the SEFs, it is necessary to measure the inlet area and the outlet area of the fan, A1 and A2; and the lengths of the inlet and outlet duct connections, L1 and L2. 6. To calculate the Fan Static Pressure: Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2 Where: Pv1 = Pv3 Since: A1 = A3 And:
AMCA 203-90 (R2007)
ρ1 = ρ3
CALCULATIONS
Due to the close proximity of Planes 2 and 5 and the fact that there is no change in area between the two planes, all conditions which exist at Plane 5 are assumed to exist at Plane 2.
Ps2 = Ps5 7. In order to compare the test results to the quoted fan curve drawn for operation at 1730 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS = = = = = = = = = = = L1 = L2 =
29.20 in. Hg 72°F 66°F 73°F -2.02 in. wg -1.92 in. wg 0.35 in. wg 0.10 in. wg 1710 rpm A2 = A3 = A5 2.64 ft2 1.5 ft, length of inlet duct 2.25 ft, length of the outlet duct
MEASURED MOTOR DATA Volts = = Amps = = NLA =
For Plane 3 conditions of: td3 = 72°F tw3 = 66°F
Therefore:
pb td3 tw3 td5 Ps1 Ps3 Pv3 Ps5 N A1
DENSITIES
227, 229, 228 228 av 12.2, 12.3, 12.4 12.3 av 7
MOTOR NAMEPLATE DATA 5 hp, 3 phase, 60 hertz 230 volts, 1760 rpm, 14.0 FLA GENERAL Fan connected to motor through belt drive.
p3 = pb + (Ps3/13.6) = 29.20 + (-1.92/13.6) = 29.06 in. Hg Use Figure N.1 in Annex N to obtain ρ3 = 0.0719 lbm/ft3. Assume that td1 = td3. ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ1 = ρ3 ⎜ s1 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d1 + 460 ⎠ ⎛ −2.02 + 13.6 × 29.20 ⎞ ⎛ 532 ⎞ = 0.0719 ⎜ ⎟ ⎜ 532 ⎟ 13.6 × 29.06 ⎝ ⎠⎝ ⎠ = 0.0719 lbm/ft 3 Assume that td2 = td5 and Ps2 = Ps5.
ρ 2 = ρ5 ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ = ρ3 ⎜ s5 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d5 + 460 ⎠ ⎛ 0.10 + 13.6 × 29.20 ⎞ ⎛ 532 ⎞ = 0.0719 ⎜ ⎟ ⎜ 533 ⎟ 13.6 × 29.06 ⎝ ⎠⎝ ⎠ = 0.0721 lbm/ft 3 FLOW RATES V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.35/0.0719)0.5 = 2418 fpm Q3 = V3A3 = 2418 × 2.64 = 6384 cfm Q = = = =
Q1 Q3 (ρ3/ρ1) 6384 (0.0719/0.0719) 6384 cfm
Q2 = = = =
Q5 Q3 (ρ3/ρ5) 6384 (0.0719/0.0721) 6366 cfm 63
AMCA 203-90 (R2007) FAN POWER INPUT Measured amps/FLA = (12.3/14.0) = 0.88 = 88% Annex K indicates that the average of the results of Equation A and Equation B will provide a reasonably accurate estimate of motor power output for a 5 hp motor operating at 88% FLA.
R-S, Figure 7.1 shows SEF 1 = 0.24 in. wg at 0.075 lbm/ft3 density. At 0.0719 lbm/ft3: SEF 1 = 0.24 (0.0719/0.075) = 0.23 in. wg For SEF 2, AMCA Publication 201-90, Figures 7.1, 8.1, and 8.4 indicate the following calculations:
Eqn A = 5 (12.3/14) (228/230) = 4.35 hp
V2 = (Q2/A2) = (6366/2.64) = 2411 fpm
Eqn B = 5 [(12.3 - 7)/(14 - 7)] (228/230) = 3.75 hp
The diameter of the fan outlet:
Hmo
= (4.35 + 3.75)/2 = 4.05 hp
Figure L.1 in Annex L indicates estimated belt drive loss of 6.3%. HL = 0.063 Hmo = 0.063 × 4.05 = 0.26 hp H = Hmo - HL = 4.05 - 0.26 = 3.79 hp SYSTEM EFFECT FACTORS To determine the value of SEF 1, calculate the velocity at the fan inlet: V1 = (Q1/A1) = (6384/2.64) = 2418 fpm
D2 = (4A2/π)0.5 = (4 × 2.64/π)0.5 = 1.83 ft Figure 8.1 shows that for velocities of 2500 fpm or less, the 100% effective duct length is 2.5 diameters: = 2.5 × 1.83 = 4.58 ft The length of the outlet duct in % effective duct length: = (L2/4.58) 100 = (2.25/4.58) 100 = 49% From Figure 8.4, for a vaneaxial fan with a 49% effective duct length between its discharge and a two piece elbow, System Effect Curve W applies. From Figure 7.1, for 2411 fpm velocity and curve W, SEF 2 is less than 0.1 in. wg, and is considered negligible. SEF 2 = 0.00
Calculate the diameter of the fan inlet: FAN STATIC PRESSURE D1 = (4A1/π)0.5 = (4 × 2.64/π)0.5 = 1.83 ft. Calculate the length of duct between the elbow and the fan inlet in terms of the fan inlet diameter: = (L1/D1) = (1.5/1.83) = 0.82 AMCA Publication 201-90, Figure 9.2, indicates that for a vaneaxial fan with a two piece elbow with a length of duct between the elbow and the fan inlet equal to 0.8 diameters, System Effect Curve R-S (estimated) applies. For 2418 fpm velocity and curve 64
Since: A 1 = A3 ρ1 = ρ3 Pv1 = Pv3 Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2 = 0.10 - (-2.02) - 0.35 + 0.23 + 0.00 = 2.00 in. wg
AMCA 203-90 (R2007) CONVERSION TO SPECIFIED CONDITIONS Qc = 6384 (1730/1710) = 6459 cfm Psc = 2.00 (1730/1710)2 (0.075/0.0719) = 2.14 in. wg Hc = 3.79 (1730/1710)3 (0.075/0.0719) = 4.09 hp
65
AMCA 203-90 (R2007)
EXAMPLE 3C: CENTRIFUGAL FAN IN A SCRUBBER SYSTEM 3
WET CELL SCRUBBER
1
2 SEF 1
PLAN VIEW
SIDE VIEW
COMMENTS 1. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located in the duct connection at the fan inlet, as shown in the diagram. Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. Ps3 is used in determining the density at the traverse plane. Measure the area of the traverse plane, A3, which is located at the tip of the Pitot-static tube. In locating Plane 3 downstream of the scrubber, changes in the composition of the air as a result of the action of the scrubber are properly taken into account in the determination of fan air flow rate. Due to the close proximity of Planes 1 and 3, and the fact that there is no change in area between the two planes, the conditions which exist at Plane 3 are assumed to exist at Plane 1.
amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K.
2. Ps2, the static pressure at the fan outlet, is zero.
Pv1 = Pv3 Ps1 = Ps3 Ps2 = 0
3. Measure td3 and tw3 in the traverse plane. Determine pb for the general vicinity of the fan. Measure td2. These measurements are used in determining densities at the planes of interest. 4. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV), and full load 66
5. SEF 1 is due to the effect of there being no duct at the fan outlet. In order to calculate the value of SEF 1, it is necessary to measure the outlet area of the fan, A2, and the blast area of the fan. 6. To calculate the Fan Static Pressure: Ps = Ps2 - Ps1 - Pv1 + SEF 1 Where:
7. In order to compare the test results to the quoted fan curve drawn for operation at 1700 rpm and 0.071 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14.
AMCA 203-90 (R2007) OBSERVATIONS SITE MEASUREMENTS = 29.80 in. Hg = 65°F = 64°F = 70°F = -8.0 in. wg = 0.337 in. wg = 1672 rpm = A3 = 7.06 ft2 A2 = 5.15 ft2 Blast Area = 3.67 ft2
pb td3 tw3 td2 Ps3 Pv3 N A1
MEASURED MOTOR DATA Volts = = Amps = =
450, 458, 462 457 av 44, 45, 44.5 44.5 av
MOTOR NAMEPLATE DATA 40 hp, 3 phase, 60 hertz 460 volts, 1780 rpm, 49 FLA GENERAL Fan connected to motor through belt drive. CALCULATIONS
⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ2 = ρ3 ⎜ s2 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d2 + 460 ⎠ ⎛ 0 + 13.6 × 29.80 ⎞ ⎛ 525 ⎞ = 0.0732 ⎜ ⎟⎜ ⎟ ⎝ 13.6 × 29.21 ⎠ ⎝ 530 ⎠ = 0.0740 lbm/ft 3 FLOW RATES V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.337/0.0732)0.5 = 2352 fpm Q3 = V3A3 = 2353 × 7.06 = 16605 cfm Q = = = =
Q1 Q3 (ρ3/ρ1) 16605 (0.0732/0.0732) 16605 cfm
Q2 = Q3 (ρ3/ρ2) = 16605 (0.0732/0.0740) = 16425 cfm FAN POWER INPUT Measured amps/FLA = (44.5/49) = 0.91 = 91%
DENSITIES
Annex K indicates that Equation A will provide a reasonably accurate estimate of motor power output for a 40 hp motor operating at 91% FLA.
For Plane 3 conditions of: td3 = 65°F tw3 = 64°F
Hmo = 40 (44.5/49) (457/460) = 36.1 hp
p3 = pb + (Ps3/13.6) = 29.80 + (-8.0/13.6) = 29.21 in. Hg Use Figure N.1 in Annex N to obtain ρ3 = 0.0732 lbm/ft3. It is assumed that: td1 = td3 Ps1 = Ps3 ρ1 = ρ3
Figure L.1 in Annex L indicates estimate belt drive loss of 4.5%. HL = 0.045 Hmo = 0.045 × 36.1 = 1.6 hp H = Hmo - HL = 36.1 - 1.6 = 34.5 hp SYSTEM EFFECT FACTOR AMCA Publication 201-90, Figures 7.1 and 8.3, indicate the following calculations:
67
AMCA 203-90 (R2007) V2 = (Q2/A2) = (16425/5.15) = 3189 fpm Blast area ratio = Blast area/A2 = 3.67/5.15 = 0.71 For a blast area ratio of 0.7 and no duct, Figure 8.3 shows System Effect Curve S applies. For 3189 fpm velocity and curve S, Figure 7.1 shows SEF 1 = 0.5 in. wg at 0.075 lbm/ft3 density. At 0.0740 lbm/ft3: SEF 1 = 0.5 (0.074/0.075) = 0.49 in. wg FAN STATIC PRESSURE Pv1 = Pv3 = 0.337 in. wg Ps = Ps2 - Ps1 - Pv1 + SEF 1 = 0 - (-8.0) - 0.337 + 0.49 = 8.15 in. wg CONVERSION TO SPECIFIED CONDITIONS Qc = 16605 (1700/1672) = 16883 cfm Psc = 8.15 (1700/1672)2 (0.071/0.0732) = 8.17 in. wg Hc = 34.5 (1700/1672)3 (0.071/0.0732) = 35.2 hp
68
AMCA 203-90 (R2007)
EXAMPLE 3D: CENTRIFUGAL ROOF VENTILATOR WITH DUCTED INLET
2 1
BACKDRAFT DAMPER 4 STATIC PRESSURE TAPS
3a
SIDE VIEW
3b
COMMENTS 1. This centrifugal roof ventilator, as supplied and rated by the manufacturer, does not include the backdraft damper. It is essential that the backdraft damper blades be fixed in their full open positions, otherwise uneven velocity distribution will occur at the inlet to the ventilator, adversely affecting its performance. 2. Normally, velocity pressure measurements would be made in a single plane, located in a duct common to all branches. In this example, a measurement plane which provides a satisfactory velocity profile cannot be located within the short length of duct between the point of connection of the branch ducts and the ventilator inlet. The alternative, as indicated in the diagram, is to make a velocity pressure measurement traverse in each branch. The velocity pressure for each branch is determined by using the root mean square of the velocity pressure measurements made in the traverse. The static pressure at each traverse plane is determined by averaging the static pressure measurements made in the same traverse. These static pressure values are used in determining the densities at the traverse planes. Procedures for traverses are described in Section 9.4. In order to determine the air flow rates, it is necessary to measure the area of each traverse plane.
3. Ps4 may be determined by averaging the static pressure measurements at each of four static pressure taps or by averaging the static pressure measurements made in a Pitot-static tube traverse of Plane 4. If a Pitot-static tube is used, it should be positioned well within the duct in which Plane 4 is located, and not project into the upstream elbows. Measure the area of Plane 1 for use in calculating Pv1. In this example, A4 = A1. Ps2, the static pressure at the outlet of the ventilator, is zero gauge pressure, referred to the atmospheric pressure in the region of the ventilator outlet. In situations such as this example, the air may be discharging from the ventilator into a region in which the atmospheric pressure is somewhat different from that to which all other pressure measurements are referred. When this possibility exists, it is essential that the static pressure in the region of the discharging air be measured, referred to the same atmospheric pressure as used in all other pressure measurements. In this case, Ps2 was measured as zero. 4. Measure the dry-bulb and wet-bulb temperatures at each velocity traverse plane. In this example, td1 and td4 are assumed to be equal to td3a. Determine pb for the general vicinity of the fan. These measurements are used in determining densities at the planes of interest. 69
AMCA 203-90 (R2007) 5. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV) and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K.
MEASURED MOTOR DATA Volts = = Amps = =
450, 455, 460 455 av 5.7, 5.85, 5.9 5.82 av
MOTOR NAMEPLATE DATA 5 hp, 3 phase, 60 hertz 460 volts, 1780 rpm, 5.95 FLA
6. Determine the backdraft damper pressure loss by using the performance ratings supplied by the manufacturer.
GENERAL
7. To calculate the Fan Static Pressure:
Fan connected to motor through belt drive. Pressure loss data supplied by manufacturer of backdraft damper.
Ps = Ps2 - Ps1 - Pv1 CALCULATIONS Where: DENSITIES Pv1 = (Q1/1096 A1)2 ρ1 Q1 = Q3a (ρ3a/ρ1) + Q3b (ρ3b/ρ1) Ps1 = Ps4 - backdraft damper pressure loss Ps2 = 0
td3a = = tw3a = =
td3b 72°F tw3b 66°F
8. In order to compare the test results to the quoted fan curve drawn for operation at 620 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14.
p3a = = = =
p3b pb + (Ps3a/13.6) 29.20 + (-0.85/13.6) 29.14 in. Hg
OBSERVATIONS SITE MEASUREMENTS pb = 29.20 in. Hg td3a = td3b = 72°F tw3a = tw3b = 66°F Ps2 = 0 in. wg Ps4 = -0.88 in. wg Ps3a = Ps3b = -0.85 in. wg Pv3a = 0.27 in. wg Pv3b = 0.275 in. wg N = 625 rpm A1 = A4 = 7.9 ft2 A3a = 3.4 ft2 A3b = 3.3 ft2
70
For Planes 3a and 3b conditions of:
Use Figure N.1 in Annex N to obtain:
ρ3a = ρ3b = 0.0721 lbm/ft3 It is assumed that: td1 = td4 = td3a = td3b Since the differences in the static pressures at Planes 1, 3a, and 4 are very small, no appreciable error will occur by assuming:
ρ1 = ρ4 = ρ3a = ρ3b FLOW RATES V3a = 1096 (Pv3a/ρ3a)0.5 = 1096 (0.27/0.0721)0.5 = 2121 fpm
AMCA 203-90 (R2007) V3b = 1096 (Pv3b/ρ3b)0.5 = 1096 (0.275/0.0721)0.5 = 2140 fpm Q3a = V3aA3a = 2121 × 3.4 = 7211 cfm Q3b = V3bA3b = 2140 × 3.3 = 7062 cfm
BACKDRAFT DAMPER LOSS The data supplied by the manufacturer of the damper indicate that the pressure loss for the damper, ΔPs, is 0.22 in. wg at the flow rate of 14273 cfm at 0.075 lbm/ft3 density. Backdraft damper loss = ΔPs (ρ4/0.075) = 0.22 (0.0721/0.075) = 0.21 in. wg FAN STATIC PRESSURE
Q
= = = =
Q1 Q3a (ρ3a/ρ1) + Q3b (ρ3b/ρ1) 7211 (0.0721/0.0721) + 7062 (0.0721/0.0721) 14273 cfm
FAN POWER INPUT Measured amps/FLA = (5.82/5.95) = 0.98 = 98%
Pv1 = (Q1/1096 A1)2 ρ1 = [14273/(1096 × 7.9)]2 0.0721 = 0.20 in. wg Ps1 = Ps4 - damper loss = -0.88 - 0.21 = -1.09 in. wg Ps = Ps2 - Ps1 - Pv1 = 0 - (-1.09) - 0.20 = 0.89 in. wg
Annex K indicates that Equation A will provide a reasonably accurate estimate of motor power output for a 5 hp motor operating at 98% FLA.
CONVERSION TO SPECIFIED CONDITIONS
Hmo = 5 (5.82/5.95) (455/460) = 4.84 hp
Qc = 14273 (620/625) = 14159 cfm
Figure L.1 in Annex L indicates estimated belt drive loss of 5.8%.
Psc = 0.89 (620/625)2 (0.075/0.0721) = 0.91 in. wg
HL = 0.058 Hmo = 0.058 × 4.84 = 0.28 hp
Hc = 4.56 (620/625)3 (0.075/0.0721) = 4.63 hp
H
= Hmo - HL = 4.84 - 0.28 = 4.56 hp
71
AMCA 203-90 (R2007)
EXAMPLE 4A: CENTRIFUGAL FAN IN A BUILT-UP AIR CONDITIONING UNIT 2 4
RETURN AIR
OUTSIDE AIR
SEF 1 L
SEF 2
PLAN VIEW
SPRAY SECTION
5
3a FAN SECTION
+
+
+
+ +
+
+ +
+
+
PREHEAT COILS FILTER SECTION
DIFFUSER PLATE
REHEAT COIL
3b
SIDE VIEW
COMMENTS 1. This is an air conditioning unit which has been assembled at the installation site. The subject of the test is the fan, which is rated by the manufacturer as free-standing, unencumbered by the cabinet in which it has been installed. The fan performance ratings are based on operation with the fan outlet ducted. Before proceeding with the test, it is essential that all dampers--outside air, return air, mixing box, multizone, face and bypass or volume control--be fixed in the positions agreed upon by all interested parties as being applicable for the installation. Also, the temperatures of the heating coils must be kept constant throughout the test period. It may be necessary to lock out, disconnect, or otherwise modify automatic control devices in order to prevent the positions of the dampers and temperatures of the coils from changing during the test. Refer to Section 17.4.3 for additional considerations affecting the test procedure for fans in this type of installation. 2. Normally, velocity pressure measurements would be made in a single plane, located in a duct common to all branches. In this example, a measurement plane which provides a satisfactory velocity profile cannot be located upstream of the fan or between the point of connection of the branch ducts and the fan outlet. The alternative, as indicated in the diagram, is to make a velocity pressure measurement traverse in each branch. The velocity pressure for each branch 72
is determined by using the root mean square of the velocity pressure measurements made in the traverse. the static pressure at each traverse plane is determined by averaging the static pressure measurements made in the same traverse. These static pressure values are used in determining the densities at the traverse planes. Procedures for traverses are described in Section 9.4. In order to determine the air flow rates, it is necessary to measure the area of each traverse plane. 3. Determine Ps4 by averaging the static pressure measurements made in a traverse of Plane 4. Determine Ps5 in a similar manner. Pitot-static tube traverses are used in determining these static pressures because the installation of suitable pressure taps is usually prevented by the insulating material encountered in this type of equipment. Due to the abrupt expansion in area from Plane 2 to Plane 5, it is assumed that there is no conversion of velocity pressure at Plane 2 to static pressure at Plane 5. Therefore, it is assumed that Ps2 = Ps5. Measure the area of Plane 4 for use in calculating Pv4. 4. Measure the dry-bulb and wet-bulb temperatures at Plane 4 and the dry-bulb temperatures at Planes 3a, 3b, and 5. Determine pb for the general vicinity of the air conditioning unit. These measurements are used in determining densities at the planes of interest.
