k fD-A±3U 253 MODIFICATION OF CABINET FANS WITH INLET AIR GUIDE I/iFAIRINGS TO IMPROVE PERFORMANCE(U) CONSTRUCTIONENGINEERING RESEARCH LAB (ARMY) CHAMPAIGN IL W H DOLANANLSIFE PR 83 CERL-TR-E-i~i F/G 13/1i N
EIND
IllIflI2.8 25
36
1111.8
1.4 L16
MICROCOPY RESOLUTION TEST CHARTNATIONAL BUREAU OF STANDARDS-1963-A
-A
construction nUftmAV
eninerngTECHNICAL REPORTE18research April 1983
(Energy Conservative Operation of
laboratory Existing Buildings and Facilities)
MODIFICATION OF CABINET FANS WITH INLET* n AIR GUIDE FAIRINGS TO IMPROVE PERFORMANCE
byWilliam H. Dolan
DTIC
MAW
849 7 12 061Approved for public release; distribution unlimited.
The contents of this report are not to be used for advertising, publication, ofpromotional purposes. Citation of trade names does not constitute anofficial indorsement or approval of the use of such commercial products.The findings of this report are not to be construed as an official Departmentof the Army position, unless so designated by other authorized documents.
J-7
HI
DESTROY THIS REPORT WHEN ITIs NO I.ONGER NEEDED
DO NOT RETURN IT TO THE ORIGIN.4 TOR
UNCLASSIFIEDSECURITY CLASSIFICATION OF THIS PAGE (When. Date Entered)
PAGE READ INSTRUCTIONSREPORT DOCUMENTATION PAEBEFORE COMPLETING FORMREPORTNUMBEROVT ACCESSION NO. 3.RCPINS CATALOG NUMBER
* ~CERL-TR-E-181A A 13_______________4. TITLE (and Subtitle) 5.TYPE OF REPORT & PERIOD COVERED
MODIFICATION OF CABINET FANS WITH INLET FINALAIR GUIDE FAIRINGS TO IMPROVE PERFORMANCE
S. PERFORMING ORG. REPORT NUMBER
*7. AUTHOR(a) S. CONTRACT OR GRANT NUMBER(&)
William H. Dolan
9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM ELEMENT PROJECT.'TASKU.S.ARMYAREA & WORK UNIT NUMBERS
CONSTRUCTION ENGINEERING RESEARCH LABORATORY 4728A4--0* P.O. BOX 4005, CHAMPAIGN, IL 61820
Ii. CONTROLLING OFFICE NAME AND ADDRESS 1.REPORT DATEp
April 198313. NUMBER OF PAGES
_______________________________________________ 15*14. MONITORING AGENCY NAME A AOORESS('lf different from Controliini Office) is. SECURITY CLASS. (of thi. report)
Unclassifiled
So. CSSIFICATION/DWNGRADINSCHEDULE
* 1S. DISTRIBUTION STATEMENT (of thle Report)
Approved for public release; distribution unlimited.
17. DISTRIBUTION STATEMENT (of the abstract enteed in Block 20. If Effafrent *oot Report)
* IS. SUPPLEMENTARY NOTES
* Copies are available from the National Technical Information ServiceSpringfield, VA 22161
IS. KEY WORDS (Continue an revere* side if noceesar and identify by block nimaher)
fansenergy consumption
2L ANSTRASI' (Cthwe - reveleef N noeemm identity by Weoek mmhee)
Cabinet fans are commonly used for central station air-handlers in commercial sizeheating, ventilating, and air-conditioning (UVAC) systems. This report describes theconception, construction, and testing of a device to improve the efficiency of cabinetfans by improving the fan inlet conditions. By observing airflow within the cabinet, a
DO ,~~F 10 EDIIONOF NOV6 ISOSOETEUNCLASS IFIEDSACUSSTV CLASSIFICATION OF THIS PAGE (11111 Deis Entered)
UINCLAS SIFILEDSECURITY CLASSIFICATION OF THIS PAGE(Wb.n Dase Matersd)
BLOCK 20 (CONT'D)
region of streamline flow and the boundary of flow separation were outlined. Fan ef-ficiency was improved by enhancing streamline flow and eliminating flow separation
* within the cabinet by installing a fiberglass fairing which guided inlet air
A fan was tested with and without the faiiing in place; an overall 20 percent im--
provement in fan efficiency was observed with the fairing in place.
IINCLASSIFTEDSECURITY CLASSIFICATION OF THIS PAGEfthm Date Etntered)
FOREWORD
This work was performed for the Assistant Chief of Engineers under Project
4A762781AT45, "Design, Construction and Operation and Maintenance Technology
for Military Engineers"; Task B, "Installation Energy Conservation Strategy"; Work
Unit 004, "Energy Conservative Operation of Existing Buildings and Facilities." Mr. '4Bernard Wasserman, DAEN-ZCF-U, was the Technical Monitor.
The work was performed by the Energy Systems (ES) Division of the U.S. Army
Construction Engineering Research Laboratory (CERL). Mr. R. G. Donaghy is Chief ofCERL-ES. Appreciation is expressed to Mr. Richard Rundus and Mr. Victor Storm of
CERL for their contributions to this work.
COL Louis 1. Circeo is Commander and Director of CERL, and Dr. L. R. Shaffer is
Technical Director.
11, -
T00
is .. 'a*
3
- - -. . . °
CONTENTS
Page
DO FORM 1473FOREWORD 3LIST OF TABLES AND FIGURES 5
IINTRODUCTION ............................................. 7BackgrondObiective
Mode of Technogyw Transfer
2 DISCUSSION................................................7GeneralExperimental DesignFan Modification and Ductwork SystemFlow VisualizationFairing Construction and DesignStatic PressureTemperature MeasurementPower/Torque MesurementAirflow MeasurementTestingEfficiency Measurement
3 TEST RESULTS.............................................13
4 CONCLUSION .............................................. 15
METRIC CONVERSION CHART
DISTRIBUTION
4
FIGURES
Number Page
I Double Width Double Inlet (DWDI) Centrifugal Fan 8
2 Standard Configuration Used for Rating Centrifugal Fans 9.
3 Experimental Facility 10
4 Torque Measurement 10
5 Flow Within Cabinet I
6 Fairing 1
TABLES
1 Summary of Fan Testing 14
* -2 Fan Testing Data 14
MODIFICATION OF CABINET FANS ApproachWITH INLET AIR GUIDE FAIRINGS TO 1. Procure and install a cabinet fan with an ap-IMPROVE PERFORMANCE propriate drive and duct network and operate it over
a typical range.
INTRODUCTION 2. Observe the inlet airflow within the cabinet byusing various flow visualization techniques.
BackgroundThe fans within the central station air handlers that 3. Based on the inlet flow patterns observed in
help heat, ventilate, and cool Army buildings use Step 2, fabricate a fairing to be positioned within
substantial amounts of electrical energy. The Army, the cabinet.as well as the private sector, operates many air-handlerfans that often consume more energy over a single 4. Procure and install instrumentation to generateheating or cooling season than all other HVAC corn- fan performance curves and fan efficiency.
ponents combined.5. Operate the test fan with and without the fairing
The fan manufacturing industry markets centri- in place to determine the effective increase in fan- fugal fans as efficient as 80 percent; however, it is efficiency.
common for fans with central air handlers to operateat static efficiencies of 50 percent or less. Several 6. If results from Step 5 are positive, determine andfactors contribute to this poor performance: acquire fans of the most common type used in Army
construction.1. Fan operation outside the region of efficient
performance. 7. Develop a mathematical expression to describethe fairing shape and determine an acceptable means
2. Nonoptimal outlet conditions. of manufacturing.
3. Nonuniform inlet flow. 8. Substantiate the improvement in fan efficiency
and fan performance on the fans acquired in Step 6.S l tfieoutfitted with fairings fabricated according to Step 7.
To find ways to reduce fan electrical consumption. by improving fan efficiency, the U.S. Army Con- 9. Develop design guidance and specifications for
struction Engineering Research Laboratory (CERL) fan cabinet fairings.
examined these causes of poor fan performance.. CERL determined that although replacing existing This report describes Steps 1 through 5, above.
fans and rebuilding central station air handlers wouldaddress the first two factors, these options were too Mode of Technology Transfercostly and usually impossible considering the existing It is recommended that the results of this study bespace in most Army buildings. However, improving the incorporated into an Engineering Technical Note oninlet conditions using a fan fairing (the third and retrofitting cabinet fans with inlet air guide fairings.fourth problem areas) could improve fan operatingefficiency without requiring any physical modificationof the fan cabinet or air handler. Presently, cabinet fan 2 DS SImanufacturers do not market such devices. DISCUSSION
ObjectiveThe overall objective of this study was to (1) General
determine if inserting an inlet fairing within a fan Most fans used for HVAC applications are thecabinet could improve fan inlet conditions and fan centrifugal type (Figure 1). A typical built-up airefficiency and (2) develop design guidance and speci- handler for a multizone system has three sections: --2.