AMCA 203-90 (R2007) 5. Measure the fan speed and motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV), and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K. 6. SEF 1 is due to the effect of insufficient distance between the fan inlets and the side walls of the fan cabinet. SEF 2 is attributed to the high degree of divergence of the transition fitting at the fan outlet. The effect created by this fitting is considered to be equivalent to the effect created by having no duct at the fan outlet. In order to determine the values of the SEFs, it is necessary to measure the diameter of an inlet of the fan, the distance between a fan inlet and a side wall of the fan cabinet, and the outlet area and blast area of the fan. 7. To calculate the Fan Static Pressure: Ps = Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2 = Ps2 - (Ps1 + Pv1) + SEF 1 + SEF 2 Where: Ps2 = Ps5 Ps1 + Pv1 = Ps4 + Pv4 Pv4 = (Q4/1096 A4)2 ρ4 Q4 = Q1 = Q3a (ρ3a/ρ1) + Q3b (ρ3b/ρ1) The calculation of Pv4 is often ignored in instances similar to this example on the basis that the calculated value of Pv4 is relatively small and its omission does not affect the test results significantly. 8. In order to compare the test results to the quoted fan curve drawn for operation at 1170 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14.
OBSERVATIONS SITE MEASUREMENTS pb = 28.72 in. Hg td3a = 59°F td3b = 90°F td4 = 56°F td5 = 58°F Ps4 = -1.75 in. wg Ps3a = 3.65 in. wg Ps3b = 3.45 in. wg Pv3a = 0.60 in. wg Pv3b = 0.47 in. wg Ps5 = 3.77 in. wg N = 1160 rpm A2 = 18.9 ft2 A3a = 7.2 ft2 A3b = 9.7 ft2 A4 = 93.2 ft2 Blast Area = 13.3 ft2 D1 = 3.92 ft, fan inlet diameter L = 2.83 ft MEASURED MOTOR DATA Volts = = Amps = =
462, 465, 465 464 av 82, 81, 83 82 av
MOTOR NAMEPLATE DATA 75 hp, 3 phase, 60 hertz 460 volts, 1780 rpm, 90.3 FLA GENERAL Fan connected to motor through belt drive. CALCULATIONS DENSITIES For Plane 4 conditions of: td4 = 56°F tw4 = 54°F p4 = pb + (Ps4/13.6) = 28.72 + (-1.75/13.6) = 28.59 in. Hg
73
AMCA 203-90 (R2007) Use Figure N.1 in Annex N to obtain ρ4 = 0.0731 lbm/ft3. It is assumed that ρ1 = ρ4. ⎛ P + 13.6 pb ⎞ ⎛ t d4 + 460 ⎞ ρ5 = ρ 4 ⎜ s5 ⎟ ⎟⎜ ⎝ 13.6 p4 ⎠ ⎝ t d5 + 460 ⎠ ⎛ 3.77 + 13.6 × 28.72 ⎞ ⎛ 516 ⎞ = 0.0731⎜ ⎟ ⎜ 518 ⎟ 13.6 × 28.59 ⎠⎝ ⎠ ⎝ = 0.0739 lbm/ft 3 ⎛ P + 13.6 pb ⎞ ⎛ t d4 + 460 ⎞ ρ3a = ρ 4 ⎜ s3a ⎟ ⎟⎜ ⎝ 13.6 p4 ⎠ ⎝ t d3a + 460 ⎠ ⎛ 3.65 + 13.6 × 28.72 ⎞ ⎛ 516 ⎞ = 0.0731⎜ ⎟ ⎜ 519 ⎟ 13.6 × 28.59 ⎝ ⎠⎝ ⎠ = 0.0737 lbm/ft 3
FAN POWER INPUT Measured amps/FLA = (82/90.3) = 0.91 = 91% Annex K indicates that Equation A will provide a reasonably accurate estimate of motor power output for a 75 hp motor operating at 91% FLA. Hmo = 75 (82/90.3) (464/460) = 68.7 hp Figure L.1 in Annex L indicates estimated belt drive loss of 4.3%. HL = 0.043 Hmo = 0.043 × 68.7 = 2.95 hp H
⎛ P + 13.6 pb ⎞ ⎛ t d4 + 460 ⎞ ρ3b = ρ 4 ⎜ s3b ⎟ ⎟⎜ ⎝ 13.6 p4 ⎠ ⎝ t d3b + 460 ⎠ ⎛ 3.45 + 13.6 × 28.72 ⎞ ⎛ 516 ⎞ = 0.0731⎜ ⎟ ⎜ 550 ⎟ 13.6 × 28.59 ⎝ ⎠⎝ ⎠ = 0.0695 lbm/ft 3
= Hmo - HL = 68.7 - 2.95 = 68.75 hp
SYSTEM EFFECT FACTORS SEF 1 is due to the effect of insufficient distance between the fan inlets and the side walls of the fan plenum. The distance is 2.83 ft, or:
FLOW RATES V3a = 1096 (Pv3a/ρ3a = 1096 (0.60/0.0737)0.5 = 3127 fpm )0.5
V3b = 1096 (Pv3b/ρ3b)0.5 = 1096 (0.47/0.0695)0.5 = 2850 fpm Q3a = V3aA3a = 3127 × 7.2 = 22514 cfm Q3b = V3bA3b = 2850 × 9.7 = 27645 cfm Q = Q1 = Q3a (ρ3a/ρ1) + Q3b (ρ3b/ρ1) = 22514 (0.0737/0.0731) + 27645 (0.0695/0.0731) = 48982 cfm Q2 = Q1 (ρ1/ρ2) = 48982 (0.0731/0.0739) = 48452 cfm
74
(2.83/3.92) = 0.72 = 72% Of the fan inlet diameter. The area of the fan inlets: A1 = 2 (π D12/4) = 2 (π × 3.922/4) = 24.1 ft2 The fan inlet velocity: V1 = (Q1/A1) = (48982/24.1) = 2032 fpm AMCA Publication 201-90, Figure 9.11A, indicates that for a plenum wall spacing of 72% of the fan inlet diameter System Effect Curve V applies. For 2032 fpm inlet velocity and curve V, Figure 7.1 shows SEF 1 = 0.06 in. wg at 0.075 lbm/ft3 density. At 0.0731 lbm/ft3: SEF 1 = 0.06 (0.0731/0.075) = 0.06 in. wg For SEF 2, AMCA Publication 201-90, Figures 7.1 and 8.3, indicate the following calculations:
AMCA 203-90 (R2007) V2 = (Q2/A2) = (48452/18.9) = 2564 fpm Blast area ratio = Blast Area/A2 = 13.3/18.9 = 0.70 For a blast area ratio of 0.7 and no duct, Figure 8.3 shows System Effect Curve S applies. For 2564 fpm velocity and curve S, Figure 7.1 shows SEF 2 = 0.33 in. wg at 0.075 lbm/ft3 density. At 0.0739 lbm/ft3:
CONVERSION TO SPECIFIED CONDITIONS Qc = 48982 (1170/1160) = 49404 cfm Psc = 5.89 (1170/1160)2 (0.075/0.0731) = 6.15 in. wg Hc = 65.75 (1170/1160)3 (0.075/0.0731) = 69.22 hp
SEF 2 = 0.33 (0.0739/0.075) = 0.33 in. wg FAN STATIC PRESSURE Pv4 = (Q4/1096 A4)2 ρ4 Since:
ρ4 = ρ1 Q4 = Q1 Pv4 = (48982/1096 × 93.2)2 0.0731 = 0.02 in. wg Ps1 + Pv1 = Ps4 + Pv4 = -1.75 + 0.02 = -1.73 in. wg Ps = = = =
Ps2 - Ps1 - Pv1 + SEF 1 + SEF 2 Ps2 - (Ps1 + Pv1) + SEF 1 + SEF 2 3.77 - (-1.73) + 0.06 + 0.33 5.89 in. wg
75
AMCA 203-90 (R2007)
EXAMPLE 4B: CENTRAL STATION AIR CONDITIONING UNIT, FACTORY ASSEMBLED DRAWTHROUGH TYPE 1
PLAN VIEW
RETURN AIR
3
STATIC PRESSURE TAPS L 5
SEF 1 +
OUTSIDE AIR
2 +
+
+
FAN SECTION SIDE VIEW
FILTER SECTION
COIL SECTION COMMENTS
1. This is a factory assembled, draw-through central station unit. The subject of the test is the fan section, which is rated by the manufacturer as an assembly of the fan and the cabinet in which the fan has been installed. As a draw-through unit, the performance ratings for the fan section are based on operation with the fan outlet ducted. Before proceeding with the test, it is essential that all dampers--outside air, return air, mixing box, multizone, face and bypass, or volume control--be fixed in the positions agreed upon by all interested parties as being applicable for the installation. Also, the temperatures of heating and cooling coils must be kept constant throughout the test period. It may be necessary to lock out, disconnect, or otherwise modify automatic control devices in order to prevent the positions of the dampers and temperatures of the coils from changing during the test. Refer to Section 17.4.2 for additional considerations affecting the test procedure in this type of installation. 2. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located near the end of a straight run of duct, as shown in the diagram. Determine Ps3 by averaging the static pressure measurements made in the same traverse. This static pressure value is used to determine the density at the traverse plane. Procedures for traverses are described in 76
Section 9.4. In order to determine the air flow rate, it is necessary to measure the area of the traverse plane. 3. Determine Ps1 by averaging the static pressure measurements made in a traverse of Plane 1. Ps5 may be determined in a similar manner or by averaging the pressure measurements at each of four static pressure taps. If it is possible to install suitable pressure taps, their use is preferred in the region of the fan outlet. due to the close proximity of Planes 2 and 5, and the fact that there is no change in area between the two planes, the conditions which exist at Plane 5 are assumed to exist at Plane 2. Measure the area of Plane 1 for use in calculating Pv1. 4. Measure the dry-bulb and wet-bulb temperatures at Plane 3 and the dry-bulb temperatures at Planes 1 and 5. Determine pb for the general vicinity of the air conditioning unit. These measurements are used to determine densities at the planes of interest. 5. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV), and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure
AMCA 203-90 (R2007) motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K. 6. SEF 1 is due to the effect of insufficient length of duct between the fan outlet and the elbow downstream of the fan. In order to determine the value of SEF 1, it is necessary to measure the outlet area of the fan, A2; the length of the outlet duct, L; and the blast area of the fan.
MEASURED MOTOR DATA Volts = = Amps = =
440, 444, 442 442 av 47.4, 47.7, 48.0 47.7 av
MOTOR NAMEPLATE DATA 40 hp, 3 phase, 60 hertz 440 volts, 1770 rpm, 49.7 FLA
7. To calculate the Fan Section Static Pressure:
GENERAL
Ps = Ps2 - Ps1 - Pv1 + SEF 1
Fan connected to motor through belt drive. CALCULATIONS
Where: Ps2 = Ps5
DENSITIES
Pv1 = (Q1/1096A1)2 ρ1
For Plane 3 conditions of:
The calculation of Pv1 is often ignored in instances similar to this example on the basis that the calculated value of Pv1 is relatively small, and it omission does not affect the test results significantly. 8. In order to compare the test results to the quoted fan section curve drawn for operation at 1430 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS = 29.27 in. Hg = 47.5°F = 49.3°F = 47.3°F = 49°F = -0.847 in. wg = 1.31 in. wg = 0.294 in. wg = 1.39 in. wg = 1402 rpm = 147.2 ft2 = A3 = A5 = 15.42 ft2 Blast Area = 9.4 ft2 L = 2.0 ft, length of outlet duct pb td1 td3 tw3 td5 Ps1 Ps3 Pv3 Ps5 N A1 A2
td3 = 49.3°F tw3 = 47.3°F p3 = pb + (Ps3/13.6) = 29.27 + (1.31/13.6) = 29.37 in. Hg Use Figure N.1 in Annex N to obtain ρ3 = 0.0762 lbm/ft3. ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ1 = ρ3 ⎜ s1 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d1 + 460 ⎠ ⎛ −0.847 + 13.6 × 29.27 ⎞ ⎛ 509.3 ⎞ = 0.0762 ⎜ ⎟ ⎜ 507.5 ⎟ 13.6 × 29.37 ⎝ ⎠⎝ ⎠ = 0.0760 lbm/ftt 3 ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ5 = ρ3 ⎜ s5 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d5 + 460 ⎠ ⎛ 1.39 + 13.6 × 29.27 ⎞ ⎛ 509.3 ⎞ = 0.0762 ⎜ ⎟ ⎜ 509 ⎟ 13.6 × 29.37 ⎝ ⎠⎝ ⎠ = 0.0763 lbm/ft 3 It is assumed ρ2 = ρ5. FLOW RATES V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.294/0.0762)0.5 = 2153 fpm
77
AMCA 203-90 (R2007) Q3 = V3A3 = 2153 × 15.42 = 33199 cfm Q = = = =
Q1 Q3 (ρ3/ρ1) 33199 (0.0762/0.0760) 33286 cfm
Q2 = = = =
Q5 Q3 (ρ3/ρ5) 33199 (0.0762/0.0763) 33155 cfm
FAN POWER INPUT Measured amps/FLA = (47.7/49.7) = 0.96 = 96% Annex K indicates that Equation A will provide a reasonably accurate estimate of motor power output for a 40 hp motor operating at 96% FLA. Hmo = 40 (47.7/49.7) (442/440) = 38.6 hp Figure L.1 in Annex L indicates estimated belt drive loss of 4.5%. HL = 0.045 Hmo = 0.045 × 38.6 = 1.74 hp H
= Hmo - HL = 38.6 - 1.74 = 36.86 hp
= 2.5 × 4.43 = 11.1 ft The length of the outlet duct in % effective duct length: = (L/11.1) 100 = (2.0/11.1) 100 = 18% Blast area ratio = Blast Area/A2 = 9.4/15.42 = 0.61 For a blast area ratio of 0.6, 18% effective duct length and elbow position A, Figure 8.5 shows System Effect Curve R applies. For 2150 fpm velocity and curve R, Figure 7.1 shows SEF 1 = 0.34 in. wg at 0.075 lbm/ft3 density. At 0.0762 lbm/ft3: SEF 1 = 0.34 (0.0762/0.075) = 0.35 in. wg FAN SECTION STATIC PRESSURE Pv1 = (Q1/1096 A1)2 ρ1 = (33286/1096 × 147.2)2 0.0760 = 0.003 in. wg It is assumed that Ps2 = Ps5 Ps = Ps2 - Ps1 - Pv1 + SEF 1 = 1.39 - (-0.847) - 0.003 + 0.35 = 2.58 in. wg
SYSTEM EFFECT FACTOR
CONVERSION TO SPECIFIED CONDITIONS
To determine SEF 1, AMCA Publication 201-90, Figures 7.1 and 8.5, indicate the following calculations:
Qc = 33286 (1430/1402) = 33951 cfm
V2 = (Q2/A2) = (33155/15.42) = 2150 fpm Duct diameter equivalent to the fan outlet area: De2 = (4 A2/π)0.5 = (4 × 15.42/π)0.5 = 4.43 ft
78
For velocities of 2500 fpm or less, the 100% effective outlet duct length is 2.5 duct diameters:
Psc = 2.58 (1430/1402)2 (0.075/0.0760) = 2.65 in. wg Hc = 36.86 (1430/1402)3 (0.075/0.0760) = 38.60 hp
AMCA 203-90 (R2007)
EXAMPLE 4C: PACKAGED AIR-CONDITIONING UNIT
3 2 L
SEF 1 PLAN VIEW
4 1 INLET PLENUM
FILTERS
FANS
5
+
+
COOLING COIL
SIDE VIEW
COMMENTS 1. The subject of the test in this example is the air conditioning unit assembly. This assembly does not include the inlet plenum. The performance ratings for the unit assembly are based on operation with the outlets of the fans ducted. Before proceeding with the test, it is essential that all system dampers be fixed in the positions agreed upon by all interested parties as being applicable for the installation. Also, the temperature of the cooling coil must be kept constant throughout the test period. It may be necessary to lock out, disconnect or otherwise modify automatic control devices in order to prevent the positions of the dampers and the temperature of the coil from changing during the test. Refer to Section 17.4.1 for additional considerations affecting the test procedure in this type of installation.
3. Ps4 may be determined by averaging the pressure measurements at each of four static pressure taps or by averaging the static pressure measurements made in a Pitot-static tube traverse of Plane 4. Ps5 is determined in a similar manner. However, if it is possible to install suitable static pressure taps, their use is preferred in the regions of the outlets of the fans. Due to the close proximity of Planes 1 and 4 and the fact that there is no change in area between the two planes, the conditions which exist at Plane 4 are assumed to exist at Plane 1. Although Plane 5 is greater in area that Plane 2, the degree of divergence is relatively small. Therefore, Ps2 will be calculated based on Ps5 and the assumption that there is no change in total pressure from Plane 2 to Plane 5.
2. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located near the end of a straight run of duct, as shown in the diagram. Determine Ps3 by averaging the static pressure measurements made in the same traverse. This static pressure value is used to determine the density at the traverse plane. Procedures for traverses are described in Section 9.4. in order to determine the air flow rate, it is necessary to measure the area of the traverse plane.
4. Measure the dry-bulb and wet-bulb temperatures at Plane 4 and the dry-bulb temperatures at Planes 3 and 5. In this example, the cooling medium, normally circulated in the coil was shut off in order to maintain constant air temperatures during the test. In order to account for water vapor which may have been added to the air as a result of evaporation of moisture previously condensed on the coil, the wet-bulb temperature at Plane 3 was measured. Determine pb for the general vicinity of the air conditioning unit. These measurements are used in determining densities at the planes of interest.
79
AMCA 203-90 (R2007) 5. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data including volts (NPV), and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K.
A 1 = A4 = 31.7 ft2 A2 = 11.5 ft2 A3 = 16.4 ft2 A5 = 14.3 ft2 Blast Area = 4.0 ft2 per fan L = 2.0 ft, length of outlet duct
6. Although an elbow is located shortly downstream of the fans, SEF 1 is judged to be more closely characterized as the effect due to insufficient lengths of duct on the outlets of the fans. In order to determine the value of SEF 1, it is necessary to measure the outlet area and the blast area of one of the fans and the length, L, of its outlet duct.