* :fications for an inlet fairing.I. A mixing box which accepts either of two
This report describes the results of investigation airstreams, or mixtures of each (usually outdoor air(I), above. and return air).
7 F,.vousA.F i7$ __.
DISCHARGE
PLATE CONNECTING FAN WHEEL TO FAN SHAFT.
ROTATING FAN WHEEL
Fig.,e 1. Double Width, Double Inlet (DWDI) centrifugal fan.
2. A fan drawing air from the mixing box. Practical considerations of space and expenseprohibit excessively large fan cabiets in air handlers.
3. C'ils into which the fan discharges and over Thus, the inlet flow is not free and unobstructed andwhich air passes (heating and cooling). I9ng diverging outlets which convert high-velocity
discharge air efficiently to static pressure are notTo control the air moved by the air-handler fan, the possible. In practice, operating a fan within an air-
fan is contained within a cabinet which, in the case of handler enclosure results in a substantial drop in fana multizone air handler, connects to the mixing box efficiencies as compared with the nearly ideal con-and coils. ditions under which the fan is tested.
Operating a centrifugal fan in this arrangement is Uterature from the Air Moving and Conditioningfar from ideal. Although this category of fan can Association (AMCA) specifically states that threeoperate at static efficiencies of 80 percent, optimum common causes of poor fan performances are (1) im-operating conditions are needed to produce this proper outlet conditions, (2) nonuniform inlet flow,high efficiency. and (3) swirl at the fan inlet.' Fans within built-up
air handlers are subject to those three problems. TheThe standard test conditions for testing centrifugal AMCA guides show the losses resulting from poor fan
fans are: inlet conditions can be equivalent to a system staticpressure drop of 0.25 in. water column (wc). For
1. A free, unobstructed inlet. example, a system with a total static pressure drop of2 in. wc would realize a reduction of up to 12 percent
2. A long, slowly diverging duct section at the fan in the power imparted to the airstream and a similaroutlet ending with a uniform register which will not savings in motor power if a fairing in the fan cabinetcause an air swirl (Figure 2). It should also be noted could reduce or eliminate this 0.25-in. wc loss.that the fan must be rotated within a certain speedrange and the static pressure rise developed across thefan must also be within a certain range, depending on tFmr and Systems, Publication 201 (Ar Moving and
the characteristics of the fan (wheel diameter, blade Colditionifl Association).
type, scroll design). *Metric conversion table is on p 15.
8.s m -O.i,
UN-OSTRUTED~. I!SYMMETRICALUN-OBSRUCTN PiROTTLING
INLET XE DEVICEK.....STRAIGHTENER9•~L "TUBEA !
TRAVERSEFAN
Figure 2. Standard configuration used for rating centrifugal fans.
The purpose of a fairing positioned within the fan 4. A fairing.cabinet is to direct the airstream entering the fan in alaminar manner. Without a fairing, airflow within a fan A double width, double inlet (DWDI) backwardcabinet is laminar in a region around the center of the incline-blade cabinet fan was available at CERL.fan inlet and nonlaminar or separated in the comers Instrumentation was acquired to monitor the fol-and center of the cabinet, creating stagnant regions. lowing parameters: fan speed, motor speed, motor
torque, air temperature rise, static pressure rise, andCERL reviewed several approaches to a fairing total flow. CERL also designed and built an air duct
design. One approach was to analytically predict the system, including dampers. The fiberglass fairing, alsoflow contours which could occur without separation; designed by CERL, was fabricated on a foam mold.basically, the inlet airstream has to accelerate aroundthe fan scroll and turn 90 degrees to enter the fan.
Fan Modification and Ductwork System"Another approach was to study and map the flow Figure 3 shows the apparatus used for this experi-
patterns occurring in the fan enclosure without a fair- ment. The first section is the horizontal ducting fit toing, and then base the fairing design on the observed the fan inlet to keep the flow entering the fan one-shape of that part of the flow stream that did not directional and free from the swirl which sometimesseparate. This latter approach was adopted because: results when a vertical duct section is fixed to the inlet
of a horizontal fan.1. The cabinet fan used for this study had to be
modified. Consequently, the rotative speed and flow The second section is the specially modified DWDIrate associated with the range of best efficiency was cabinet fan. Because one of the two fan inlets wasnot certain and so had to be determined experimen- obstructed by bearings, scroll supports, and frametally. Analytical prediction of optimal fan performance members, the fan was capped, which resulted in a no-for a modified fan plus flow patterns was considered flow stall condition in one-half of the fan. (Test resultsto overcomplicate the study objective, proved this was not to be a problem; see Chapter 3.)
The active half of the fan performed in a manner2. Flow separation is a relatively simple concept consistent with the fan laws.
which can be observed using a variety of techniques.The fan was driven by a 5-hp motor powered by a
Experimental Deign variable frequency power supply. Fan and motorThe equipment listed below was used to evaluate speed were regulated by a potentiometer located near
the merit of enhancing flow characteristics by inserting the motor. Torque developed by the fan motor wasa contoured fairing in a cabinet fan: measured by a balancing-type scale (Figure 4).
1. A cabinet fan ' it could be f*, 4th a fairing. The final section of the test facility was the outletsection which:
2. Instrumentation aluate the fan performance.1. Spanned about 15 equivalent diameters after the
3. An air duct system capable of adjusting the total fan discharge (intended to ensure a uniform outletsystem resistance. flow profile).
9
.- . . .
VARIABLE SPEEDPOWER SUPPLY
MOTOR /F,".,
SCALE A07TEMPERATR
-- -- 15 EQU. DIAMETERS /
/I FLOW MEASURES STATIONLINLET DUCT SECTION
Figure 3. Experimental facility.
WEIGHTS TOVARY BELTTENSION
16 17/32
Fipre 4. Torque measurement.
2. Incorporated a pitot tube rack to sample the air 3. Yam sampling by a single traversing probe.velocity at 24 points toniformly spaced across the ductcross section. Although laser anemometry is a sophisticated,
noninterfering method, it was rejected because it is3. Discharged through two slide dampers located overly complex and expensive, and had a resolution
at the end of the ductwork. These dampers were used greater than the needs of this study. Smoke generators,to control the airflow rate by varying the overall both single high-output source and single-traversingresistance of the duct section. probe, successfully identified the overall characteristics
of the airflow as well as the boundaries of streamlineFlow Visualization and turbulent flow.
The methods considered for observing the airflowwithin the fan cabinet were:
The high-output smoke source was bottled carbonI. Particulate tracing with laser anemometry. dioxide with a moderately convergent nozzle which
produced condensate or clouds in the airstream.2. Smoke generation by (a) fixed multiyle-point Pictures and notes were recorded of the streamline
sources, (b) a traversing point source, and (c) a single flow pattern observed through plexiglass windowshigh-output source. on the fan cabinet.
10
The single traversing probe was made from i18-in.OD copper tuLAng about 3 ft long; it output smoke atthc tip. The probe's small diameter had no appreciableeffect on the flow and clearly identified the boundaryof streamlined and separated flow.
The point-source traversing smoke generator, plus DISCHARGEsmall pieces of yarn suspended from a thin probe,were used to map the shape of the unseparated stream- ___
line flow entering the fan. ./ //
Fairing Construction and Design / FANThe fairing shape was based on the results of the .
flow visualization tests and elementary principles of // /low-velocity airflow; i.e., the duct design of smooth I /uniform sections and gradual turns to impede flow I I "separation. Figure 5 is a two-dimensional representa- I, ,tion of airflow within the cabinet. Layered airflow I \ "
from the free stream into the fan inlet described anarc closely approximated by a circle with tangentsat the left border of the fan cabinet and at the rear %
of the fan inlet. Figure S. Flow within cabinet.
The fairing's physical shape was defined by the arc fan shaft was cut during assembly; a slice to the holeon the converging section. The inlet was flared to form as well as the hole were cut during installation with aa rectangle the same size of the cabinet inlet (Figure 6). low-power reciprocating saw.
The fairing was made of fiberglass formed over a Static Pressuremold. Fiberglass proved to be well suited to this Static pressure was sampled at the fan dischargeapplication; it has excellent strength, flexibility which using several static port pitot tubes. Static pressureeased the fairing installation, and a smooth interior at the fan inlet was assumed to be at atmosphericfinish. The mold was carved from a block of rigid pressure because of the low velocities in the 5-ftpolyurethane foam, then painted with multiple coats section of ducting on the inlet. The static pressureof epoxy paint and waxed. The fiberglass was placed was observed on a manometer with a finely graduatedover the mold, allowed to harden, then separated from rule.the mold with 120 psi air.
Temperature Measurement
The fairing did not have to be fastened in place It was of interest to measure the rise in temperaturebecause it fit snugly into the cabinet. A hole for the experienced by the airstream as it passed through theto
IgIP.
! " %\ %
HOLE FOR SHAFTFipre 6. Fairing.