Volts = = Amps = =
7. To calculate the static pressure for the unit assembly:
MEASURED MOTOR DATA 460, 455, 465 460 av 38.2, 38, 37.9 38.0 av
MOTOR NAMEPLATE DATA 25 hp, 3 phase, 60 hertz 460 volts, 1760 rpm, 39.5 FLA GENERAL
Ps = Ps2 - Ps1 - Pv1 + SEF 1 Fans connected to motor through belt drive. Where: CALCULATIONS Ps1 = Ps4
DENSITIES
Pv1 = (Q1/1096A1)2 ρ1 Ps2 = Ps5 + Pv5 - Pv2 Pv2 and Pv5 are calculated in manners similar to the calculation of Pv1. 8. In order to compare the test results to the quoted unit assembly curve drawn for operation at 1050 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS pb td3 tw3 td4 tw4 td5 Ps3 Pv3 Ps4 Ps5 N 80
= = = = = = = = = = =
29.65 in. Hg 75.0°F 59.5°F 72.5°F 58.5°F 74.5°F 2.02 in. wg 0.35 in. wg -0.32 in. wg 2.11 in. wg 1025 rpm
For Plane 3 conditions of: td3 = 75.0°F tw3 = 59.5°F p3 = pb + (Ps3/13.6) = 29.65 + (2.03/13.6) = 29.80 in. Hg Use Figure N.1 in Annex N to obtain ρ3 = 0.0736 lbm/ft3. For Plane 4 conditions of: td4 = 72.5°F tw4 = 58.5°F p4 = pb + (Ps4/13.5) = 29.65 + (-0.32/13.6) = 29.63 in. Hg Use Figure N.1 in Annex N to obtain ρ4 = 0.0735 lbm/ft3. It is assumed that ρ1 = ρ4.
AMCA 203-90 (R2007) ⎛ P + 13.6 pb ⎞ ⎛ t d3 + 460 ⎞ ρ5 = ρ3 ⎜ s5 ⎟⎜ ⎟ ⎝ 13.6 p3 ⎠ ⎝ t d5 + 460 ⎠ ⎛ 2.11 + 13.6 × 29.65 ⎞ ⎛ 535 ⎞ = 0.0736 ⎜ ⎟ ⎜ 534.5 ⎟ 13.6 × 29.80 ⎝ ⎠⎝ ⎠ = 0.0737 lbm/ft 3
SYSTEM EFFECT FACTOR To determine SEF 1, AMCA Publication 201-90, Figures 7.1 and 8.3, indicate the following calculations:
It is assumed ρ2 = ρ5.
V2 = (Q2/A2) = (39143/11.5) = 3404 fpm
FLOW RATES
Duct diameter equivalent to the outlet area of one fan:
V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.35/0.0736)0.5 = 2390 fpm
De2 = (4A2/2π)0.5 = (4 × 11.5/2π)0.5 = 2.71 ft
Q3 = V3A3 = 2390 × 16.4 = 39196 cfm
Figure 8.3 shows that for velocities over 2500 fpm, 100% effective duct length is one diameter for every 1000 fpm:
Q2 = = = =
Q5 Q3 (ρ3/ρ5) 39196 (0.0736/0.0737) 39143 cfm
Q = = = =
Q1 = Q4 Q3 (ρ3/ρ4) 39196 (0.0736/0.0735) 39249 cfm
FAN POWER INPUT Measured amps/FLA = (38.0/39.5) = 0.96 = 96%
= De2 (V2/1000) = 2.71 (3404/1000) = 9.22 ft L in % effective duct length: = (L/9.22) 100 = (2.0/9.22) 100 = 22% Blast area ratio = Blast area/A2 = (2 × 4.0)/11.5 = 0.70
Annex K indicates that Equation A will provide a reasonably accurate estimate of motor power output for a 25 hp motor operating at 96% FLA.
For a blast area ratio of 0.7, and 22% effective duct length Figure 8.3 shows System Effect Curve W applies. For 3404 fpm velocity and curve W, Figure 7.1 shows SEF 1 = 0.13 in. wg at 0.075 lbm/ft3 density. At 0.0737 lbm/ft3:
Hmo = 25 (38.0/39.5) (460/460) = 24.1 hp
SEF 1 = 0.13 (0.0737/0.075) = 0.13 in. wg
Figure L.1 in Annex L indicates estimated belt drive loss of 4.8%.
STATIC PRESSURE OF UNIT
HL = 0.048 Hmo = 0.048 × 24.1 = 1.2 hp H = Hmo - HL = 24.1 - 1.2 = 22.9 hp
Pv5 = (Q5/1096 A5)2 ρ5 = (39143/1096 × 14.3)2 0.0737 = 0.46 in. wg Pv2 = (Q2/1096 A2)2 ρ2 = (39143/1096 × 11.5)2 0.0737 = 0.71 in. wg
81
AMCA 203-90 (R2007) Ps2 + Pv2 = Ps5 + Pv5 Ps2 = Ps5 + Pv5 - Pv2 = 2.11 + 0.46 - 0.71 = 1.86 in. wg
CONVERSION TO SPECIFIED CONDITIONS
Pv1 = (Q1/1096 A1)2 ρ1 = (39249/1096 × 31.7)2 0.0735 = 0.09 in. wg
Psc = 2.22 (1050/1025)2 (0.075/0.0735) = 2.38 in. wg
Ps = Ps2 - Ps1 - Pv1 + SEF 1 = 1.86 - (-0.32) - 0.09 + 0.13 = 2.22 in. wg
82
Qc = 39249 (1050/1025) = 40206 cfm
Hc = 22.9 (1050/1025)3 (0.075/0.0735) = 25.1 hp
AMCA 203-90 (R2007)
EXAMPLE 4D: PACKAGED AIR-CONDITIONING UNIT
3a 3b
PLAN VIEW
STATIC PRESSURE TAPS 2
5
L
FILTER SECTION 1
SEF 1 +
+
INLET LOUVER
HEATING COIL SIDE VIEW
COMMENTS 1. The subject of the test in this example is the air conditioning unit assembly. This assembly includes the filter section and the inlet louver. The performance ratings for the unit assembly are based on operation with the outlets of the fans ducted. Before proceeding with the test, it is essential that all system dampers be fixed in the positions agreed upon by all interested parties as being applicable for the installation. Also, the temperature of the heating coil must be kept constant throughout the test period. It may be necessary to lock out, disconnect or otherwise modify automatic control devices in order to prevent the positions of the dampers and the temperature of the coil from changing during the test. Refer to Section 17.5.1 for additional considerations affecting the test procedure in this type of installation. 2. Normally, velocity pressure measurements would be made in a single plane, located in a duct common to all branches. In this example, a measurement plane which provides a satisfactory velocity profile cannot be located upstream of the fans or between the point of connection of the branch ducts and the outlets of the fans. The alternative, as indicated in the diagram, is to make a velocity pressure measurement traverse in each of two branches. the velocity pressure for reach branch is determined by using the root mean square of the velocity pressure measurements made in the traverse. The static
pressure at each traverse plane is determined by using the root mean square of the velocity measurement traverse in each of two branches. The velocity pressure for each branch is determined by using the root mean square of the velocity pressure measurements made in the traverse. The static pressure at each traverse plane is determined by averaging the static pressure measurements made in the same traverse. These static pressure values are used in determining the densities at the traverse planes. Procedures for traverses are described in Section 9.4. In order to determine the air flow rates, it is necessary to measure the area of each traverse plane. 3. Determine Ps5 by averaging the pressure measurements at each of four static pressure taps located in the duct fitting at the outlets of the fans. The conditions which exist at Plane 5, including the static pressure, are assumed to exist at Plane 2, based on their close proximity and the fact that there is no change in area between the two planes. In situations such as this example, it is important to be certain that all pressure measurements are referred to the same atmospheric pressure. 4. Measure the dry-bulb and wet-bulb temperatures at Plane 1 and the dry-bulb temperatures at Planes 3a, 3b, and 5. Determine pb for the general vicinity of 83
AMCA 203-90 (R2007) the air conditioning unit. These measurements are used to determine densities at the planes of interest. 5. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV), and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Motor performance data, supplied by the motor manufacturer, are used in the determination of motor power output in this example. 6. SEF 1 is due to the effect of insufficient length of duct between the outlets of the fans and the elbow downstream of the fans. In order to determine the value of SEF 1, it is necessary to measure the outlet area and the blast area of one of the fans and the length of the duct, L, between the fan and the elbow. 7. The sum of the static pressure, Ps1, and velocity pressure, Pv1, at the inlet to the unit assembly is considered to be equal to the sum of the static pressure, Psx, and velocity pressure, Pvx, at a point sufficiently distant from the inlet as to be in still air. At this point, the static pressure is zero, and the velocity pressure in still air is zero. Ps1 + Pv1 = Psx + Pvx = 0 This consideration, which is the same as that used in the methods for testing this type of unit for performance rating purposes, charges to the unit losses incurred in accelerating the air into its inlet and eliminates the inaccuracies which arise in any attempt to measure the velocity pressure and static pressure at the inlet. To calculate the static pressure for the unit assembly: Ps = Ps2 - Ps1 - Pv1 + SEF 1 = Ps2 - (Ps1 + Pv1) + SEF 1 Since: Ps1 + Pv1 = 0 Ps = Ps2 + SEF 1 Where: Ps2 = Ps5
8. In order to compare the test results to the quoted performance curve for the packaged unit drawn for operation at 1720 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS pb td1 tw1 td5 td3a td3b Ps5 Ps3a Ps3b Pv3a Pv3b N A2
= 29.65 in. Hg = 72°F = 61°F = 85°F = 82.5°F = 83°F = 1.25 in. wg = 1.15 in. wg = 1.22 in. wg = 0.56 in. wg = 0.60 in. wg = 1710 rpm = A5 = 5.64 ft2 A3a = 3.1 ft2 A3b = 2.2 ft2 Blast Area = 2.5 ft2 per fan L = 0.96 ft, length of outlet duct MEASURED MOTOR DATA Volts = = Amps = =
460, 458, 462 460 av 10.0, 10.0, 9.8 9.9 av
MOTOR NAMEPLATE DATA 10 hp, 3 phase, 60 hertz 460 volts, 1750 rpm, 13.5 FLA GENERAL Fans connected to motor through belt drive. The following motor performance data was supplied by the motor manufacturer: Motor Efficiency: 82.5% at 1/2 load 84.5% at 3/4 load 84.5% at full load Power Factor = 0.85
84
AMCA 203-90 (R2007) DENSITIES For Plane 1 conditions of: td1 = 72°F tw1 = 61°F p1 = pb = 29.65 in. Hg Use Figure N.1 in Annex N to obtain ρ1 = 0.0735 lbm/ft3. ⎛ P + 13.6 pb ⎞ ⎛ t d1 + 460 ⎞ ρ5 = ρ1 ⎜ s5 ⎟ ⎟⎜ ⎝ 13.6 p1 ⎠ ⎝ t d5 + 460 ⎠ ⎛ 1.25 + 13.6 × 29.65 ⎞ ⎛ 532 ⎞ = 0.0735 ⎜ ⎟ ⎜ 545 ⎟ 13.6 × 29.65 ⎠⎝ ⎠ ⎝ = 0.0720 lbm/ft 3 It is assumed that ρ2 = ρ5 ⎛ P + 13.6 pb ⎞ ⎛ t d1 + 460 ⎞ ρ3a = ρ1 ⎜ s3a ⎟ ⎟⎜ ⎝ 13.6 p1 ⎠ ⎝ t d3a + 460 ⎠ ⎛ 1.15 + 13.6 × 29.65 ⎞ ⎛ 532 ⎞ = 0.0735 ⎜ ⎟ ⎜ 542.5 ⎟ 13.6 × 29.65 ⎝ ⎠⎝ ⎠ = 0.0723 lbm/ft 3 ⎛ P + 13.6 pb ⎞ ⎛ t d1 + 460 ⎞ ρ3b = ρ1 ⎜ s3b ⎟ ⎟⎜ ⎝ 13.6 p1 ⎠ ⎝ t d3b + 460 ⎠ ⎛ 1.22 + 13.6 × 29.65 ⎞ ⎛ 532 ⎞ = 0.0735 ⎜ ⎟ ⎜ 543 ⎟ 13.6 × 29.65 ⎝ ⎠⎝ ⎠ = 0.0722 lbm/ft 3 FLOW RATES V3a = 1096 (Pv3a/ρ3a)0.5 = 1096 (0.56/0.0723)0.5 = 3050 fpm V3b = 1096 (Pv3b/ρ3b)0.5 = 1096 (0.60/0.0722)0.5 = 3159 fpm Q3a = V3aA3a = 3050 × 3.1 = 9455 cfm Q3b = V3bA3b = 3159 × 2.2 = 6950 cfm
Q = = = =
Q1 Q3a (ρ3a/ρ1) + Q3b (ρ3b/ρ1) 9455 (0.0723/0.0735) + 6950 (0.0722/0.0735) 16128 cfm
Q2 = = = =
Q5 Q1 (ρ1/ρ5) 16128 (0.0735/0.0720) 16464 cfm
FAN POWER INPUT Measured amps/FLA = (9.9/13.5) = 0.73 = 73% The data supplied by the motor manufacturer indicate power factor of 0.85 and motor efficiency of 84.5% for the motor operating at 73% FLA. Using the appropriate equation in Section 10.2.2: Hmo = (3)0.5 × 9.9 × 460 × 0.85 × 0.845/746 = 7.59 hp Figure L.1 in Annex L indicates estimated belt drive loss of 5.6%. HL = 0.056 Hmo = 0.056 × 7.59 = 0.43 hp H
= Hmo - HL = 7.59 - 0.43 = 7.16 hp
SYSTEM EFFECT FACTOR SEF 1 is due to the effect of insufficient lengths of duct between the outlets of the fans and the elbow downstream of the fans. AMCA Publication 201-90, Figures 7.1, 8.1, and 8.5 indicate the following calculations: V2 = (Q2/A2) = (16464/5.64) = 2919 fpm Duct diameter equivalent to the outlet area of one fan: De2 = (4A2/2π)0.5 = (4 × 5.64/2π)0.5 = 1.89 ft Figure 8.1 shows that for velocities over 2500 fpm 100% effective duct length is one diameter for every 1000 fpm: 85
AMCA 203-90 (R2007) STATIC PRESSURE OF UNIT = De2 (V2/1000) = 1.89 (2919/1000) = 17% L, in % effective duct length: = (L/5.52) 100 = (0.96/5.52) 100 = 17% Blast area ratio = Blast Area/A2 = (2 × 2.5)/5.64 = 0.89 For a blast area ratio of 0.89, 17% effective duct length and elbow position C, Figure 8.5 shows System Effect Curve S applies. For 2919 fpm velocity and curve S, Figure 7.1 shows SEF 1 = 0.43 in. wg at 0.075 lbm/ft3 density. At 0.0720 lbm/ft3: SEF 1 = 0.43 (0.0720/0.075) = 0.41 in. wg
86
Ps2 = Ps5 = 1.25 in. wg Ps = Ps2 + SEF 1 = 1.25 + 0.41 = 1.66 in. wg CONVERSION TO SPECIFIED CONDITIONS Qc = 16128 (1720/1710) = 16222 cfm Psc = 1.66 (1720/1710)2 (0.075/0.0735) = 1.71 in. wg Hc = 7.16 (1720/1710)3 (0.075/0.0735) = 7.44 hp
AMCA 203-90 (R2007)
EXAMPLE 4E: CENTRAL STATION AIR CONDITIONING UNIT, FACTORY ASSEMBLED BLOWTHROUGH TYPE 2
1
STATIC PRESSURE TAPS
5
PLAN VIEW
RETURN AIR
3b
SPRAY SECTION
3a
HEATING COIL + +
+
OUTSIDE AIR
+ + + +
FILTER SECTION
+
FAN SECTION
COOLING COIL
SIDE VIEW
COMMENTS 1. This is a factory assembled, blow-through central station unit. The subject of the test is the fan section, which is rated by the manufacturer as an assembly of the fan and the cabinet in which the fan has been installed. As a blow-through unit, the performance ratings for the fan section are based on operation without the fan outlet ducted. Before proceeding with the test, it is essential that all dampers (outside air, return air, mixing box, multizone, face and bypass, or volume control) be fixed in the positions agreed upon by all interested parties as being applicable for the installation. Also, the temperatures of heating and cooling coils must be kept constant throughout the test period. It may be necessary to lock out, disconnect, or otherwise modify automatic control devices in order to prevent the positions of the dampers and temperatures of the coils from changing during the test. In instances in which a cooling coil is located between a velocity pressure traverse plane and the fan, as in this example, the flow of the cooling medium should be stopped or its temperature raised to a level sufficient to prevent condensation on the cooling coil, otherwise the moisture condensed will not be properly taken into account in the determination of fan air flow rate. Refer to Section 17.5.2 for additional considerations affecting the test procedure in this type of installation. 2. Normally, velocity pressure measurements would
be made in a single plane, located in a duct common to all branches. In this example, a measurement plane which provides a satisfactory velocity profile cannot be located upstream of the fan or between the point of connection of the branch ducts and the fan outlet. The alternative, as indicated in the diagram, is to make a velocity pressure measurement traverse in each branch. The velocity pressure for each branch is determined by using the root mean square of the velocity pressure measurements made in the traverse. The static pressure at each traverse plane is determined by averaging the static pressure measurements made in the same traverse. These static pressure values are used in determining the densities at the traverse plane. Procedures for traverses are described in Section 9.4. In order to determine the air flow rates it is necessary to measure the area of each traverse plane. 3. Determine Ps1 by averaging the static pressure measurements made in a traverse of Plane 1. Ps5 may be determined in a similar manner or by averaging the pressure measurements at each of four static pressure taps. If it is possible to install suitable pressure taps, their use is preferred in the regions of the fan outlet. Due to the abrupt expansion in area from Plane 2 to Plane 5, it is assumed that there is no conversion of velocity pressure at Plane 2 to static pressure at Plane 5. Therefore, it is assumed 87
AMCA 203-90 (R2007) that Ps2 = Ps5. Measure the area of Plane 1 for use in calculating Pv1. 4. Measure the dry-bulb and wet-bulb temperatures at Planes 1, 3a, and 3b. Determine pb for the general vicinity of the air conditioning unit. These measurements are used to determine densities at the planes of interest. The measurements of additional wet-bulb temperatures were made in this example in order to provide data which may be used to determine whether the moisture content of the air changed between Plane 1 and Planes 3a and 3b. 5. Measure the fan speed and the motor amps, volts, and if possible, watts. Record all pertinent motor nameplate data, including volts (NPV), and full load amps (FLA). If the motor power output is to be estimated by using the phase current method described in Annex K, it is not necessary to measure motor watts; however, it may be necessary to disconnect the drive and measure the no load amps (NLA) if the motor is not operating at or near its full load point. Refer to Annex K.
tw3a = td3b = tw3b = Ps1 = Ps5 = Ps3a = Ps3b = Pv3a = Pv3b = N = A1 = A3a = A3b =
71.5°F 60°F 58°F -2.43 in. wg 6.55 in. wg 5.35 in. wg 5.1 in. wg 0.53 in. wg 0.60 in. wg 1695 rpm 68.9 ft2 5.37 ft2 6.78 ft2
MEASURED MOTOR DATA Volts = = Amps = = NLA =
570, 575, 565 570 av 81.5, 82.5, 81 81.7 19
MOTOR NAMEPLATE DATA 6. Since the performance ratings for the fan section are based on operation without the fan outlet ducted, an SEF does not apply for the unducted position.
100 hp, 3 phase, 60 hertz 575 volts, 1790 rpm, 95 FLA
7. To calculate the Fan Section Static Pressure:
GENERAL
Ps = Ps2 - Ps1 - Pv1
Fan connected to motor through belt drive.