II
fan. This temperature rise can be used to calculate fan Testingefficiency (temperature rise from ideal compvession Fan performance was recorded with and withoutdivided by actual temperature rise). Two temperature- the fairing in place by generating standard fan curvesmeasuring methods were tried: at various rotative speeds (static pressure vs flow for
a fixed rotative speed). Air temperature rise across the1. Four-wire platinum probes were used along with fan and motor torque were also recorded. Thus, fan
a Ilewlett-Packard 3052A data acquisition system. For efficiency could be calculated by two means: - -
the purpose of a differential temperature measurement.these probes displayed excellent accuracy,. (le probes 1. Power imparted to the air divided by motorwere matched within an adiabatic environment of shaft power.O.05°F agreement.) Ilowever. the fast response of theseprobes sensed minor fluctuations occurring in the inlet 2. Temperature rise associated with ideal com-airstream, which meant minor temperature fluctuations pression divided by actual temperature rise.had to be averaged out in the inlet airstream.
Efficiency Measurement
2. High-accuracy glass bulb thermometers (gradua- If the process of the compression of air by the testtions to 0.18 0 F) were matched within an adiabatic fan is assumed to occur at constant density, a simpleenvironment to 0.09'F agreement. The relatively slow expression for fan efficiency as a function of pressureresponse of the glass bulb thermometers tended to rise and temperature rise can be derived. For ideal gasaverage out the minor temperature variations observed compression, the density will increase according to:in the inlet airstream by the four-wire platinum probes.
The glass bulb thermometers were selected for use P 4P "[Eqfor the remainder of the study. P kP2 /
Power/Torque Measurement where:
Fan shaft power was determined by measuring therotative speeds of the fan and motor shafts and the P = absolute pressuretorque developed by the motor. Although this methoddid not account for the belt drive losses, those losses P = densitywere believed to be negligible. They also would be thesame for each experimental condition. k = gas constant.
Motor torque was measured by recording a force However, the actual process of compression by a fantransmitted by a rigid member to a balance scale with will involve additional heat added to the gas because ofa contact point 17.07 in. from the motor shaft center. fan inefficiencies, which will cause a temperature rise
and corresponding reduction in density according -..
The rotative speeds of the fan and motor were to Eq 2:measured using a digital photo tachometer accurateto within 1 rpm. Power was calculated as the product T2 PL [Eq 21of torque and rotative speed. T, p2
Airflow Measurement where T = absolute temperature.An airflow measuring station manufactured by
Cambridge Filter Corporation was used to record Conveniently, the process occurring in the test fanflow. This device uses 36 pitot probes evenly posi- having an approximate efficiency of 35 percent resultstioned across a I x I ft cross section. Stagnation and in the effects of ideal compression and constantstatic pressures are transmitted by averaging manifolds, pressure heating on gas density negating one another.The measuring station was located 15 equivalent This is evident by examining the test fan at two opera.diameters downstream of the fan discharge to ensure ting points in the range of interest of this study:a uniform flow profile. The manometer supplied by
* the manufacturer was calibrated in units of flow 1. A relatively high compression of 2.38 in. wcintended specifically for the flow station. where a temperature rise of 2.36°F was observed.
12
- - . . - .
2. A relatively low compression of 1 in. wc where An expression for an efficiency is formed by com-a temperature rise of 1.62 0 F was observed. bining Eqs 5 and 6:
Using ideal gas laws n, efficiency power imparted the airshaft power
p, =Pl T2 -"P2 P2T [Eq3] 0.00015"73 cfm AP .-
n= 0.000402 cfm ATand laboratory conditions of absolute pressure (29.21
in. Hg), the resultant change in P2for Case 1 was 0.999 n = 0.391 - [Eq 7]
and Case 2 was 1.000. Thus, over the operating rangeof interest of the test fan, the change in density of the where:air was negligible.
AP = pressure rise (in. wc)Continuing in the derivation of an expression
for fan efficiency, Eq 4 defines the rate at which AT = temperature rise *Fenergy is delivered by the fan shaft and realized by atemperature rise in the airstream (i.e.. change in To verify Eq 7, the case of an ideal fan having anenthalpy of the airstream): efficiency of I was examined. A relationship of AT =
0.39lAP should exist. From ideal gas laws, the re-P = (cfm)(PXcpXAT) [Eq 4J lationship of pressure and temperature is as follows:
where: T2 PA
P = rate of shaft work T, Pl J Eq8 ]
Using the laboratory conditions of T, = 535°R*,cfm = volumetric flow rate P1 = 397 in. wc, and k for air = 1.4, a change in P of
I in. wc (P 2 = 398 in. wc) produces a change in tem-p = density perature of 0.386*F, which is in excellent agreement
with the expression stated by Eq 7 derived indepen-AT = temperature rise dent of ideal gas laws.
cp = specific heat of the gas In conclusion, excellent confidence is held in themeasurement of fan efficiency using the observed
Using a density for air of 0.071 lbm/cu ft (based on temperature and pressure rise and assuming a constantabsolute pressure of 29.21 in. Hg, 750F, and 65 per- density process.cent relative humidity) and a specific heat of 0.024Btu/ibm°F, Eq 4 simplifies to
P = 0.000402 cfm AT [Eq 51 3 TEST RESULTS
where P = power (hp).
1. Fan performance improved with the inlet airEq 6 is a hydraulic definition for power imparted to guide fairing in place. For a fixed rotative speed and
a fluid as a function of volumetric flow rate and fixed airflow rate, the fan developed an additional
pressure rise. static pressure rise with the inclusion of the fairing.P 0.0001573cfmAP This indicates an increase in the fan capacity. More
PfEq6] significantly, the fan efficiency improved and the
region of efficient operation increased. This wouldwhere: P = power (hp)
AP = pressure rise (in. wc) It= "F+460.
13
* . .. .... \- * . ..-. .
.6 K 3
S
be reflected in reduced power requirements and wide-open flow was selected as operational test points.operating costs. Since ventrifugal fans are designed to move air against
resistance, it is inappropriate to test fan performance2. The experimental test apparatus did not truly near wide-open flow. Operating centrifugal fans in a
represent a typical DWDI centrifugal cabinet fan since range of less than 25 percent wide-open flow is notone fan inlet was capped off. However. the test results recommended because surging usually occurs. Table Ishow the positive effects of the inlet air guide fairing summarizes the results of operating points over theand it is expected a conventional DWDI fan would flow range recorded for each of three rotative speeds.benefit to an even greater extent from the inclusion Although an absolute value of improved efficiencyof two inlet guide fairings. could not be quantified for this series of trials, 10 of
the II trials resulted in an increase in fan capacity3. The testing determined that fan performance and efficiency with the fan retrofit with the inlet air
could be improved using a flow contouring insert guide fairing (Table 2). The average efficiency for thewithin the fan cabinet. The test fan operated at apeak static efficiency of 38 percent in a standardconfiguration. With the inlet air guide fairing in place, Table ithe test fan operated at 43 percent peak static ef- Sunmary of Fan Testing .ficiency. This represents an improvement in peak Flow Average Peakefficiency of 13 percent. The average efficiency of the Rotative Range Efficiency Efficiencytest fan improved by 20 percent with the fairing in Speed (% ofWide- With Faiing/ Withou Fairingplace due to the increase in the region of efficient (rpm) Open Flow) Without Fain Without Faiingoperation. 1500 30-73 41/31 43/38
4. Testing involved operating the fan at three 1750 44-78 42/35 43/37rotative speeds, through a range of no-flow to wide-
- open flow. An operating range of 25 to 75 percent of 1990 29-69 37/35 42/38
Table 2Fan Testing Data
Static StaticPNMe Premure Static Static
Rotative With Without Efficiency EfficencySpeed Flow Faing Faking With Without(rpm) (Win) (in. wc) (In. we) pawn Faidg
1500 1200 2.00 1.94 43 28
1500 1500 1.75 1.75 42 38
1500 2300 1.47 1.50 40 36
1500 2900 1.09 .87 40 21
1750 2000 2.44 2.38 41 37
1750 2500 2.19 2.06 43 37
1750 3000 1.88 1.75 41 35
1750 3500 1.38 1.34 43 32
1990 1500 3.50 3.38 32 29
19 2000 2.88 2.75 42 38
1990 3500 2.19 2.13 37 38
Average: 40.4% 33.6%
14
I I trials without the fairing was 33.6 percent; with the METRIC CONVERSION CiHARTfairing in place the average efficiency was 40.4 percent.
I in. wc = 249 Pa5. For a system where energy to operate fans
is half the total annual 14VAC energy usage, a 20 1 in. Hg = 3385 Papercent reduction in fan power would result in anoveiall energy consumption reduction of 10 percent I hp = 0.74 kWfor the HVAC system, a substantial and favorableimprovement. I in. = 25.4 mm
6. Further refinement and evaluation of fairing I ft = 0.3048minserts is necessary to more accurately quantify ef-ficiency improvements and cost-benefit ratios. I psi = 6.894 kPa
(°F.32)0.55 =C
4 CONCLUSION I rpm 0.104 rad/s
I Ibm/cu ft = 16 kg/m 3
The results of this study indicate that the efficiencyof cabinet fans common to Army air-handler systemscan be improved 20 percent overall by inserting a fiber-glass fairing to guide inlet air.