Where:
CALCULATIONS
Ps2 = Ps5 Pv1 = (Q1/1096 A1)2 ρ1 Q1 = Q3a (ρ3a/ρ1) + Q3b (ρ3b/ρ1)
DENSITIES
The calculation of Pv1 is often ignored in instances similar to this example on the basis that the calculated value of Pv1 is relatively small, and its omission does not affect the test results significantly.
td1 = 65°F tw1 = 60°F
8. In order to compare the test results to the quoted fan section curve drawn for operation at 1650 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14.
For Plane 1 conditions of:
p1 = pb + (Ps1/13.6) = 28.85 + (-2.43/13.6) = 28.67 in. Hg Use Figure N.1 in Annex N to obtain ρ1 = 0.0720 lbm/ft3. For Plane 3a conditions of:
OBSERVATIONS SITE MEASUREMENTS pb td1 tw1 td3a 88
= = = =
28.85 in. Hg 65°F 60°F 100°F
td3a = 100°F tw3a = 71.5°F p3a = pb + (Ps3a/13.6) = 28.85 + (5.35/13.6) = 29.24 in. Hg
AMCA 203-90 (R2007) Use Figure N.1 in Annex N to obtain ρ1 = 0.0720 lbm/ft3.
Hmo
For Plane 3b conditions of:
Reference to Figure L.1 in Annex L indicates estimated belt drive loss of 4.2%.
td3b = 60°F tw3b = 58°F p3b = pb + (Ps3b/13.6) = 28.85 + (5.1/13.6) = 29.23 in. Hg Use Figure N.1 in Annex N to obtain ρ3b = 0.0741 lbm/ft3. FLOW RATES V3a = 1096 (Pv3a/ρ3a)0.5 = 1096 (0.53/0.0691)0.5 = 3035 fpm V3b = 1096 (Pv3b/ρ3b = 1096 (0.60/0.0741)0.5 = 3119 fpm )0.5
Q3a = V3aA3a = 3035 × 5.37 = 16298 fpm Q3b = V3bA3b = 3119 × 6.78 = 21147 cfm Q = Q1 = Q3a (ρ3a/ρ1) + Q3b (ρ3b/ρ1) = 16298 (0.0691/0.0720) + 21147 (0.0741/0.0720) = 37405 cfm
= (85.3 + 81.8)/2 = 83.6 hp
HL = 0.042 Hmo = 0.042 × 83.6 = 3.5 hp H = Hmo - HL = 83.6 - 3.5 = 80.1 hp FAN SECTION STATIC PRESSURE Pv1 = (Q1/1096 A1)2 ρ1 = (37405/1096 × 68.9)2 0.0720 = 0.02 in. wg It is assumed that Ps2 = Ps5 Ps = Ps2 - Ps1 - Pv1 = 6.55 - (-2.43) - 0.02 = 8.96 in. wg CONVERSION TO SPECIFIED CONDITIONS Qc = 37405 (1650/1695) = 36412 cfm Psc = 8.96 (1650/1695)2 (0.075/0.0720) = 8.84 in. wg Hc = 80.1 (1650/1695)3 90.075/0.0720) = 77.0 hp
FAN POWER INPUT Measured amps/FLA = (81.7/95) = 0.86 = 86% Annex K indicates that the average of the results of Equation A and Equation B will provide a reasonably accurate estimate of motor power output for a 100 hp motor operating at 86% of FLA. Eqn. A = 100 (81.7/95) (570/575) = 85.3 hp Eqn. B = 100 [(81.7 - 19)/(95 - 19)] (570/575) = 81.8 hp
89
AMCA 203-90 (R2007)
EXAMPLE 5A: FREE INLET, FREE OUTLET ROOF VENTILATOR
2
1
3 2 De
TEMPORARY DUCT WITH SQUARE CROSS-SECTION, De = EQUIVALENT DIAMETER OF DUCT
1.5 De
COMMENTS 1. The subject of the test in this example is the roof ventilator assembly. Before proceeding with the test, refer to Section 17.4 for considerations affecting the test procedure in this type of installation. 2. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located in the duct which has been installed on the inlet side of the ventilator. Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. Measure the area of the traverse plane, A3, which is located at the tip of the Pitot-static tube. The duct, temporarily installed for purposes of the test, is square in cross-section. Its cross-sectional dimensions were selected as the maximum permissible for its installation into the opening in the ventilator mounting curb. The length of the duct is twice its equivalent diameter and the entrance to the duct is flared in oder to reduce inlet losses. The installation of a duct of this size and cross-sectional configuration is judged as creating no significant effect on the performance of the ventilator in this example. 3. Ps2, the static pressure at the outlet of the ventilator, is zero gauge pressure, referred to the atmospheric pressure in the region of the ventilator outlet. In situations such as this example, the air may 90
be discharging from the ventilator into a region in which the atmospheric pressure is somewhat different from that to which all other pressure measurements are referred. When this possibility exists, it is essential that the static pressure in the region of the discharging air be measured, referred to the same atmospheric pressure as used in all other pressure measurements. In this example, Ps2 was measured, referred to the same atmospheric pressure as in the static pressure measurements made at Plane 3. 4. Measure the dry-bulb and wet-bulb temperatures at the velocity traverse plane. Determine pb for the general vicinity of the ventilator. These measurements are used to determine densities at the planes of interest. 5. Measure the fan speed and the motor amps and volts. Record all pertinent motor nameplate data. For the horsepower rating of the motor in this example, it is recommended that the fan power input be determined by using the measured watts input to the motor and motor performance data, obtained from the motor manufacturer. 6. To calculate the Fan Static Pressure: Ps = Ps2 - Ps1 - Pv1 = Ps2 - (Ps1 + Pv1)
AMCA 203-90 (R2007) Where:
FLOW RATE
Ps1 + Pv1 = Ps3 + Pv3
V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.077/0.0727)0.5 = 1128 fpm
7. In order to compare the test results to the quoted fan curve drawn for operation at 1180 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS pb td3 tw3 Ps2 Ps3 Pv3 N A3
= = = = = = = =
29.37 in. Hg 73.5°F 58.1°F 0.037 in. wg -0.085 in. wg 0.077 in. wg 1177 rpm 5.58 ft2
Q = = = =
Q1 = Q3 V3A3 1128 × 5.58 6294 cfm
FAN POWER INPUT At the measured power input value of 755 watts, the data supplied by the motor manufacturer indicate efficiency of 61% for the motor. Hmo = (755 × 0.61)/746 = 0.62 hp Since the fan is direct connected to the motor, there is no drive loss, and:
MEASURED MOTOR DATA
H = Hmo = 0.62 hp
Volts = 235, 230, 230 = 232 av Watts = 755
FAN STATIC PRESSURE
MOTOR NAMEPLATE DATA 1 hp, 3 phase, 60 hertz 230 volts, 1175 rpm, 3.6 FLA
Ps1 + Pv1 = Ps3 + Pv3 = -0.085 + 0.077 = -0.008 in. wg Ps = Ps2 - (Ps1 + Pv1) = 0.037 - (-0.008) = 0.045 in. wg
General CONVERSION TO SPECIFIED CONDITIONS Fan direct connected to motor. Motor efficiency data supplied by motor manufacturer.
Qc = 6294 (1180/1177) = 6310 cfm
CALCULATIONS DENSITIES For Plane 3 conditions of: td3 = 73.5°F tw3 = 58.1°F
Psc = 0.045 (1180/1177)2 (0.075/0.0727) = 0.047 in. wg Hc = 0.62 (1180/1177)3 (0.075/0.0727) = 0.64 hp
p3 = pb + (Ps3/13.6) = 29.37 + (-0.085/13.6) = 29.36 in. Hg Use Figure N.1 in Annex N to obtain ρ3 = 0.0727 lbm/ft3. It is assumed that ρ1 = ρ3. 91
AMCA 203-90 (R2007)
EXAMPLE 5B: FREE INLET, FREE OUTLET PROPELLER FAN
2 De
2 3
1.5 De
D2
TEMPORARY DUCT WITH SQUARE CROSS-SECTION, De = EQUIVALENT DIAMETER OF DUCT
COMMENTS 1. The subject of the test in this example is the propeller fan assembly. Before proceeding with the test, refer to Section 17.4 for considerations affecting the test procedure in this type of installation. 2. Determine Pv3 by using the root mean square of the velocity pressure measurements made in a traverse of Plane 3, located in the duct which has been installed on the inlet side of the fan. Determine Ps3 by averaging the static pressure measurements made in the same traverse. Procedures for traverses are described in Section 9.4. Measure the area of the traverse plane, A3, which is located at the tip of the Pitot-static tube. The duct, temporarily installed for purposes of the test, is square in cross-section, with side dimension of 1.5 D2. The shape and area of the duct cross-section were selected on the basis of minimizing the effect of the duct on the performance of the fan while providing velocity pressure readings of measurable magnitudes. The length of the duct is twice its equivalent diameter, and the entrance to the duct is flared in order to reduce inlet losses. The installation of the duct is judged as creating no significant effect on the performance of the fan in this example.
such as this example, the air may be discharging from the fan into a region in which the atmospheric pressure is somewhat different from that to which all other pressure measurements are referred. When this possibility exists, it is essential that the static pressure in the region of the discharging air be measured, referred to the same atmospheric pressure as used in all other pressure measurements. In this example, Ps2 was measured, referred to the same atmospheric pressure as in the static pressure measurements made at Plane 3. 4. Measure the dry-bulb and wet-bulb temperatures at the velocity traverse plane. Determine pb for the general vicinity of the fan. These measurements are used to determine densities at the planes of interest. 5. Measure the fan speed and the motor amps and volts. Record all pertinent motor nameplate data. For the horsepower rating of the motor in this example, it is recommended that the fan power input be determined by using the measured watts input to the motor and motor performance data obtained from the motor manufacturer. 6. To calculate the Fan Static Pressure:
3. Ps2, the static pressure at the outlet of the fan, is zero gauge pressure, referred to the atmospheric pressure in the region of the fan outlet. In situations 92
Ps = Ps2 - Ps1 - Pv1 = Ps2 - (Ps1 + Pv1)
AMCA 203-90 (R2007) Where:
FLOW RATES
Ps1 + Pv1 = Ps3 + Pv3
V3 = 1096 (Pv3/ρ3)0.5 = 1096 (0.025/0.0715)0.5 = 648 fpm
7. In order to compare the test results to the quoted fan curve drawn for operation at 1725 rpm and 0.075 lbm/ft3 density, it is necessary to convert the results to the specified conditions. The basis for the calculations is described in Section 14. OBSERVATIONS SITE MEASUREMENTS pb = td3 = tw3 = Ps2 = Ps3 = Pv3 = N = A3 =
29.65 in. Hg 85°F 74°F 0 in. wg -0.027 in. wg 0.025 in. wg 1775 rpm 5.06 ft2
Q = = = =
Q1 = Q3 V 3 A3 648 × 5.06 3279 cfm
FAN POWER INPUT At the measured power input value of 637 watts, the data supplied by the motor manufacturer indicate efficiency of 65% for the motor. Hmo = (637 × 0.65)/746 = 0.56 hp Since the fan is direct connected to the motor, there is no drive loss, and:
MEASURED MOTOR DATA
H = Hmo = 0.56 hp
Volts = 230, 225, 230 = 228 av Watts = 637
FAN STATIC PRESSURE
MOTOR NAMEPLATE DATA 3/4 hp, 3 phase, 60 hertz 230 volts, 1760 rpm, 4.8 FLA GENERAL Fan direct connected to motor. Motor efficiency data supplied by motor manufacturer.
Ps1 + Pv1 = Ps3 + Pv3 = -0.027 + 0.025 = -0.002 in. wg Ps = Ps2 - (Ps1 + Pv1) = 0 - (-0.002) = 0.002 in. wg This small value is attributed to the loss at the duct inlet, and the fan is considered to be operating at free delivery (Ps = 0).
CALCULATIONS CONVERSION TO SPECIFIED CONDITIONS DENSITIES For Plane 3 conditions of: td3 = 85°F tw3 = 74°F p3 = pb + (Ps3/13.6) = 29.65 + (-0.027/13.6) = 29.65 in. Hg
Qc = 3279 (1725/1775) = 3187 cfm Psc = 0 in. wg Hc = 0.56 (1725/1775)3 (0.075/0.0715) = 0.54 hp
Use Figure N.1 in Annex N to obtain ρ3 = 0.0715 lbm/ft3. It is assumed that ρ1 = ρ3 93
AMCA 203-90 (R2007)
EXAMPLE 5C: FREE INLET, FREE OUTLET ROOF VENTILATOR
2
1 3
COMMENTS 1. The subject of the test in this example is the roof ventilator assembly. Before proceeding with the test, refer to Section 17.1 for considerations affecting the test procedure in this type of installation. 2. Ps3, the static pressure in the vicinity of the ventilator inlet, would normally be determined by averaging the static pressure measurements made in a Pitot tube traverse. But in this example, a temporary duct was not installed and the Pitot tube traverse could not be accomplished. In this method for testing a nonducted fan, consider the fan static pressure (Ps) as the differential pressure, as read on a manometer, between the pressure measured inside the room (Ps3) and the pressure measured outside the room in the vicinity of the ventilator outlet (Ps2). These pressures are measured at a sufficient distance from the ventilator so as to be unaffected by the velocity of the entering or leaving air. 3. Ps2 is considered to be zero gauge pressure, but since this measurement is actually part of the differential pressure described in paragraph 2, it is necessary to make only one density correction; the correction is to the differential pressure, which is the fan static pressure.
94
4. Measure the dry-bulb and wet-bulb temperatures in the region of the inside pressure measurement. Also, determine pb in the same vicinity. 5. Measure the fan speed and the motor amps and volts. Record all pertinent motor nameplate data. For the horsepower rating of the motor in this example, it is recommended that the fan power input be determined by using the measured watts input to the motor and motor performance data obtained from the motor manufacturer. 6. Airflow rates are determined from the fan manufacturer’s certified performance ratings. Draw a fan performance curve from these ratings converted to operation at the test values of fan speed and entering air density. The basis for these calculations is described in Section 14. The fan airflow rate is then determined by entering this curve at the test values of fan static pressure and fan power input. OBSERVATIONS SITE MEASUREMENTS pb = 29.19 in. Hg td3 = 79°F tw3 = 63°F Ps2 - Ps3 = 0.13 in. wg N = 1735 rpm
AMCA 203-90 (R2007) MEASURED MOTOR DATA
FAN STATIC PRESSURE
Volts = 229, 229, 232 = 230 av Watts = 1390
The fan static pressure is considered to be the differential static pressure.
MOTOR NAMEPLATE DATA
Ps = Ps2 - Ps3 = 0.13 in. wg
1.5 hp, 3 phase, 60 hertz 230 volts, 1740 rpm, 4.8 FLA
It is assumed that Ps1 = Ps3
GENERAL
CONVERSION OF MANUFACTURER’S RATINGS TO OPERATING CONDITIONS
Fan direct connected to motor. Motor efficiency data supplied by motor manufacturer.
Rating Point #1
Fan performance, at standard air density, as supplied by fan manufacturer for 1750 rpm. Point
CFM
Ps
HP
1) 2) 3)
8900 8520 8060
0 1/8 1/4
1.45 1.50 1.55
CALCULATIONS DENSITIES For Plane 3 conditions of: td3 = 79°F tw3 = 63°F pb3 = pb + (Ps2 - Ps1)/13.6 = 29.19 + (0.13/13.6) = 29.2 in. Hg Use Figure N.1 in Annex N to obtain ρ3 = 0.0715 lbm/ft3. It is assumed that ρ1 = ρ3. FAN POWER INPUT At the measured power input value of 1395 watts, the data supplied by the motor manufacturer indicate efficiency of 77% for the motor. Hmo = (1390 × 0.77)/746 = 1.43 hp Since the fan is direct connected to the motor, there is no drive loss, and: H
= Hmo = 1.43 hp
FAN
Q1c = 8900 (1735/1750) = 8824 cfm Ps1c = 0 H1c = 1.45 (1735/1750)3 (0.0715/0.0750) = 1.35 hp Rating Point #2 Q2c = 8520 (1735/1750) = 8447 cfm Ps2c = 0.125 (1735/1750)2 (0.0715/0.0750) = 0.117 in. wg H2c = 1.50 (1735/1750)3 (0.0715/0.0750) = 1.39 hp Rating Point #3 Q3c = 8060 (1735/1750) = 7991 cfm Ps3c = 0.25 (1735/1750)2 (0.0715/0.0750) = 0.234 in. wg H3c = 1.55 (1735/1750)3 (0.0715/0.0750) = 1.44 hp Draw a performance curve for these operating conditions. Enter the measured values for static pressure and horsepower on the appropriate curves. Ideally, these two points will coincide at the same cfm. However, usually they will not coincide and should be averaged to determine the fan airflow rate. If this difference is small, such as in this example, it is only a reflection of test inaccuracies. If, however, these differences exceed 10%, the system should be reanalyzed for SEFs that may have been overlooked, or for procedural errors in the initial testing. 95
AMCA 203-90 (R2007) Qa = 8070 cfm (based upon horsepower) Qb = 8400 cfm (based upon static pressure) Use: Q = 8235 cfm (average of above).
x
x BHP
x 1.25
.30
STATIC PRESSURE IN. WG (Ps)
x 1.00
.20
x
.10
0 7000
SP
x 8000
9000 CFM(Q)
Fan Performance at 0.0715 Air Density
96
BHP (H)
1.50
.40
AMCA 203-90 (R2007)
Annex B. Pitot Static Tubes 16D
8D 0.8D
0.5D Radius
0.4D D
3D Radius
Head shall be free from nicks and burrs. 90° ± 0.1°
All dimensions shall be within ±2%. SECTION A-A
Static Pressure
8 holes - 0.13D, not to exceed 0.04 in., diameter equally spaced and free from burrs. Hole depth shall not be less than the hole diameter.
Note: Surface finish shall be 32 micro in. or better. The static orifices may not exceed 0.04 in. diameter. The minimum Pitot tube stem diameter recognized under this standard shall be 0.10 in. In no case shall the stem diameter exceed 1/30 of the test duct diameter.
Total Pressure
PITOT-STATIC TUBE WITH SPHERICAL HEAD All other dimensions are the same as for spherical head pitot-static tubes. 8D
D
X
0.2D Diameter V
X/D
V/D
X/D
V/D
0.000 0.237 0.336 0.474 0.622
0.500 0.496 0.494 0.487 0.477
1.602 1.657 1.698 1.730 1.762
0.314 0.295 0.279 0.266 0.250
0.741 0.936 1.025 1.134 1.228
0.468 0.449 0.436 0.420 0.404
1.796 1.830 1.858 1.875 1.888
0.231 0.211 0.192 0.176 0.163
1.313 1.390 1.442 1.506 1.538 1.570
0.388 0.371 0.357 0.343 0.333 0.323
1.900 1.910 1.918 1.920 1.921
0.147 0.131 0.118 0.109 0.100
ALTERNATE PITOT-STATIC TUBE WITH ELLIPSOIDAL HEAD Figure B.1 97
AMCA 203-90 (R2007)
Annex C. Double Reverse Tubes AIR FLOW TUBE ENDS MUST BE SMOOTH AND FREE FROM BURRS
IMPACT TUBE
REVERSE TUBE
SECTION VIEW
STAINLESS STEEL TUBING PREFERRED APPROX. 0.375 in. OD
FLEXIBLE TUBING
TOTAL PRESSURE = READING A CORRECTED FOR MANOMETER CALIBRATION
ING B
READING A
READ
VELOCITY PRESSURE = READING B CORRECTED FOR MANOMETER CALIBRATION AND CALIBRATION FACTOR FOR THE DOUBLE REVERSE TUBE.