15
CERL DISTRIBUTIONChief of Engineers 8th USA. Korea MYICAr Li: Tech Mnitor AiThg: CAFE:N 96271 ATTN: RMCS 20315ATiM: DALN-ASI-L 121 A -i EAFE 9625 ATTN: Faclisties EngineerATTN: DAEN-CCP ATTN: EAWC-t 96212 Oakland Army Base 946265.TM: OAEN-CW Bayo NOT 07002ATTN: DAE-Cut oRX/us C'jinAd Forces Command 96301 Sunny Foint NOT 21461ATTN: 0MMI-Cdt-SR ATTN: EUSA-IUC-CFC/EngrATTN: DAM-Ceo kkADCOM, ATTN: DRDNA-F 071160ATTN: OAL-CP USA Japan (USM.J)ATTN: CaLm-CC Ch,. FE Div, MJEN-FE 94343 TARCOM. Fec. DIV. 48090ATTN: OMEM-ECC Fat Caqr (lOetthul 96343AiM: DAEN-ECE Fac Eng (Okinsal 96331 TRADOCATTN: OALII.ZCF 140. TRADOC. AlT:4 LiEN-FEATTN: DAEN-ECI ok t. T 00 ATTN, Facilities EngineerATTN: DA~I-UO For Ara 90 elveir 22060Aflhl: DAiRCArea Engineer. ACO-Aree Office Fort Bumming 3190S
ATM AN-0M Arnold Air Force Station, TO! 37339 Fort Bliss 79916ATTN: DAEM-UI Carlisle Barracks 17013ATTN: DALM..ZCZ Western Area Office. CE Fort Chiaffee 72902ATTN: DAEM-ZCE Vandenberg AI, CA 93437 Fort Dia DO64ATTN: DALM-ZCI Fort Eustis 23604Aiim: DAEM-ZCM 416th Engineer Comiand 60623 Fort Bard.. 3090
ATTN.: Facilities Engineer Frt Hmilttoa 11212FESA. AiiM: Library 22060 Fort Benana Harriso" 46216
US Military Acdem 10996 Frt Jackson 29207RFESA. AiiMI: OCT 111 19906 AM;i Facilities Engineer Fort Knox 40121
ATTN: Oupt of 6Oeogrpy A Fort Leavmmoit. 66027US Army Engineer Districts Computelr Science Fort Lee 23801
ATTN: Library ATTN: D1CPCR/IPEM-A Fort MoClell~it 3620Alaska 99501 Fort Monroe 23651Al Satin 09616 Engr. Studies Center 2031S Fort sucker 36362
*Albuquerque 81103 ATTN: Library Fort Sill 73503Baltinore 21203 F. Leonerd Maod 65473Buffalo 14207 AWK. ATTN: COM-WE 02172Charleston 29402 TSAICOM. ATTN: STSAS-F 63120Chicago 60604 USA AISCI 61299Detroit 48231 ATTN: BAIlS-RI-I USACCFar Cast 96301 ATTN: BalM-IS ATTN: Facilities EngineerFort Worth 76102 Fort Nueciwa 51613Galveston 775S0 OAR= - Dir.. laIst.. 6 Svcs. Fort Ritchie 21719Huntington 25721 ATTN: Facilities EngineerJacksonville 32232 ANOON 0781 ETOJapan 964 MArdeen Proving Grown 21006 AMi: Facilities EngineerKanasC City 64104 Army MitlS. and Mechanics e. Ctr. Fart Shifter 96668Little Rock 72203 Corpus Chfistl Atop OGpe 78419 ATTN: APEN-INLos Angeles 9003 Hesry BiaMM Laboratories 20763Loisville 40201 11 yPryn Ground 6402 SHAPEf 0906Mophis% 38103 ja~ro rsn rud4726 ATTN: Survivability Section. CCX-IPSMobile 36626 Fort Mtouth 0703 Infrastructure Brach. LAUBANashville 37202 Le-tterel As Bee 7201Mew England 02154 %sic ADD Ctr. 0176 N0 UKIICOM 09126Mew Orleans 70160 How Co el AMW Onset 17070 ATTN: CCJ 4/7-LOEMX" York 10007 Pueblo AoW DeOW 6101Norfolk 23510 Bad liser Asr Ones 76901 Fart Selveir. VA 22060Onh 66102 Moisten Arsenal 3660 ATTN: ATZA-OT-4IPhiladelpia 19106 Re1ck Iland Arsenal 6129 ATTN: ATZA-lh-SiPittsburgh 15222 Savannsoop BeN Get 6104 AM*i. hilA-FEPortland 97206 SMarg AIW lega 9633 ATTN: Engr. LibraryRiyadh 09036 Seneca LoW lega 14641 ATTN: Cendon Liaison Office (2)Rock Island 61201 