Notes: 1. For use in dirty or wet gas streams.
2. The double reverse tube must be calibrated and used in the same orientation as used in its calibration 3. Also referred to as impact reverse tube, combined reverse tube, and type S tube. Figure C.1 - Double Reverse Tube 98
AMCA 203-90 (R2007)
Annex D. Pitot-Static Tube Holder
0.312 in. DIA.
PITOT-STATIC TUBE SPLIT BRASS BUSHING PRESS TO FIT INTO TUBING
THERMOCOUPLE
DUCT WALL
1½ in. PIPE HALF-COUPLING WELDED TO DUCT BRASS BUSHINGS
1½ in. PIPE NIPPLE 12 in. LONG
STAINLESS STEEL TUBING 1 in. OUTSIDE DIA. × 8 ft. LONG SLIP FIT IN BRASS BUSHINGS
Notes: ¼ in. OUTSIDE DIA. STAINLESS STEEL TUBING FOR GAS SAMPLING
SPLIT BRASS BUSHING
1. Apparatus for mounting Pitot-static tube on duct 2. For use in large ducts or high velocity gas streams 3. 1 in. diameter tube slides inside 1.5 in. pipe, which can be unscrewed and moved to another traverse location 4. The gas sampling tube and thermocouple may be omitted if these data are obtained in other manners
CUT-OFF AND REBRAZE AFTER ASSEMBLY
Figure D.1 - Pitot-Static Tube Holder (Typical) 99
AMCA 203-90 (R2007)
Annex E. Static Pressure Tap
DUCT WALL
MAXIMUM 0.125 in. DIAMETER FOR USE IN RELATIVELY CLEAN GASES. MAY BE NECESSARY TO INCREASE TO 0.312 in. DIAMETER FOR DIRTY OR WET GASES ½ in. PIPE HALF-COUPLING OR SIMILAR ARRANGEMENT
INSIDE SURFACE OF DUCT AND EDGE OF HOLE ARE TO BE SMOOTH AND FREE FROM BURRS
Figure E.1 - Static Pressure Tap
MINIMUM OF FOUR TAPS, LOCATED 90° APART AND NEAR THE CENTER OF EACH WALL
STATIC PRESSURE MEASUREMENT REQUIRED AT EACH TAP. USE THE AVERAGE OF THE MEASUREMENTS AS THE STATIC PRESSURE FOR THE PLANE
Figure E.2 - Locations of Static Pressure Taps 100
AMCA 203-90 (R2007)
Annex F. Pitot-Static Tube Connections
PLANE 2
PLANE 1
PLANE 4
PLANE 3
*SEF 1 Ps4 FAN STATIC PRESSURE Ps = - Ps1 - Pv1 + SEF 1 where Ps1 = Ps4 Pv1 = Pv3 Figure F.1 - Fan with Inlet Duct Only Ps2 = 0 PLANE 3
PLANE 5
Ps3
Ps3
P v3 *SEF 1 is due to no duct at fan outlet PLANE 2
PLANE 1
FAN STATIC PRESSURE Ps = Ps2 where Ps2 = Ps5 Pt1 = 0
Ps5
P v3
Figure F.2 - Fan with Outlet Duct Only ALTERNATE PLANE 5 PLANE 3
Ps5
PLANE 2
PLANE 1
FAN STATIC PRESSURE Ps = Ps2 - Ps1 - Pv1 where Ps2 = Ps5 Ps1 = Ps4 Pv1 = Pv3
PLANE 4
PLANE 3
Ps3
Ps4
P v3
Figure F.3 - Fan with Inlet Duct and Outlet Duct 101
AMCA 203-90 (R2007)
Annex G. Manometer Data
10 in. wg 1:1 SLOPE RATIO
2 in. wg 5:1 SLOPE RATIO
0.5 in. wg 20:1 SLOPE RATIO
Figure G.1 - Manometer Data
102
1 in. wg 10:1 SLOPE RATIO
AMCA 203-90 (R2007) PERCENT UNCERTAINTY IN VELOCITY DETERMINATION USING PITOT-STATIC TUBE AND MANOMETER DUE TO MANOMETER SLOPE Based on an uncertainty equivalent to an indicating column length of 0.05 in. wg in a vertical manometer (1:1 slope ratio)
.01
.02
VELOCITY PRESSURE READING, in. wg .04 .06 0.1 0.2 0.4 0.6 1 2
3 4
6 8 10
10.0 8.0 6.0 5.0
3.0
2.0
R TE ME TIO NO RA MA OPE SL 1:1
1.0 0.8 0.6 0.5 2:1
:1 10
:1
0.3
5:1
0.4
20
% UNCERTAINTY IN VELOCITY DETERMINATION
4.0
0.2
0.3
0.4
0.6
0.8
1
2
3
4
6
8
10
15
STANDARD AIR VELOCITY, fpm (×1000)
Figure G.2 - Uncertainty in Velocity Determination
103
AMCA 203-90 (R2007)
Annex H. Distribution of Traverse Points In order to obtain a representative average velocity in a duct, it is necessary to locate each traverse point accurately. It is recommended that the number of traverse points increase with increasing duct size. The distributions of traverse points for circular ducts, as indicated below, are based on log-linear Pitot traverse method.
X1
60º
X2
X3 X4
D
Xn
Xa = D × Ka Where: D is the inside diameter of the duct Ka is the factor corresponding to the duct size and the traverse point location as indicated in the table below NUMBER OF TRAVERSE INSIDE DIAMETER POINTS IN K1 OF DUCT EACH OF 3 DIAMETERS
K2
K3
K4
K5
K6
K7
K8
K9
K10
K11
K12
K14
K15
K16
LESS THAN 8 ft.
8
.021 .117 .184 .345 .655 .816 .883 .979
8 ft. THROUGH 12 ft.
12
.014 .075 .114 .183 .241 .374 .626 .759 .817 .886 .925 .986
GREATER THAN 12 ft.
16
.010 .055 .082 .128 .166 .225 .276 .391 .609 .724 .775 .834 .872 .918 .945 .990
Figure H.1 - Distribution of Traverse Points for Circular Ducts 104
K13
AMCA 203-90 (R2007) The recommended minimum number of traverse points for rectangular ducts is indicated below in Figure H.3. For rectangular ducts with cross-sectional areas of 24 square feet and less, the recommended minimum number is 24. For cross-sectional areas greater than 24 square feet, the minimum number of points increases as indicated in Figure H.3. The points are to be located in the centers of equal areas with the areas as nearly square as practical (see Figure H.2). If the flow conditions at the traverse plane are less than satisfactory, the accuracy of the determination of flow rate may be improved by using more than the recommended minimum number of points. Fewer points may be used if the flow is very uniform; however, the maximum area covered per point should not exceed 3 square feet. Y
Y 2
X 2 X
Figure H.2 - Distribution of Traverse Points for Rectangular Duct
NUMBER OF TRAVERSE POINTS
100 90 80 70 60 50 40 30 25 20 15
10 10
15
20
25 30
40
50 60 70 80 100
150
200 250 300
DUCT CROSS-SECTIONAL AREA, ft2
Figure H.3 - Recommended Minimum Number of Traverse Points for Rectangular Ducts 105
AMCA 203-90 (R2007)
Annex J. Instrumentation Characteristics
Table J.1 - Temperature Measurement
No. Measurement Means 1. Glass-stem thermometers Mercury-glass thermometer
Application Temp of gases and liquids by contact
Alcohol-glass thermometer Pentane-glass thermometers Jena or quartz mercury nitrogen thermometers 2. Gas thermometer 3. Resistance thermometers Platinum-resistance thermometer
Nickel-resistance thermometer
Precision F
Limitations
-38/575
Less than 0.1 to 10
In gases, accuracy affected by radiation
” ”
-100/100 -200/70
”
-38/1000 -459/1000
Primary standard
”
High cost; accuracy affected by radiation in gases Accuracy affected by radiation in gases
Less than 0.02 to 5
Remote readings; temp by contact
-150/300
0.3
Up to 600
0.1 0.1 to 5
General testing of high temp; remote rapid readings by direct contact ” Same as above, especially suited for low temp
Up to 2200
0.1 to 15
Up to 1500 Up to 700
0.1 to 15 0.1 to 15
5. Beckman thermometers (metastatic)
For differential temp in same applications as in glass stem thermometer
9 diff
0.018
6. Bimetallic thermometers
For approx temp
0/1000
1, usually much more
7. Pressure-bulb thermometers Gas-filled bulb
Remote-testing
-100/1000
2
” ” For intensity of narrow spectra band of high temp radiation (remote)
20/500 -50/2100
2 2 15
9. Radiation pyrometers
For intensity of total high temp radiation (remote)
Any range
10. Seger cones (fusion pyrometers)
Approx temp (within temp source)
1000/3600
50
11. Indicating crayons
Approx temp (in surface)
125/900
12. Melting and boiling points of materials
Standards
All except extremely high temp
±1% Extremely precise
Iron-constantain thermocouple Copper-constantan thermocouple Chromel-constantan thermocouple
Vapor-filled bulb Liquid-filled bulb 8. Optical pyrometers
Standard for thermocouples
1500 upward
Reprinted by permission from ASHRAE Handbook - 1989 Fundamentals
”
Requires considerable skill to use
-320/1800
”
” ”
Less than 0.01
500/3000
Chromel-alumel thermocouple
106
” ”
Precision; remote readings; temp of fluids or solids by contact
Thermistors 4. Thermocouples Pt-Pt-Rh thermocouple
Approximate Range F
High cost; also, requires expensive measuring device Less accurate than above Subject to oxidation
Must be set for temp to be measured Time lag; unsuitable for remote use; unreliable Caution must be exercised so that installation is correct ” ”
For laboratory use only
AMCA 203-90 (R2007)
Table J.2 - Differential Pressure Measurement
No.
Measurement Means
Application
Range
Precision
Limitations
1. Micromanometer
Very low press. diff.
0 to 6 in. H20
0.005 to 0.001 in. H20
Not readily portable; not easy to use with pulsating pressure
2. Draft gauges
Moderately low press. diff.
0 to 10 in. H20
0.005 to 0.05 in. H20
Must be leveled carefully
3. Manometer
Medium press diff.
0 to 100 in. H20 or Hg
0.05 in.
Where used with liquid must be compensated for liquid density
4. Swinging-vane-type gauge
Moderately low press. diff.
0 to 0.5 in. H20 0 to 20 in. H20
5%
Generally usable to atmospheric pressure only
5. Bourdon-tube type
Medium to high press. diff., usually to atmosphere
Any
0.05 to 5%
Subject to damage due to over press-shock or pulsation
6. Pressure transducersstrain gauge, capacity, potentiometer, crystal, magnet
Remote reading, responds to rapid changes of pressure
0.05 to 50,000 psi
0.1 to 0.5%
Requires electronic amplifier and readout device
Table J.3 - Velocity Measurement No.
Measurement Means
Application
Range
Precision
5 to 50
10 to 20%
Limitations
1. Smoke puff or airborne solid tracer
Low air velocities in rooms; highly directional
2. Deflecting-vane anemometer
Air velocities in rooms, at outlets, etc; directional
30 to 24,000
5%
Not well suited for duct readings; needs periodic check calibration
3. Revolving-vane anemometer
Moderate air velocities in ducts and rooms; somewhat directional
100 to 3000
5 to 20%
Extremely subject to error with variations in velocities with space or time; easily damaged; needs periodic calibration
4. Pitot tube
Std instrument for measurement of duct velocities
180 to 10,000 with micromanometer 600 to 10,000 with draft gauges; 10,000 up with manometer
1 to 5%
Accuracy falls off at low end of range
5. Impact tube and sidewall or other static tap
High velocities, small tubes and where air direction may be variable
120 to 10,000 with micromanometer; 600 to 10,000 with draft gauges; 10,000 up with manometer
1 to 5%
Accuracy depends upon constancy of static pressure across stream section
6. Heated thermocouple anemometer
Air velocities in ducts, velocity distributions
10 to 2000
3 to 20%
Accuracy of some types not good at lower end of range; steady state measurements only
7. Hot-wire anemometer
(a) Low air velocities; directional and nondirectional available
1 to 1000
1 to 20%
Requires accurate calibration at frequent intervals; complex, costly
up to 60,000
1 to 20%
(b) High air velocities
Awkward to use but valuable in tracing air movement
(c) Transient velocity and turbulence
Reprinted by permission from ASHRAE Handbook - 1989 Fundamentals
107
AMCA 203-90 (R2007)
Annex K. Phase Current Method for Estimating the Power Output of Three Phase Fan Motors
Use Equation A to estimate the Hmo for motors of 5 horsepower and greater, operating at 90% or more of FLA. The uncertainties will be less than 5%.
The power output of three phase motors can be estimated based on the relationship of motor current and motor power output. Two equations can be used in estimating the motor power output. The equations are as follows:
Use the average of Equation A and Equation B to estimate the Hmo for all motors operating at less than 90% of FLA and for 3 horsepower and smaller motors operating above 90% of FLA. An estimated Hmo less than 50% of NPH can contain 15% uncertainties or greater.
Equation A: ⎛ Measured amps ⎞ ⎛ Measured volts ⎞ Hmo = NPH ⎜ ⎟⎜ ⎟ FLA NPV ⎝ ⎠⎝ ⎠ Where: Hmo = motor power output NPH = nameplate horsepower FLA = full load amps NPV = nameplate volts measured volts = average of the measured phase volts measured amps = average of the measured phase amps Equation B: ⎛ Measured amps - NLA ⎞ ⎛ Measured volts ⎞ Hmo = NPH ⎜ ⎟⎜ ⎟ FLA - NLA NPV ⎝ ⎠⎝ ⎠
Where: NLA = average of the measured phase values of no load amps NPH = nameplate horsepower FLA = full load amps NPV = nameplate volts NLA can usually be obtained with the motor operating and the motor shaft coupling or belt drive disconnected. In the case where the fan impeller is mounted directly on the motor shaft, it will be necessary to remove the impeller in order to obtain NLA measurements.
108
Figure K.1 represents the relationship of motor current and motor power output. The “dashed” lines between 0% NPH and 100% NPH for motor sizes shown represents Equation B. The solid lines between these same end points for the motor sizes shown represent the general shape of typical motor calibration amp/load curves. The solid line from 100% NPH and 100% FLA to 0% NPH and 0% FLA represents Equation A. These curves indicate that if you average the results of Equation A and Equation B for a specific measured amp draw, that your results approach the typical calibration curve. It also points out that the uncertainties are low if just Equation A is used above 90% FLA, especially in the larger integral motor horsepowers. Many fractional horsepower and small integral horsepower motors do not have a significant change in current from no load to full load. The actual ampsload characteristics for motors of the same horsepower rating can vary greatly from motor manufacturer to motor manufacturer. No load amperage (NLA) varies significantly for the same size motor between manufacturers. In addition, various motor design requirements result in different ampload characteristics even though the horsepower ratings of the motors are the same. These are some of the reasons that Figure K.1 cannot be used to determine the motor output directly. The chart is only intended to indicate the accuracy and suitability of using the above equations for estimating motor power output.
AMCA 203-90 (R2007) GENERALIZED CURVES ILLUSTRATING THE RELATIONSHIP OF HORSEPOWER TO AMPS FOR THREE PHASE MOTORS Do not use for determining actual motor horsepower DOTTED LINES PER EQUATION B: Hmo ∝ MEASURED AMPS - NLA/FLA - NLA 100
90
RATED HORSEPOWER 1 2
80
70 3 60 5 50 10 40 400 30 2500 20
10
0
0
10
20
30
40
50
60
70
80
90
100
% NAMEPLATE HORSEPOWER PER EQUATION A: Hmo ∝
MEASURED AMPS FLA
CAUTION: THIS CHART IS REPRESENTATIVE ONLY! SINCE THE AMP-LOAD CHARACTERISTICS OF THE SAME SIZE MOTOR WILL VARY BETWEEN THE VARIOUS MOTOR MANUFACTURERS, IT CANNOT BE USED TO DETERMINE THE HORSEPOWER OUTPUT OF A MOTOR. USE THE EQUATIONS AS DIRECTED ON THE PREVIOUS PAGE. 109
AMCA 203-90 (R2007)
Annex L. Estimated Belt Drive Loss
3) A larger belt section than required will increase the drive loss.
Drive loss is defined as follows: Percent drive loss equals power to driving sheave minus power from driven sheaves times 100, divided by power to driving sheave. There are several things which can affect belt drive efficiencies. Some of these are: 1) Over-designed drives. This was considered good practice at one time because the drive would last longer. It will still last longer but it is more inefficient. 2) Multiple belts on subminimum diameter sheaves are less efficient than fewer belts on larger diameter sheaves. Both the National Electric Motor Association and the Rubber Manufacturer’s Association publish data dealing with minimum recommended sheave diameters. As these minimum sheave diameters are approached, the drive loss becomes greater.
110
4) A badly undertensioned drive will increase the drive loss. 5) Misaligned drives will increase the drive loss. Drive loss is manifested as heat in belt drives. Under ambient conditions of less than 100°F, well designed drives that operate efficiently will be warm to the touch immediately after being shut down. If the drive is uncomfortable to the touch (approximately 140°F or more), then the drive loss is high. Obviously poorly tensioned and misaligned drives should be corrected before estimating brake horsepowers and drive losses.
AMCA 203-90 (R2007)
100 80
DRIVE LOSS, % MOTOR POWER OUTPUT*
60 40 30 RANGE OF DRIVE LOSSES FOR STANDARD BELTS
20 15 10 8 6 4 3 2 1.5 1 0.3 0.4 0.6 0.8 1
2
3
4
6
8 10
20
30 40
60 80 100
200 300 400
600
MOTOR POWER OUTPUT, hp
HIGHER BELT SPEEDS TEND TO HAVE HIGHER LOSSES THAN LOWER BELT SPEEDS AT THE SAME HORSEPOWER *Drive losses are based on the conventional V-belt, which has been the “work horse” of the drive industry for several decades. EXAMPLE • Motor power output, Hmo, is determined to be 13.3 hp • The belts are the standard type and just warm to the touch immediately after shutdown • From chart, drive loss = 5.1% • Drive loss, HL = 0.051 × 13.3 = 0.7 hp • Fan power input, H = 13.3 - 0.7 = 12.6 hp Figure L.1 - Estimated Belt Drive Loss
111
AMCA 203-90 (R2007)
Annex M. Density Determinations
p 1 = pb = 28.60 in. Hg
M.1 General The wet-bulb depression is: This annex contains examples illlustrating the procedures for determining densities. Determinations of densities are shown for air and for gases other than air.
td1 - tw1 = 78 - 62 = 16°F
M.2 Determination of the density of air, general case
For wet-bulb depression of 16°F, dry-bulb temperature of 78°F and absolute pressure of 28.60 in. Hg, obtain ρ1 = 0.0701 lbm/ft3 by using the Psychrometric Density Chart in Figure N.1 in Annex N.
Determine air density by using the Psychrometric Density Chart, shown in Figure N.1 in Annex N, the Psychrometric Density Table, shown in Annex N, or a calculation procedure which makes use of perfect gas relationships and the modified Apjohn equation for partial vapor pressure. Examples of the use of these procedures are included in this section. Each of the procedures requires knowledge of the pressure, dry-bulb temperature and wet-bulb temperature of the air. The Psychrometric Density Chart and the Psychrometric Density Table are limited to the temperatures and pressures normally encountered in fan applications. Limit the use of the calculation procedure that is based on perfect gas relationships and illustrated in Example M2.3, to instances in which the dry-bulb temperature is 180°F or less. Accurate wet-bulb temperature measurements are difficult to obtain when the dry-bulb temperature exceeds 180°F. When the dry-bulb temperature exceeds 180°F, it may be necessary to rely on site personnel for the water vapor content of the air. Alternately, commercially available instrumentation for dew point determination may be used. For the procedure required to determine density based on the data provided in either of the above cases, refer to Psychrometric Tables and Charts by Zimmerman and Lavine.1 EXAMPLE M2.1
The conditions at a fan inlet, located at an elevation of 1000 ft above sea level are: Ps1 = -3.45 in. wg td1 = 85°F tw1 = 75°F Barometric pressure, obtained from a nearby airport, is 29.82 in. Hg at sea level. Using the data in Figure N.3 in Annex N, the barometric pressure at 1000 ft above sea level is: pb = 29.82 × 0.964 = 28.75 in. Hg The absolute pressure at the fan inlet is: p1 = pb + (Ps1/13.6) = 28.75 + (-3.45/13.6) = 28.50 in. Hg The wet-bulb depression is: td1 - tw1 = 85 - 75 = 10°F For dry-bulb temperature of 85°F, absolute pressure of 28.50 in. Hg and wet-bulb depression of 10°F, use the Psychrometric Density Table in Figures N.5 in Annex N to obtain:
The conditions that exist at the inlet of a fan that is not ducted on the inlet side are:
ρ1 = 0.06829 + 10 × 0.000041 = 0.0687 lbm/ft3
td1 = 78°F tw1 = 62°F
Example M2.3
Since: Ps1 = 0
112
EXAMPLE M2.2
The conditions at a fan inlet are: Ps1 = -8.75 in. wg td1 = 146°F tw1 = 93°F 1. O. T. Zimmerman and I. Lavine, Psychrometric Tables and Charts, 2nd ed. (Dover, N.H.: Industrial Research Service Inc., 1964)
AMCA 203-90 (R2007) The barometric pressure, pb, measured for the atmosphere to which Ps1 is referred, is 28.15 in. Hg.
Barometric pressure, obtained from a nearby airport, is 29.24 in. Hg at sea level.
The absolute pressure at the fan inlet is:
Using the data in Figure N.3 in Annex N, the barometric pressure at 1000 ft above seal level is:
p1 = pb + (Ps1 /13.6) = 28.15 + (-8.75/13.6) = 27.51 in. Hg
pb = 29.24 × 0.964 = 28.19 in. Hg
Use Figure N.2 in Annex N to obtain saturated vapor pressure, pe, of 1.562 in. Hg for the wet-bulb temperature of 93°F. Use the modified Apjohn equation for partial vapor pressure, pp, to obtain: pp = pe - p1 (td1 - tw1)/2700 = 1.562 - 27.51 (146 - 93)/2700 = 1.022 in. Hg
ρ1 is calculated by using perfect gas relationships:
ρ1 =
=
1.3257 ( p1 − 0.378 pp )
( td1 + 460 )
1.3257 ( 27.51 − 0.378 × 1.022 )
(146 + 460 )
= 0.0593 lbm/ft 3
The absolute pressure at the fan inlet is: p1 = pb + (Ps1/13.6) = 28.19 + (-15/13.6) = 27.09 in. Hg Dry air at 29.92 in. Hg and 70°F has a density of 0.075 lbm/ft3. Consider the density of air to be directly proportional to absolute pressure and inversely proportional to absolute temperature. The density of the air at the fan inlet is calculated as follows:
ρ1 = 0.075 (p1/29.92) [(70 + 460)/(td1 + 460)] = 0.075 (27.09/29.92) [530/(95 + 460)] = 0.0648 lbm/ft3 EXAMPLE M3.2 Saturated air is enterting a fan inlet, located at an elevation of 1500 ft above sea level. The pressure and temperature at the inlet are:
M.3 Determination of the density of air, special cases
Ps1 = - 6.75 in. wg td1 = 103°F
The procedures for the determination of the density of air that are described in Section M.2 are valid for dry air, air that is saturated with water vapor and air that is partially saturated with water vapor. This section contains alternate procedures for cases in which it is known that the air is either dry or saturated. Knowledge that the air is either dry or saturated eliminates the usual requirement of the wet-bulb temperature determination; however, it should be noted that an incorrect assumption of either of these conditions can result in a significant uncertainty in the density determination.
Barometric pressure, obtained from a nearby airport, is 29.66 in. Hg at sea level.
EXAMPLE M3.1 Dry air is entering a fan inlet, located at an elevation of 1000 ft above sea level. The pressure and temperature at the inlet are: Ps1 = -15 in. wg td1 = 95°F
Using the data in Figure N.3 in Annex N, the barometric pressure at 1500 ft above sea level is: pb = 29.66 × 0.947 = 28.09 in. Hg The absolute pressure at the fan inlet is: p1 = pb + (Ps1/13.6) = 28.09 + (-6.75/13.6) = 27.59 in. Hg Refer to Figure N.4 in Annex N to obtain saturated air density of 0.06868 at 103°F and 29.92 in. Hg. Assuming the density of saturated air to be directly proportional to absolute pressure, the density at the fan inlet is calculated as follows: 113
AMCA 203-90 (R2007)
ρ1 = 0.06868 (p1/29.92) = 0.06868 (27.59/29.92) = 0.0633 lbm/ft3 Assuming the density of saturated air to be directly proportional to absolute pressure is an approximation. The uncertainty in the density determination as a result of this approximation increases with increasing temperature and increases with increasing variation between the actual absolute pressure and 29.92 in. Hg, which is the stated pressure for the data in Figure N.4. The uncertainty will be approximately 1% or less under the following conditions: • At 120°F and at an absolute pressure within 20% of 29.92 in. Hg • At 150°F and at an absolute pressure within 10% of 29.92 in. Hg • At 180°F and at an absolute pressure within 4% of 29.92 in. Hg M.4 DETERMINATION OF THE DENSITY OF A GAS OTHER THAN AIR The determination of the density of a gas other than air may require the use of complex equipment. Unless specifically qualified, an expert should be consulted for the proper use of the equipment. If the gas is a complex mixture of various consitutuents, as found in certain industrial processes, it is suggested that the company chemist be consulted for the gas analysis. Particular care should be used if the gas is toxic, corrosive or explosive; and in these cases, consideration should be given to substituting air for the test.
The apparent molecular weight of the gas is determined as follows: Volume Molecular Component Fraction × Weight = lb/mole CO2 CO O2 H2 N2
0.055 0.01 0.15 0.01 0.775
44 28 32 2 28
1.00
2.42 0.28 4.80 0.02 21.70 29.22
Apparent molecular weight = (29.22/1.00) = 29.22 The density of the gas at 70°F and 29.92 in. Hg is calculated as follows: Apparent molecular weight 29.22 = 386.7 386.7 = 0.0756 lbm/ft 3
Using the data in Figure N.3 in Annex N, the barometric pressure at 2000 ft above sea level is: pb = 29.92 × 0.930 = 27.83 in. Hg
The first two examples in this section illustrate gas density determinations based on analyses that provide the relative amounts of the gas constituents. Typical flue gas density data, which is provided in Figure N.6 in Annex N, is illustrated in Example M4.3. Since the actual density may be significantly different from the density determined by using typical data, it is recommended that the typical data be used only in the even that more specific information is not available.
The absolute pressure at the fan inlet is:
EXAMPLE M4.1
ρ1 = 0.0756 (p1/29.92)[(70 + 460)/(td1 + 460)] = 0.0756 (26.21/29.92) [530/(230 + 460)] = 0.0509 lbm/ft3
A gas is entering a fan inlet located at an elevation of 2000 ft above sea level. The pressure and temperature at the inlet are: Ps1 = - 22 in. wg td1 = 230°F 114
Barometric pressure, obtained from a nearby airport, is 29.92 in. Hg at sea level. The composition of the gas is 5.5% CO2, 1% CO, 15% O2, 1% H2, and 77.5% N2, by volume.
p1 = pb + (Ps1/13.6) = 27.83 + (-22/13.6) = 26.21 in. Hg Consider the density of the gas to be directly proportional to absolute pressure and inversely proportional to absolute temperature. The density of the gas at the fan inlet is calculated as follows:
EXAMPLE M4.2 The conditions that exist at the inlet of a fan are Ps1 = -19.5 in. wg and td1 = 240°F. The barometric pressure,
AMCA 203-90 (R2007) pb, measured for the atmospheric to which Ps1 is referred is 29.35 in. Hg. The composition of the gas is 5.5% CO2, 1% CO, 15% O2, 1% H2, and 77.5% N2 by weight. The apparent molecular weight of the gas is determined as follows: Volume Molecular Component Fraction × Weight = lb/mole CO2 CO O2 H2 N2
0.055 0.01 0.15 0.01 0.775
44 28 32 2 28
1.00
0.00125 0.00036 0.0047 0.005 0.0277 0.0390
Apparent molecular weight = 1/0.0390 = 25.6 The density of the gas at 70°F and 29.92 in. Hg is calculated as follows: Apparent molecular weight 25.6 = 386.7 386.7 = 0.0662 lbm/ft 3
EXAMPLE M4.3 Flue gas is flowing at Plane 3, the Pitot traverse measurement plane. The flue gas is the result of using natural gas as the fuel. The conditions that exsit at Plane 3 are: Ps3 = 5.74 in. wg td3 = 680°F The barometric pressure, pb, measured for the atmosphere to which Ps3 is referred is 28.85 in. Hg. The absolute pressure at Plane 3 is: p3 = pb + (Ps3/13.6) = 28.85 + (5.74/13.6) = 29.27 in. Hg Refer to Figure N.6 in Annex N to obtain typical flue gas density when natural gas is used as the fuel of 0.0725 lbm/ft3 at 70°F and 29.92 in. Hg. Consider the density of the flue gas to be directly proportional to absolute pressure and inversely proportional to absolute temperature. The density of the gas at Plane 3 is calculated as follows:
ρ1 = 0.0725 (p3/29.92)[(70 + 460)/(td3 + 460)] = 0.0725 (29.27/29.92) [530/(680 + 460)] = 0.0330 lbm/ft3
The absolute pressure at the fan inlet is: p1 = pb + (Ps1/13.6) = 29.35 + (-19.5/13.6) = 27.92 in. Hg Consider the density of the gas to be directly proportional to absolute pressure and inversely proportional to absolute temperature. The density of the gas at the fan inlet is calculated as follows:
ρ1 = 0.0662 (p1/29.92)[(70 + 460)/(td1 + 460)] = 0.0662 (27.92/29.92) [530/(240 + 460)] = 0.0468 lbm/ft3
115
AMCA 203-90 (R2007)
116
Fold out for Figure N.1 - Psychrometric Density Charts
Annex N. Density Charts and Tables
AMCA 203-90 (R2007)
117
118
.1646 .1724 .1805 .1879 .1956 .2036 .2118 .2204 .2292 .2384 .2478 .2576 .2678 .2783 .2892 .3004 .3121 .3241 .3365 .3494 .3626 .3764 .3905 .4052 .4203 .4359 .4520 .4687 .4859 .5036
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59
85 86 87 88 89
80 81 82 83 84
75 76 77 78 79
70 71 72 73 74
65 66 67 68 69
60 61 62 63 64
tw °F
1.214 1.254 1.294 1.336 1.379
1.033 1.067 1.103 1.139 1.176
.8757 .9053 .9359 .9673 .9997
.7397 .7653 .7917 .8188 .8468
.6225 .6445 .6667 .6906 .7148
.5219 .5408 .5603 .5804 .6011
pe in. Hg
115 116 117 118 119
110 111 112 113 114
105 106 107 108 109
100 101 102 103 104
95 96 97 98 99
90 91 92 93 94
tw °F
2.999 3.085 3.173 3.263 3.356
2.599 2.675 2.753 2.833 2.915
2.246 2.313 2.381 2.452 2.525
1.935 1.994 2.054 2.117 2.180
1.662 1.714 1.767 1.821 1.877
1.423 1.468 1.515 1.562 1.611
pe in. Hg
145 146 147 148 149
140 141 142 143 144
135 136 137 138 139
130 131 132 133 134
125 126 127 128 129
120 121 122 123 124
tw °F
6.689 6.860 7.034 7.212 7.394
5.889 6.043 6.199 6.359 6.522
5.173 5.310 5.450 5.593 5.740
4.531 4.654 4.779 4.908 5.038
3.960 4.069 4.180 4.295 4.412
3.451 3.548 3.647 3.749 3.853
pe in. Hg
175 176 177 178 179 180
170 171 172 173 174
165 166 167 168 169
160 161 162 163 164
155 156 157 158 159
150 151 152 153 154
tw °F
13.69 14.00 14.32 14.64 14.94 15.31
12.21 12.50 12.79 13.08 13.38
10.88 11.13 11.40 11.66 11.94
9.665 9.898 10.14 10.38 10.63
8.569 8.779 8.994 9.213 9.437
7.580 7.770 7.963 8.161 8.362
pe in. Hg
Figure N.2 - Thermodynamic Properties of Water at Absolute Vapor Pressures, Inches of Mercury
Adapted from ASHRAE Handbook - 1989 Fandamentals
pe in. Hg
tw °F
AMCA 203-90 (R2007)
WET-BULB DEPRESSION, °F
74
72
70
68
66
64
62
60
58
56
54
52
50
48
46
44
42
0
Figure N.1 - Psychrometric Density Chart
DRY-BULB TEMPERATURE, °F
0.080
0.079 76
2 78
0.078
4
80
0.077
6
82
0.076
8
84
0.075
10
86
0.074
0.070
0.069
0.068
0.067
0.066
12
88
Wet-bulb depression = 4°F; proceed horizontally to 54°F dry-bulb temperature; read vertically to 29.9 in. Hg; read horizontally to the density -- ρ = 0.0769 lbm/ft3.
Solution:
•
0.073
90
td = 54°F; tw = 50°F; pb = 29.9 in. Hg
Given:
•
0.065
0.064
14
92
4.
0.072
94
Read vertically to the absolute pressure. Then read horizontally to the density.
3.
0.063
0.062
16
96
Proceed horizontally to the appropriate dry-bulb temperature.
2.
Example
Calculate wet-bulb depression. Enter chart at the left.
1.
0.061
0.060
0.071
98
28.0 28.2 28.4 28.6 28.8 29.0 29.2 29.4 29.6 29.8 30.0
g
18
20
22
24
26
28
30
32
34
36
38
40
TE
OLU
ABS
in. H
URE
PRE SS
AMCA 203-90 (R2007)
AIR DENSITY, lbm/ft3
AMCA 203-90 (R2007)
ALTITUDE ft.
SPECIFIC GRAVITY
PRESSURE in. Hg
ALTITUDE ft.