TabyhMi L.W legal 18466 ATTN: 10 LibrarySacrmento 95614 Towe AtW Capel 346174San Francisco 94105 Watarvliet Arsenl 12166 Cold 4egion ReseacN Enining Lab 03765Savannah 31402 van Proving Broun ow3 ATTN: LibrarySeattle 96124 MIND SendB Missile 111n10 0021St. Louis 63101 EMI. Lii: Library 2206St. Paul 55101 OA ATTN: ILL-NI 22314Tulsa 74102 Waterways Experiment station 39160Vicksburg 39160 FORSCOM ATTN LibraryWalla Walla 99362 FORICU Eginwe. LAM: Mill-FEWilmington 26401 AT: Facililist EIgioser IQ 59111 Airborne Carps and 2B307
Fert Inciana 06934 Ot. Ing"US Army Engi near Divisions Fart era"g 211307 AM: AFlA-FE-UAT: Library Fart Cmie) I 6232
Europe 09757 Fort Corna 8113 Cbsaut MSil. IL 618WHuntsville 31807 Fert eaees 01433 3366 CZS/SE Step 27LeorS? isissippi Valley 36160 Fart Drum 13601Middle East 09036 Fert Nee 7#644 sarten wU 92609Middle East (Rear) 22601 Fen rt ndatm eGp 17063 ATTN: AM 49E-/lEMissouri River 611101 Fart Irain 861311North Atlantic 10007 Fort Sen Meaon 1334 Tyndall Aft. FL. 32403Eath Central 60606 Fort Lenis IS43 MISC/Engirn 89 Service LabNarth Pacific 972061 Fart Mayo 5466Oii iser 411201 Fart OWimeN" 30330 NAVECPacific Ocean BIDS6 Fart BeimrD 6. tede 207%6 ATTN: ISTU Liaison OfficeSouth Atlantic 3033 Fert Ord 93941 Atlantic Division 23611Soush Pacific 94111 Fart Polk 71469 Cmepaoe Division 303745411401 term 7620 Fent Rickeraei 911906 leather, Division 29411
Fort RWay 06441 Pacific SIVISIon 916614JS tug uoe PM14idi of son Francieve "I12 SartheaDivision 19112
%O Ith Atop Training CNdN 09114 Fern Shiiden 8603 "weaen Division 64owMii: A~IiS..BEM1 I I Fert s34eft 31313 AMi: Sr. Iock. FM-f4W 22332
NO. 7th ARy UDCS/Egr. 09013 Fart "Immight 9616 ATTN: AssI. COR R0. FAC-O3 22332ATN: AAME-EM (4) leanceeer s4,. 90000
V. Carps 09079 11M 9241AMi: MuLTCN (S) gg90 ATTN: LIbr"r (Cede LOSII
Il. Corps 091114 AiM: MILO-V M834ATTN: AETSIEM (5) AMi: F691iliti141 EngiW nerefanee Teafeical lnte. Cante 22314
21st Soppert Cmnd 09321 FlsaimmsAne am 803TTNi: MSA (12)ATTN: IEN (5) Wlter ReedN MEIR51K Berlin 0974 Cnnall ing Secialee Library 10017LAMi: AEE (21 INS= -ch. ail. Div. iou yaen. BY
Souther European Toat Farce 091" AiiTN Facilities EngineeAMi: LEKt-Eng (3) Arlingto "ell Italian Q2) 2212 waetaioe Gerd Steem 20310
Installatioin Support Activity 0943 Vit Mill FrnmsStatism 2216 Installation DivisionAMi: ACS-W
pow US Government Printing Office 22NIsoh USA. KOMa ATTN: Facilities Enginee Reiving sectioni/9eItry Capie (21
AMi: EWE (B) 96301 C"Mn station 2214AMi: CAFE-1 94136 Fart Lestey J. Muir 20319 6ATTN: EWEC-lI 96124
Dolan, William H4Mlodification of cabinet fans with inlet air guide fairings to improve
performance. - Champaign, Ill :Construction Engineering Research Laboratory;available from NTIS, 1983.
15 p. (Technical report / Construction Engineering Research LaboratoryE-181)
1. Fans (machinery) -- performance. 1. Title. II. Series: Technicalreport (Construction Engineering Research Laboratory) ; -181.
bk*" -;
II