SPECIFIC GRAVITY
PRESSURE in. Hg
0 100 200 300 400
1.00 0.996 0.993 0.989 0.986
29.92 29.81 29.70 29.60 29.49
3000 3200 3400 3600 3800
0.896 0.890 0.833 0.877 0.870
26.82 26.62 26.42 26.23 26.03
500 600 700 800 900
0.982 0.979 0.975 0.971 0.968
29.38 29.28 29.17 29.07 28.96
4000 4200 4400 4600 4800
0.864 0.857 0.851 0.845 0.838
25.84 25.65 25.46 25.27 25.08
1000 1100 1200 1300 1400
0.964 0.961 0.957 0.954 0.950
28.86 28.75 28.65 28.54 28.44
5000 5200 5400 5600 5800
0.832 0.826 0.820 0.814 0.807
24.90 24.71 24.52 24.34 24.16
1500 1600 1700 1800 1900
0.947 0.944 0.940 0.937 0.933
28.33 28.23 28.13 28.02 27.92
6000 6500 7000 7500 8000
0.801 0.786 0.772 0.757 0.743
23.98 23.53 23.09 22.65 22.22
2000 2100 2200 2300 2400
0.930 0.926 0.923 0.920 0.916
27.82 27.72 27.62 27.52 27.42
8500 9000 9500 10000 15000
0.729 0.715 0.701 0.688 0.564
21.80 21.39 20.98 20.58 16.89
2500 2600 2700 2800 2900
0.913 0.909 0.906 0.903 0.899
27.32 27.21 27.11 27.01 26.91
20000 25000 30000 35000 40000
0.460 0.371 0.297 0.235 0.185
13.75 11.10 8.89 7.04 5.54
Note: Specific gravity of standard air at sea level and 29.92 in. Hg = 1.00 Figure N.3 - Relative Specific Gravity of Air at Various Altitudes1
1. Robert Jorgensen, ed., Fan Engineering, 7th ed. (Buffalo, NY, Buffalo Forge Co., 1970) p.8 - Reprinted by Permission
119
AMCA 203-90 (R2007)
PROPERTIES OF SATURATED AIR2
Temp °F
WEIGHT IN A CUBIC FOOT OF MIXTURE
VOLUME ft3/lb
WEIGHT OF THE VAPOR
OF TOTAL lb/lb lb/lb DRY AIR VAPOR WEIGHT DRY AIR OF OF lb lb lb DRY AIR MIXTURE
Temp °F
WEIGHT IN A CUBIC FOOT OF MIXTURE
VOLUME ft3/lb
WEIGHT OF THE VAPOR
OF TOTAL lb/lb lb/lb DRY AIR VAPOR WEIGHT DRY AIR OF OF lb lb lb DRY AIR MIXTURE
-25 -20 -15 -10 -5
.09134 .09025 .08922 .08820 .08723
.000018 .000024 .000031 .000041 .000053
.09136 .09027 .08925 .08824 .08728
10.95 11.07 11.21 11.34 11.46
.00020 .00027 .00035 .00046 .00061
.00020 .00027 .00035 .00046 .00061
46 47 48 49 50
.07768 .00750 .07731 .07714 .07694
.000509 .000527 .000545 .000567 .000587
.07819 .07803 .07785 .07771 .07753
12.87 12.90 12.93 12.96 12.99
.00655 .00680 .00705 .00734 .00762
.00651 .00675 .00700 .00728 .00756
0 5 10 15 20
.08625 .08529 .08434 .08340 .08247
.000068 .000087 .000110 .000140 .000176
.08632 .08538 .08445 .08354 .08264
11.59 11.72 11.85 11.99 12.12
.00080 .00102 .00130 .00168 .00213
.00080 .00102 .00130 .00168 .00213
51 52 53 54 55
.07676 .07657 .07637 .07620 .07600
.000608 .000632 .000651 .000675 .000700
.07737 .07720 .07702 .07687 .07670
13.02 13.06 13.09 13.12 13.15
.00792 .00823 .00854 .00884 .00921
.00786 .00819 .00845 .00877 .00913
21 22 23 24 25
.08230 .08210 .08193 .08173 .08156
.000185 .000193 .000202 .000213 .000222
.08248 .08229 .08213 .08194 .08178
12.15 12.18 12.20 12.23 12.26
.00225 .00235 .00246 .00260 .00272
.00224 .00234 .00245 .00259 .00271
56 57 58 59 60
.07582 .07562 .07544 .07524 .07506
.000723 .000749 .000775 .000801 .000829
.07654 .07637 .07622 .07604 .07589
13.19 13.22 13.25 13.29 13.32
.00952 .00989 .01026 .01063 .01103
.00943 .00980 .01016 .01052 .01091
26 27 28 29 30
.08136 .08117 .08099 .08083 .08063
.000233 .000243 .000254 .000264 .000277
.08159 .08141 .08124 .08109 .08090
12.29 12.32 12.34 12.37 12.40
.00285 .00300 .00314 .00328 .00345
.00284 .00299 .00313 .00327 .00344
61 62 63 64 65
.07486 .07468 .07447 .07429 .07408
.000857 .000886 .000916 .000947 .000979
.07572 .07557 .07539 .07524 .07506
13.35 13.39 13.42 13.46 13.49
.01143 .01185 .01229 .01273 .01320
.01130 .01171 .01214 .01257 .01303
31 32 33 34 35
.08043 .08025 .08006 .07989 .07970
.000290 .000303 .000315 .000327 .000339
.08072 .08055 .08038 .08022 .08004
12.43 12.46 12.49 12.51 12.54
.00362 .00378 .00393 .00409 .00426
.00361 .00376 .00392 .00408 .00425
66 67 68 69 70
.07390 .07369 .07350 .07330 .07310
.001012 .001045 .001080 .001115 .001152
.07491 .07473 .07458 .07441 .07425
13.53 13.57 13.60 13.64 13.68
.01368 .01417 .01468 .01520 .01576
.01349 .01397 .01447 .01497 .01551
36 37 38 39 40
.07952 .07933 .07916 .07897 .07880
.000353 .000364 .000380 .000394 .000409
.07987 .07969 .07954 .07936 .07921
12.57 12.60 12.63 12.66 12.69
.00444 .00460 .00480 .00499 .00519
.00442 .00458 .00478 .00496 .00516
71 72 73 74 75
.07290 .07270 .07250 .07229 .07208
.001189 .001229 .001268 .001310 .001352
.07409 .07393 .07377 .07360 .07343
13.71 13.75 13.79 13.83 13.87
.01630 .01691 .01748 .01812 .01876
.01604 .01662 .01717 .01780 .01841
41 42 43 44 45
.07860 .07843 .07825 .07805 .07788
.000425 .000440 .000456 .000473 .000491
.07902 .07887 .07871 .07852 .07837
12.72 12.75 12.78 12.81 12.84
.00541 .00561 .00583 .00606 .00630
.00538 .00558 .00579 .00602 .00626
76 77 78 79 80
.07188 .07166 .07144 .07124 .07104
.001395 .001439 .001485 .001532 .001579
.07328 .07310 .07293 .07277 .07262
13.91 13.95 13.99 14.03 14.08
.01941 .02008 .02079 .02150 .0223
.01904 .01968 .02036 .02106 .02174
Figure N.4 - Weights of Air, Water Vapor, and Saturated Mixture of Air and Water Vapor at Different Temperatures and 29.92 in. Hg
2. Jorgensen, op. cit., pp 15-17
120
Reprinted by Permission
AMCA 203-90 (R2007)
PROPERTIES OF SATURATED AIR2
Temp °F
WEIGHT IN A CUBIC FOOT OF MIXTURE
VOLUME ft3/lb
WEIGHT OF THE VAPOR
OF TOTAL lb/lb lb/lb DRY AIR VAPOR WEIGHT DRY AIR OF OF lb lb lb DRY AIR MIXTURE
Temp °F
WEIGHT IN A CUBIC FOOT OF MIXTURE
VOLUME ft3/lb
WEIGHT OF THE VAPOR
OF TOTAL lb/lb lb/lb DRY AIR VAPOR WEIGHT DRY AIR OF OF lb lb lb DRY AIR MIXTURE
81 82 83 84 85
.07081 .07059 .07038 .07015 .06993
.001629 .001680 .001733 .001785 .001840
.07244 .07227 .07211 .07193 .07177
14.12 14.16 14.21 14.26 14.30
.02301 .02380 .02462 .02545 .02631
.02249 .02325 .02403 .02482 .02566
116 117 118 119 120
.06186 .06154 .06124 .06092 .06060
.004427 .004548 .004669 .004794 .004921
.06629 .06609 .06591 .06571 .06552
16.16 16.24 16.32 16.41 16.50
.07157 .07390 .07625 .07869 .08121
.06678 .06882 .07084 .07296 .07511
86 87 88 89 90
.06970 .06947 .06925 .06902 .06880
.001898 .001954 .002014 .002072 .002139
.07160 .07142 .07126 .07109 .07094
14.34 14.39 14.44 14.48 14.53
.02723 .02813 .02908 .03002 .03109
.02651 .02736 .02826 .02915 .03015
121 122 123 124 125
.06027 .05995 .05960 .05927 .05892
.005049 .005183 .005319 .005456 .005598
.06532 .06513 .06492 .06473 .06452
16.58 16.68 16.77 16.87 16.96
.08376 .08646 .08925 .09204 .09502
.07729 .07958 .08194 .08428 .08677
91 92 93 94 95
.06855 .06832 .06809 .06785 .06760
.002201 .002267 .002334 .002404 .002474
.07075 .07058 .07042 .07025 .07007
14.58 14.63 14.69 14.73 14.79
.03211 .03318 .03428 .03543 .03660
.03111 .03212 .03314 .03422 .03531
130 135 140 145 150
.05713 .05524 .05319 .05100 .04865
.006355 .007195 .008128 .009162 .010303
.06349 .06244 .06132 .06016 .05895
17.49 18.10 18.79 19.60 20.55
.11125 .13026 .15280 .17966 .21178
.10010 .11523 .13255 .15230 .17478
96 97 98 99 100
.06736 .06711 .06688 .06660 .06634
.002546 .002620 .002692 .002770 .002853
.06991 .06973 .06957 .06931 .06919
14.84 14.90 14.95 15.01 15.07
.03780 .03904 .04025 .04159 .04300
.03642 .03757 .03870 .03993 .04124
155 160 165 170 175
.04612 .04340 .04048 .03734 .03398
.011547 .012937 .014436 .016118 .017926
.05767 .05634 .05492 .05346 .05191
21.67 23.03 24.69 26.77 29.43
.25038 .29810 .35660 .43168 .52750
.20022 .22962 .26285 .30150 .34530
101 102 103 104 105
.06610 .06583 .06557 .06530 .06504
.002937 .003019 .003106 .003193 .003283
.06904 .06885 .06868 .06849 .06832
15.12 15.18 15.25 15.31 15.37
.04443 .04586 .04737 .04890 .05048
.04255 .04385 .04523 .04662 .04806
180 185 190 195 200
.03035 .02645 .02228 .01779 .01297
.019905 .022062 .024393 .026957 .029730
.05036 .04851 .04667 .04475 .04270
32.94 37.78 44.85 56.20 77.11
.65580 .83410 1.0948 1.5153 2.2923
.39525 .45425 .52270 .60240 .69660
106 107 108 109 110
.06477 .06451 .06421 .06394 .06364
.003375 .003470 .003568 .003666 .003766
.06814 .06798 .06778 .06761 .06741
15.44 15.50 15.57 15.64 15.71
.05212 .05379 .05556 .05734 .05917
.04953 .05105 .05264 .05422 .05587
205 210 212
.00782 .00232 .00000
.032715 .035942 .037298
.04064 .03836 .03730
127.9 431.0 ____
4.1838 15.493 Inf.
.80500 .93700 1.0000
111 112 113 114 115
.06336 .06306 .06278 .06247 .06216
.003872 .003978 .004085 .004199 .004311
.06723 .06704 .06686 .06667 .06647
15.78 15.85 15.93 16.00 16.08
.06111 .06308 .06507 .06722 .06935
.05760 .05934 .06110 .06299 .06486
Figure N.4 - Weights of Air, Water Vapor, and Saturated Mixture of Air and Water Vapor at Different Temperatures and 29.92 in. Hg
2. Jorgensen, op. cit., pp 15-17
Reprinted by Permission
121
AMCA 203-90 (R2007)
Density of Saturated Air for Various Barometric Conditions - lbm/ft3 Dry-Bulb Temp. °F
Barometric Pressure in. Hg
Approximate average increase in Increase in density per density per °F wet-bulb 0.1 in. pressure depression
28.5
29.0
29.5
30.0
30.5
31.0
30 31 32 33 34
.07703 .07687 .07671 .07654 .07638
.07839 .07822 .07806 .07789 .07772
.07974 .07957 .07940 .07924 .07907
.08110 .08093 .08075 .08058 .08041
.08245 .08228 .08210 .08193 .08175
.08380 .08363 .08345 .08327 .08310
.00027 .00027 .00027 .00027 .00027
.000017 .000017 .000017 .000018 .000018
35 36 37 38 39
.07621 .07605 .07589 .07573 .07557
.07756 .07739 .07723 .07706 .07690
.07890 .07873 .07856 .07840 .07823
.08024 .07807 .07990 .07973 .07956
.08158 .08141 .08123 .08106 .08089
.08292 .08274 .08257 .08239 .08222
.00027 .00027 .00027 .00027 .00027
.000018 .000018 .000019 .000019 .000019
40 41 42 43 44
.07541 .07525 .07509 .07493 .07477
.07674 .07657 .07641 .07625 .07609
.07806 .07790 .07773 .07757 .07740
.07939 .07922 .09705 .07889 .07872
.08072 .08055 .08038 .08021 .08004
.08205 .08187 .08170 .08153 .08135
.00027 .00026 .00026 .00026 .00026
.000019 .000020 .000020 .000020 .000020
45 46 47 48 49
.07461 .07445 .07429 .07413 .07397
.07592 .07576 .07560 .07544 .07528
.07724 .07707 .07691 .07674 .07658
.07855 .07838 .07822 .07805 .07788
.07986 .07970 .07953 .07936 .07919
.08118 .08101 .08084 .08066 .08049
.00026 .00026 .00026 .00026 .00026
.000020 .000021 .000021 .000021 .000022
50 51 52 53 54
.07381 .07366 .07350 .07334 .07318
.07512 .07496 .07479 .07464 .07447
.07642 .07625 .07609 .07593 .07576
.07772 .07755 .07739 .07722 .07706
.07902 .07885 .07868 .07852 .07835
.08032 .08015 .07998 .07981 .07964
.00026 .00026 .00026 .00026 .00026
.000022 .000022 .000023 .000023 .000023
55 56 57 58 59
.07302 .07287 .07271 .07255 .07240
.07431 .07415 .07399 .07383 .07367
.07560 .07544 .07528 .07512 .07495
.07689 .07673 .07656 .07640 .07623
.07818 .07801 .07784 .07768 .07751
.07947 .07930 .07913 .07896 .07879
.00026 .00026 .00026 .00026 .00026
.000024 .000024 .000025 .000025 .000025
60 61 62 63 64
.07224 .07208 .07193 .07177 .07161
.07352 .07336 .07320 .07304 .07288
.07479 .07463 .07447 .07430 .07414
.07607 .07590 .07574 .07557 .07541
.07734 .07718 .07701 .07684 .07668
.07862 .07845 .07828 .07811 .07794
.00026 .00026 .00026 .00026 .00026
.000026 .000026 .000027 .000027 .000028
Note: Approximate average decrease in density per 0.1°F rise in dry-bulb temperature equals .000017 lbm/ft3. Figure N.5 - Psychrometric Density Table (I-P) 122
AMCA 203-90 (R2007)
Psychrometric Density Table (I-P) Density of Saturated Air for Various Barometric Conditions - lbm/ft3 Dry-Bulb Temp. °F
Barometric Pressure in. Hg
Approximate average increase in Increase in density per density per °F wet-bulb 0.1 in. pressure depression
28.5
29.0
29.5
30.0
30.5
31.0
65 66 67 68 69
.07145 .07130 .07114 .07098 .07083
.07272 .07256 .07240 .07224 .07208
.07398 .07382 .07366 .07350 .07333
.07525 .07508 .07492 .07475 .07459
.07651 .07634 .07618 .07601 .07584
.07770 .07760 .07744 .07727 .07710
.00026 .00026 .00026 .00026 .00026
.000028 .000029 .000029 .000030 .000030
70 71 72 73 74
.07067 .07051 .07035 .07020 .07004
.07192 .07176 .07160 .07144 .07128
.07317 .07301 .07285 .07268 .07252
.07442 .07426 .07410 .07393 .07377
.07568 .07551 .07534 .07517 .07501
.07693 .07676 .07659 .07642 .07625
.00026 .00025 .00025 .00025 .00025
.000031 .000031 .000032 .000033 .000033
75 76 77 78 79
.06988 .06972 .06956 .06940 .06925
.07112 .07096 .07080 .07064 .07048
.07236 .07220 .07203 .07187 .07171
.07360 .07343 .07327 .07310 .07294
.07484 .07467 .07451 .07434 .07417
.07603 .07591 .07574 .07557 .07540
.00025 .00025 .00025 .00025 .00025
.000034 .000034 .000035 .000036 .000036
80 81 82 83 84
.06909 .06893 .06877 .06861 .06845
.07032 .07015 .07000 .06983 .06967
.07155 .07138 .07122 .07105 .07089
.07277 .07261 .07244 .07227 .07211
.07400 .07383 .07366 .07349 .07333
.07523 .07506 .07489 .07472 .07454
.00025 .00025 .00024 .00024 .00024
.000037 .000038 .000039 .000039 .000040
85 86 87 88 89
.06829 .06812 .06796 .06780 .06764
.06950 .06934 .06917 .06901 .06885
.07072 .07056 .07039 .07022 .07005
.07194 .07177 .07160 .07143 .07126
.07316 .07299 .07281 .07264 .07247
.07437 .07420 .07403 .07385 .07368
.00024 .00024 .00024 .00024 .00024
.000041 .000042 .000043 .000043 .000044
90 91 92 93 94
.06748 .06731 .06715 .06698 .06682
.06868 .06852 .06835 .06818 .06801
.06989 .06972 .06955 .06938 .06921
.07109 .07092 .07075 .07058 .07041
.07230 .07213 .07195 .07178 .07161
.07351 .07333 .07316 .07298 .07280
.00024 .00024 .00024 .00024 .00024
.000045 .000046 .000047 .000048 .000049
95 96 97 98 99
.06665 .06648 .06632 .06615 .06598
.06785 .06768 .06751 .06734 .06717
.06904 .06887 .06870 .06853 .06835
.07024 .07006 .06989 .06972 .06954
.07143 .07126 .07108 .01091 .07073
.07263 .07245 .07227 .07209 .07191
.00024 .00024 .00024 .00024 .00024
.000050 .000051 .000052 .000053 .000054
100
.06581
.06700
.06818
.06937
.07055
.07174
.00024
.000055
Note: Approximate average decrease in density per 0.1°F rise in dry-bulb temperature equals .000017 lbm/ft3. Figure N.5 - Psychrometric Density Table (I-P) 123
AMCA 203-90 (R2007)
FUEL
FLUE GAS DENSITY lbm/ft3
COAL
0.078
OIL
0.075
NATURAL GAS
0.0725
BAGASSE
0.070
BLAST FURNACE GAS
0.076
LIGNITE
0.073
WOOD
0.070
The above densities at 70°F and 29.92 in. Hg are based on average fuel analyses and moisture contents Figure N.6 - Typical Densities for Various Flue Gases
124
AMCA 203-90 (R2007)
Annex P. Diffusion at Fan Outlets
BLAST AREA DISCHARGE DUCT CUTOFF
OUTLET AREA
25% 50% 75% CENTRIFUGAL FAN 100% EFFECTIVE DUCT LENGTH AXIAL FAN
To calculate 100% effective duct length, assume a minimum of 2½ duct diameters for 2500 fpm or less. Add 1 duct diameter for each additional 1000 fpm. Example: 5000 fpm = 5 equivalent duct diameters If the duct is rectangular, with side dimensions equal to a and b, the equivalent duct diameter is equal to (4ab/π)0.5
Figure P.1 - Controlled Diffusion and Establishment of a Uniform Velocity Profile in a Straight Length of Outlet Duct
125
AMCA 203-90 (R2007)
Annex R. Terminology for Fans and Air Handling Units CASING
BACKPLATE RIM INLET
HUB
MOTOR GUIDE VANE
BLADE IMPELLER
INLET BELL
Tubular Centrifugal Fan - Direct Drive CASING
BLADE DIFFUSER HUB
MOTOR
IMPELLER CASING
Tubeaxial Fan-Direct Drive (Impeller Downstream)
BEARING CASING BELT TUBE BLADE
HUB
GUIDE VANE
Vaneaxial Fan-Belt Drive IMPELLER
INLET BOX
BEARINGS
FAN CASING
GUIDE VANES
MECHANISM FOR CONTROLLING BLADE ANGLE
INNER CYLINDER
IMPELLER
DIFFUSER
Vaneaxial Mechanical Draft Fan
Figure R.1 - Common Terminology for Axial and Tubular Centrifugal Fans 126
AMCA 203-90 (R2007)
HOUSING
DIVERTER CU
TO
FF
CENTER PLATE BLAST AREA DISCHARGE OUTLET AREA SIDE SHEET BACKPLATE
FF
BLADE
TO
CU
INLET
SCROLL IMPELLER FRAME RIM BEARING SUPPORT INLET COLLAR
Figure R.2 - Common Terminology for Centrifugal Fan 127
AMCA 203-90 (R2007)
Figure R.3 - Common Terminology for Centrifugal Fan Appurtenances 128
AMCA 203-90 (R2007)
HEATING AND VENTILATING DRAW-THROUGH UNIT FS
BELT GUARD FS CS
EXT F & BP
MB
FB
INT F & BP
HC
MB
FB
AS
+
+
HEATING AND VENTILATING BLOW-THROUGH UNIT ZONE DAMPERS
FS HC
BYPASS COLD DECK
+ +
HOT DECK
+
+
AIR-CONDITIONING DRAW-THROUGH UNIT FS
AS
MB
FB
CC
HC
MB
FB
ELIM
+
+
+
SS
+
+ +
+
+
DRIP TRAY
AIR-CONDITIONING BLOW-THROUGH UNIT DIFFUSER PLATE
ZONE DAMPERS
FS HC
CC
+
+
+
HOT DECK
+
HC
FB
MB
+
COLD DECK CC
+
+
+
FLEXIBLE CONNECTION AS CS CC HC
ACCESS SECTION COIL SECTION COOLING COIL HEATING COIL
EXT F & BP INT F & BP ELIM
EXTERNAL FACE AND BYPASS DAMPER INTERNAL FACE AND BYPASS DAMPER ELIMINATORS
FS FB MB SS
FAN SECTION FILTER BOX MIXING BOX SPRAY SECTION
Figure R.4 - Common Terminology for Central Station Air-Handling Units 129
AMCA 203-90 (R2007)
Annex S. Typical Format for Field Test Data Sheet
FIELD TEST DATA SHEET JOB DESCRIPTION: Location, User, Contractor, Engineer, . . . . . FAN DESCRIPTION: Mfgr., Size, Type, Ident. No., . . . . . MOTOR DESCRIPTION: Mfgr., Nameplate Data (Ident. No., hp, volts, FLA, . . . ), Performance Data Reference, . . . . . DRIVE DESCRIPTION: Type, Mfgr., Ident. No., Size, . . . . . REFERENCE DRAWINGS OR SKETCHES OF INSTALLATION: System Configuration with Dimensions, Measurement Plane Locations, . . . . . MEASUREMENTS AMBIENT DATA: Barometric Pressure, Dry-Bulb Temp., Wet-Bulb Temp, . . . . . MOTOR DATA: volts, amps, watts, rpm, . . . . . FAN SPEED GAS DENSITY DATA: GAS TEMPERATURES AT MEASUREMENT PLANES:
READING
Ps1 or Ps4
Ps2 or Ps5
Ps3
Pv3
Pv3
1 2 3 4 5
• • • • n TOTAL AVERAGE
CALCULATIONS: (Refer to the various sections of this publication for the appropriate calculation procedures.)
Figure S.1 - Typical Format for Field Test Data Sheet 130
AMCA 203-90 (R2007)
Annex T. Uncertainty Analysis T.1 Introduction In an attempt to determine the range of uncertainties likely to be encountered in field testing of fans, a statistical uncertainty analysis was undertaken. Maximum and minimum uncertainties were assigned to each quantity to be measured based on the degree of difficulty in measuring the quantity, the previously specified accuracies of instruments and the conditions expected to be encountered in field testing. These individual maximum and minimum uncertainties were then combined statistically to arrive at the probable range of overall uncertainties for the fan flow rate, fan static pressure, and fan power input. It would be unlikely, however, that any particular field installation would have all minimum or all maximum uncertainties occurring simultaneously. Therefore, an agreement by the parties as to acceptable measurement tolerances for a given installation should be established prior to testing. In Type A tests, it may be sufficient to accept the results of any field test without consideration of the probable uncertainties in the results. For Type B and Type C tests, it may be necessary to calculate the uncertainties. To do this, each measured quantity is assigned an estimated uncertainty by agreement of the parties involved and the overall uncertainty is calculated as outlined in this annex.
T.2 General This analysis is based on the assumption that fan perfomance can be treated as a statistical quantity and that the performances derived from repeated tests would have a normal distribution. The most probable performance would, therefore, be the mean results based on repeated observations at each point of operation. Only one set of observations is specified in this publication. This analysis deals, therefore, with the probable uncertainty in the results obtained from a single set of observations. The results of a fan field performance test for a single point of operation are a combination of variables which are normally presented graphically. Test results will be considered to be the fan static pressure versus flow rate and fan power input versus flow rate. The uncertainty in results will be expressed in terms of fan flow rate, fan static pressure, and fan power input. The accuracies specified in this publication are based upon two standard deviations. This means that there should be a 95% probability that the actual uncertainties will be less than the specified value.
This applies only to random uncertainties. Systematic uncertainties should be eliminated by the use of properly calibrated test instruments. This analysis considers only the uncertainties inherent in testing. This publication specifies uncertainties in percent. These are, of course, per unit uncertainties, multiplied by 100. Absolute uncertainties which bear the units of the quantity being measured or calculated, are equal to the per unit uncertainty multiplied by the measured or calculated quantity. Since the tolerance on measured values is specified on the basis of 95% confidence limits, the actual deviations in results will be less than the calculated deviations 95% of the time. For the purposes of a field test, an uncertainty range will be defined with minimum and maximum values. This range of possible uncertainty is necessary to cover the varying degrees of difficulty encountered in performing tests in field installations. Field test conditions range from near ideal to near impossible.
T.3 Symbols In the analysis that follows, certain symbols and notations are used in addition to those shown in Annex Q. Symbol
Quantity
ex ΔX R
Per Unit Uncertainty in X Absolute Uncertainty in X Gas Constant (ft-lb/lbm —°R)
Subscript
Description
A b d f g h H N P Q w x ρ
area Barometric Pressure Dry-bulb Temperature Velocity Pressure Static Pressure Power Input Fan Power Input Fan Speed Fan Static Pressure Fan Flow Rate Wet-bulb Depression Generalized Quantity (A, b, ..., ρ) Density
T.4 Measurement uncertainties The various measurement uncertainty ranges used in this publication are listed below. The considerations that led to their adoption include difficulties in field testing generally not encountered in laboratory testing. 131
AMCA 203-90 (R2007) T.4.1 Barometric pressure. The estimated uncertainty in measuring barometric pressure is between 0.3% minimum and 0.7% maximum. eb = 0.003 (min) to 0.007 (max) Barometric pressure is generally obtained by portable aneroid barometer, on-site barometer (mercury or aneroid) or by use of data obtained from a nearby airport. The uncertainty range above is estimated based on the use of portable or on-site instrumentation and applicable corrections. T.4.2 Dry-bulb temperature. The estimated uncertainty in measuring dry-bulb temperature is between 0.5% of absolute temperature minimum and 2.0% of absolute temperature maximum. ed = 0.005 (min) to 0.02 (max) The estimated uncertainty range is based on a broad temeprature range and the likelihood of stratification. T.4.3 Web-bulb depression. The estimated uncertainty in measuring wet-bulb depression is between 5°F minimum and 10°F maximum. ew = 5/(td - tw) (min) to 10/(td - tw) (max) The estimated uncertainty range is based on a broad temperature range with the associated difficulties in determining wet-bulb readings at high or low temperatures and the likelihood of stratification. T.4.4 Fan speed. The estimated uncertainty in measuring fan speed is between 0.5% minimum and 1.0% maximum. eN = 0.005 (min) to 0.01 (max) The uncertainty range in fan speed is estimated on the basis of portable instrumentation accuracy and an allowance for fluctuation in fan speed. T.4.5 Power input. The estimated uncertainty in measuring power input is betwen 3.0% minimum and 7.0% maximum. eh = 0.03 (min) to 0.07 (max) The estimated uncertainty range is based on the various measurement methods and their respective accuracies, estimated drive losses, and the broad horsepower range encountered in the field. T.4.6 Pitot traverse. A properly performed field traverse is estimated to have an accuracy of 1.5% minimum to 7.5% maximum. 132
ec = 0.015 (min) to 0.075 (max) The uncertainty range in the Pitot traverse is estimated on the basis of traverse location, broad range of duct sizes, nonuniform velocity profiles, and turbulence. T.4.7 Flow measurement area. The estimated uncertainty in the flow measurement area is between 1.0% minimum to 2.0% maximum. eA = 0.010 (min) to 0.020 (max) The estimated uncertainty is based on a broad range of duct sizes, accessibility, and the rigidity of ducts under pressure. T.4.8 Velocity pressure. An allowance of 2.0% minimum to 5.0% maximum of the reading is estimated for the mental averaging performed on a fluctuating reading. An allowance of 1.0% minimum to 2.0% maximum of the reading is estimated for calibrated manometer uncertainty and relocation of the instrument after calibration. In addition, an allowance of 0.5% minimum to 10.0% maximum of the reading is estimated for instrument precision. No allowance is included for yaw on the assumption that the Pitot-static tube is aligned within 10 degrees of streamlines. A combined uncertainty can be written as: ef (min) = [(0.02)2 + (0.01)2 + (0.005)2]0.5 = 0.0229 ef (max) = [(0.05)2 + (0.02)2 + (0.10)2]0.5 = 0.1136 T.4.9 Static pressure. An allowance of 1.0% minimum to 5.0% maximum of the reading is estimated for the mental averaging performed on a fluctuating reading. An allowance of 1.0% minimum to 2.0% maximum of the reading is estimated for calibrated manometer uncertainty and relocation of the instrument after . In addition, a tolerance of 10% minimum to 20.0% maximum of the fan velocity pressure should cover the influence of Pitot-static tube yaw or velocity influence on static pressure taps and other possible effects. A combined uncertainty can be written as: eg (min) = {(0.01)2 + (0.01)2 + (0.005)2 + [0.1 Pv/(Ps2 - Ps1)]2}0.5 = {0.000225 + [0.1 Pv/(Ps2 - Ps1)]2}0.5 eg (max) = {(0.05)2 + (0.02)2 + (0.02)2 + [0.2 Pv/(Ps2 - Ps1)]2}0.5 = {0.0033 + [0.2 Pv/(Ps2 - Ps1)]2}0.5
AMCA 203-90 (R2007) Where the denominator in the final term in each equation will involve Ps2 or Ps5 and Ps1 or Ps4, whichever are measured.
Assuming Δ70.73 and ΔR are both zero:
The estimated uncertainty range is based on an allowance for fluctuation in the fan-system operation, lack of ideal measurement locations, turbulence, and the relocation of instrumentation after calibration.
It can be shown that: ev2 = [(0.00000725 tw - 0.0000542) Δ(td - tw)]2 Where:
T.5 Combined uncertainties The uncertainties in the test performance are the result of using various values, each of which contains a probable uncertainty. The combined uncertainty for each of the fan performance variables is given below. T.5.1 Density. Air density involves the various psychrometric measurements and the approximate formula:
ρ=
eρ = (eb2 + ev2 + ed2)0.5
70.73 pbV R ( t d + 460 )
Δ(td - tw) = Absolute uncertainty in wet-bulb depression. Other methods for determining density are assumed to have equal accuracy. T.5.2 Fan flow rate. Fan flow rate directly involves the area at the flow measuring station, the Pitot traverse, the square root of the pressure measurement for flow, and the square root of the density. Uncertainties in fan speed will produce a first-power uncertainty in flow rate when making the fan law conversions. Combining: eQ = [ec2 + eA2 (ef/2)2 + (eρ/2)2 + eN2]0.5
Where: V = 1.0 - 0.378 {(pe/pb) - [(td - tw)/2700]} For random and independant uncertainties in products, the combined uncertainty is determined as follows: Δρ/ρ = {(Δ70.73/70.73)2 + (Δpb/pb)2 + (ΔV/V)2 + (ΔR/R)2 + [Δtd/(td + 460)]2}0.5
T.5.3 Fan static pressure. Fan static pressure directly involves static pressure measurements. Uncertainties in density will produce a first-power uncertainty in fan static pressure while uncertainties in fan speed will produce a second-power uncertainty in fan static pressure when making fan law conversions. Combining: ep = [eg2 + eρ2 + (2eN)2]0.5
Table T.1 Measurement eb ed** eW eN eh ec eA ef eg
Minimum
Maximum
0.003 0.005 5/(td - tw) 0.005 0.030 0.015 0.010 0.0229 {0.000225 + [0.1 Pv/(Ps2 - Ps1)]2}0.5
0.007 0.020 10/(td - tw) 0.010 0.070 0.075 0.020 0.1136 {0.0033 + [0.2 Pv/(Ps2 - Ps1)]2}0.5
* These uncertainties do not account for the effect of swirl at the fan inlet. This situation must be corrected in order to produce acceptable fan-system performance (see Section 5). ** Based on absolute temperature 133
AMCA 203-90 (R2007) In order to simplify the application of this uncertainty analysis to the results of field tests, the above equation was developed on the basis of tests in which static pressure measurements are made at a single plane, as would be the case in which a fan is ducted on one side only. However, the equation is reasonably accurate for all other fan-system configurations.
The uncertainty calculations lead to absolute uncertainties in fan flow rate, fan static pressure, and fan power input that can be applied directly to the corresponding test results. The uncertainty results can then be plotted as rectangles around the test point. Intersection of the rectangles with the quoted fan performance within the limitations of a field test. See the examples in Section T.7.
Although in most cases the determination of fan static pressure involves Pv1, the uncertainty in determining Pv1 is not included in the above equation on the basis that it normally has a very small effect on the overall uncertainty in fan static pressure.
T.7 Examples
For purposes of this publication, eP is applied directly to Psc, which may include System Effect Factors. T.5.4 Fan power input. Fan power input directly involves the power measurement; in addition, when making fan law conversions, density has a first-power effect and speed has a third-power effect on fan power input. Combining: eH = [eh2 + eρ2 + (3eN)2]0.5
T.6 Summary The minimum and maximum measurement uncertainties (See Table T.1) were defined earlier in Section T.4. Summarizing, the per unit uncertainties are as shown in Table T.1.
134
Two examples of the calculation of uncertainties and the method of comparison with the quoted fan curve are included in this section. Uncertainty calculations and comparisons have been developed for Examples 2B and 2C of Annex A. Uncertainty calculations for Example 2B utilize all minimum uncertainty tolerances. Uncertainty calculations for Example 2C utilize all maximum uncertainty tolerances. It would be unlikely that any field installation would lend itself to all minimum or all maximum measurement tolerances. Agreement of the parties as to acceptable measurement tolerances for a given installation should be established prior to testing.
AMCA 203-90 (R2007) EXAMPLE 1: CALCULATION OF UNCERTAINTIES IN TEST RESULTS BASED ON MINIMUM MEASUREMENT UNCERTAINTY TEST VALUES Reference: Example 2B in Annex A SITE MEASUREMENTS td2 = tw2 = Ps1 = Ps2 = Pv3 = A2 = A3 = ρ2 = ρ3 =
91.3°F 70.4°F -11.4 in. wg 0.1 in. wg 1.24 in. wg 1.40 ft2 1.57 ft2 0.0714 lbm/ft3 0.0705 lbm/ft3
CONVERTED RESULTS Qc = 7114 cfm Psc = 11.42 in. wg Hc = 18.90 hp MEASUREMENT UNCERTAINTIES Reference: Minimum values per Section T.6 eb ed ew eN eh ec eA ef eg
= = = = = = = = =
0.003 0.005 5/(td2 - tw2) 0.005 0.030 0.015 0.010 0.0229 {0.000225 + [0.1 Pv/(Ps2 - Ps1)]2}0.5 CALCULATIONS
Pv = = = =
Pv2 Pv3 (A3/A2)2 (ρ3/ρ2) 1.24 (1.57/1.40)2 (0.0705/0.0714) 1.54 in. wg
eg = {0.000225 + [0.1 Pv/(Ps2 - Ps1)]2}0.5 = {0.000225 + [(0.1 × 1.54)/(0.1 + 11.4)]2}0.5 = 0.02011
ev2 = [(0.00000725 tw - 0.0000542) Δ(td - tw)]2 = [(0.00000725 × 70.4 - 0.0000542) 5]2 = 0.00000520 eρ = [eb2 + ev2 + ed2)0.5 = (0.0032 + 0.00000520 + 0.0052)0.5 = 0.006261 eP = [eg2 + eρ2 + (2eN)2]0.5 = [0.020112 + 0.0062612 + (2 × 0.005)2]0.5 = 0.0233 eQ = [ec2 + eA2 + (ef/2)2 + (eρ/2)2 + eN2]0.5 = [0.0152 + 0.0102 + (0.0229/2)2 + (0.006261/2)2 + 0.0052]0.5 = 0.0222 eH = [eh2 + eρ2 + (3eN)2]0.5 = [0.0302 + 0.0062612 + (3 × 0.005)2]0.5 = 0.0341 ΔP = ePPsc = 0.0233 × 11.42 = 0.27 in. wg Psc + ΔP = 11.42 + 0.27 = 11.69 in. wg Psc - ΔP = 11.42 - 0.27 = 11.15 in. wg ΔQ = eQQc = 0.0222 × 7114 = 158 cfm Qc + ΔQ = 7114 + 158 = 7272 cfm Qc - ΔQ = 7114 - 158 = 6956 cfm ΔH = eHHc = 0.0341 × 18.90 = 0.64 hp Hc + ΔH = 18.90 + 0.64 = 19.54 hp Hc - ΔH = 18.90 - 0.64 = 18.26 hp
135
AMCA 203-90 (R2007)
GRAPHICAL PRESENTATION
Psc
Psc + ∆P
TEST POINT MINIMUM UNCERTAINTY RANGE
Ps, FAN STATIC PRESSURE
Psc - ∆P
Qc = 7114 cfm ΔQ = 158 cfm Psc = 11.42 in. wg ΔP = 0.27 in. wg Hc = 18.90 hp ΔH = 0.64 hp
Qc + ∆Q
Qc - ∆Q Qc
QUOTED FAN PERFORMANCE CURVES
Q, FAN FLOW RATE
H, FAN POWER INPUT
Hc + ∆H Hc Hc - ∆H
Qc + ∆Q
Qc - ∆Q
Qc Q, FAN FLOW RATE
Figure T.1
136
AMCA 203-90 (R2007) EXAMPLE 2: CALCULATION OF UNCERTAINTIES IN TEST RESULTS BASED ON MAXIMUM MEASUREMENT UNCERTAINTIES TEST VALUES Reference: Example 2C in Annex A SITE MEASUREMENTS
eQ = [ec2 + eA2 + (ef/2)2 + (eρ/2)2 + eN2]0.5 = [0.0752 + 0.0202 + (0.1136/2)2 + (0.02176/2)2 + 0.0102]0.5 = 0.0973 eH = [eh2 + eρ2 + (3eN)2]0.5 = [0.0702 + 0.021762 + (3 × 0.010)2]0.5 = 0.0792
td3 = 86.5°F tw3 = 75.5°F Ps4 = -1.57 in. wg Ps5 = 1.22 in. wg Pv2 = 0.61 in. wg
ΔP = eP Psc = 0.0780 × 2.54 = 0.20 in. wg
CONVERTED RESULTS
Psc - ΔP = 2.54 - 0.20 = 2.34 in. wg
Psc + ΔP = 2.54 + 0.20 = 2.74 in. wg
Qc = 25964 cfm Psc = 2.54 in. wg Hc = 17.11 hp MEASUREMENT UNCERTAINTIES Reference: Maximum values per Section T.6 eb ed eW eN eh ec eA ef eg
= = = = = = = = =
0.007 0.020 10/(td3 - tw3) 0.010 0.070 0.075 0.020 0.1136 {0.0033 + [0.2 Pv/(Ps5 - Ps4)]2}0.5 CALCULATIONS
ΔQ = eQQc = 0.0973 × 25964 = 2526 cfm Qc + ΔQ = 25964 + 2526 = 28490 cfm Qc - ΔQ = 25964 - 2526 = 23438 cfm ΔH = eHHc = 0.0792 × 17.11 = 1.36 hp Hc + ΔH = 17.11 + 1.36 = 18.47 hp Hc - ΔH = 17.11 - 1.36 = 15.75 hp
eg = {0.0033 + [0.2 Pv/(Ps5 - Ps4)]2}0.5 = {0.0033 + [(0.2 × 0.61)/(1.22 + 1.57)]2}0.5 = 0.07219 ev2 = [(0.00000725 tw - 0.0000542) Δ(td - tw)]2 = [(0.00000725 × 75.5 - 0.0000542) 10]2 = 0.0000243 eρ = (eb2 + ev2 + ed2)0.5 = (0.0072 + 0.0000243 + 0.0202)0.5 = 0.02176 eP = [eg2 + eρ2 + (2eN)2]0.5 = [0.072192 + 0.021762 + (2 × 0.010)2]0.5 = 0.0780
137
AMCA 203-90 (R2007)
GRAPHICAL PRESENTATION
TEST POINT MAXIMUM UNCERTAINTY RANGE Qc = 25964 cfm ΔQ = 2526 cfm Psc + ∆P
Psc = 2.54 in. wg ΔP = 0.20 in. wg
Ps, FAN STATIC PRESSURE
Psc
Hc = 17.11 hp ΔH = 1.36 hp
Psc - ∆P
Qc - ∆Q
Qc + ∆Q
Qsc
H, FAN POWER INPUT
Q, FAN FLOW RATE
QUOTED FAN PERFORMANCE CURVES
Hc + ∆H Hc Hc - ∆H Qc - ∆Q
Qc + ∆Q Qsc
Q, FAN FLOW RATE
Figure T.2
138
AIR MOVEMENT AND CONTROL ASSOCIATION INTERNATIONAL, INC. 30 West University Drive Arlington Heights, IL 60004-1893 U.S.A.
Tel: (847) 394-0150 E-Mail :
[email protected]
Fax: (847) 253-0088 Web: www.amca.org
The Air Movement and control Association International, Inc. is a not-for-profit international association of the world’s manufacturers of related air system equipment primarily, but limited to: fans, louvers, dampers, air curtains, airflow measurement stations, acoustic attenuators, and other air system components for the industrial, commercial and residential markets.