1、SAE Technical Standards Board Rules provide that: “This report is published by SAE to advance the state of technical and engineering sciences. The use of this report is entirelyvoluntary, and its applicability and suitability for any particular use, including any patent infringement arising therefro
2、m, is the sole responsibility of the user.”SAE reviews each technical report at least every five years at which time it may be reaffirmed, revised, or cancelled. SAE invites your written comments and suggestions.QUESTIONS REGARDING THIS DOCUMENT: (724) 772-8512 FAX: (724) 776-0243TO PLACE A DOCUMENT
3、 ORDER; (724) 776-4970 FAX: (724) 776-0790SAE WEB ADDRESS http:/www.sae.orgCopyright 1992 Society of Automotive Engineers, Inc.All rights reserved. Printed in U.S.A.SURFACEVEHICLE400 Commonwealth Drive, Warrendale, PA 15096-0001INFORMATIONREPORTAn American National StandardJ2082ISSUEDJUN92Issued 199
4、2-06COOLING FLOW MEASUREMENT TECHNIQUESForewordThis report is part of a series of SAE Reports dealing with aerodynamic testing of road vehicles, whichare as follows to date:SAE HS1566 JAN86Aerodynamic Flow Visualization Techniques and ProceduresSAE J1594 JUN87Vehicle Aerodynamics TerminologySAE J207
5、1 FEB90Open-Jet Wind Tunnel Boundary InterferenceSAE J2084 DRAFTTesting Methods ProceduresSAE J2085 FEB90Solid-Wall Tunnel Boundary InterferenceTABLE OF CONTENTS1. Scope . 22. References . 22.1 Applicable Publications. 22.2 Symbols 33. North American Practice 43.1 Siemens Automotive. 43.2 General Mo
6、tors. 43.3 Chrysler-Jeep Truck (Formerly AMC). 93.4 Ford Motor Company 103.5 National Research Council of Canada (NRCC) 144. European Practice 164.1 Audi A.G . 164.2 Mercedes-Benz (MB) 214.3 Motor Industry Research Association (MIRA) 214.4 Volvo. 215. Key Words 22APPENDIX AWork Statement #1 Ram Air
7、Effect on Cooling Flow Measurement. 23APPENDIX BWork Statement #2 Computer Controlled 5-Tube Probe Positioning and Flow Vector Measurement 28SAE J2082 Issued JUN92-2-1. ScopeThis SAE Information Report has been prepared by the Standards Committee on Cooling FlowMeasurement (CFM) at the request of th
8、e SAE Road Vehicle Aerodynamics Forum Committee (RVAC). Thecommittee was formed in January 1985 for the purpose of investigating what measuring techniques are usedby automotive product manufacturers to determine air cooling air flow rates and, if possible, to synthesize theseinto a recommended pract
9、ice report.Although a great deal is already known about engine cooling, recent concern with fuel conservation hasresulted in generally smaller air intakes whose shape and location are dictated primarily by low vehicle drag/high forward speed requirements. The new vehicle intake configurations make i
10、t more difficult to achieveadequate cooling under all conditions. They cause cooling flow velocity profiles to become distorted andunderhood temperatures to be excessively high. Such problems make it necessary to achieve much betteraccuracy in measuring cooling flows.As the following descriptions sh
11、ow, each company or institution concerned with this problem has invested a lotof time and as a result gained considerable experience in developing measuring techniques that appear toachieve reliable results. There is, however, little uniformity at the present time among the methods used bydifferent
12、companies and no indication at this time of a trend towards a simple and universally acceptablemeasuring technique. If one can make generalizations, it seems that the North American industry appears touse vane anemometers, whereas the European industry appears to favor pressure measurements fordeter
13、mining cooling air flows.Usually cooling flow measurement makes use of ensemble average calibrations of arrayed sensors. The majordrawback of ensemble averaging usage of sensors appears to be the need for prior calibration of eachcombination of sensor array and radiator/vehicle front-end configurati
14、on. While there appears to be a growinginterest in also knowing flow distributionswhich necessitates the use of a plurality of sensing devicesverylittle is made of area-averaging techniques.Given the present state of the technology, this report covers simply an overview of the different measuringtec
15、hniques deployed in the industry, and it is left to the future to report on developments towards a unifiedcooling flow measurement method. Appendices A and B present a number of problem statements that wereidentified during the present CFM review and whose solution would promote a better understandi
16、ng of coolingflow measurement generally.2. References2.1 Applicable PublicationsThe following publications form a part of this specification to the extent specifiedherein.1. SCHAUB, U. W., CHARLES, H. N. “Ram Air Effects on the Air Side Cooling System Performance of aTypical North American Passenger
17、 Car.“ SAE 800032. February 25, 1980.2. WILLIAMS, J. E. “An Automatic Front-end Design Approach for Improved Aerodynamics but can and have been obtained in-house experimentally. For the experiments use is made of a small wind tunnel. The wind tunnel nozzle isattached and sealed to the front of the c
18、ar. The pressure loss is measured across the radiator and is thencalibrated against the volume flux, as obtained from a metering nozzle that is part of the small wind tunnel.Vehicle air path cooling hardware, such as grille openings, fan shrouds, cooling air discharge ducts andopenings, and other in
19、ternal air path elements for ducted and nonducted airflows need to be optimized duringfull scale vehicle tests in order to avoid problems with Reynolds number similarity. Internal/external flow fieldinteractions are investigated and optimized by means of 1/4 scale model tests.These model tests invol
20、ve the use of a radiator simulator, which is adjustable with respect to pressure loss inorder to maintain similarity in the extended pressure loss coefficient (radiator pressure loss coefficient x thesquare of the ratio of cooling air exit area to radiator face area).The simulator consists of a shor
21、t duct with three inserts: a variable area slide valve and two flow straighteners(honey combs) at the entry and exit planes of the valve for elimination of flow irregularities. The vehicle modelexterior is proportioned to provide the correct ratio of cooling flow engine bay exit area to vehicle fron
22、tal area.The slide valve is then adjusted to generate the same extended pressure loss coefficient as for the full scaleradiator at the full scale vehicle approach flow Reynolds number. The technique of using a radiator simulatoravoids the dynamic scaling problems generally associated with incorrect
23、modelling of internal flow Reynoldsnumbers. The one-fourth scale cooling flow is then calculated by means of measured static pressures acrossthe radiator simulator and from the set (known) total pressure loss coefficient.SAE J2082 Issued JUN92-17-FIGURE 14EFFECT OF APPROACH AND COOLING AIR SPEEDS ON
24、 COOLING AIRFLOWSSAE J2082 Issued JUN92-18-FIGURE 15FREE AIR OPEN PROPELLERTYPE ANEMOMETER CALIBRATIONS ANEMOMETERS MOUNTED INSIDE TRAVERSING FRAMESAE J2082 Issued JUN92-19-FIGURE 16PITCH SENSITIVITY OF A FRAMED PROPELLER ANEMOMETERSAE J2082 Issued JUN92-20-FIGURE 17YAW SENSITIVITY OF A FRAMED PROPE
25、LLER ANEMOMETERSAE J2082 Issued JUN92-21-Optical measurements (i.e., non-intrusive) with laser two focus (L2F) instrumentation were made at the VonKarman Institute for Fluid Dynamics (VKI) by measuring the time of flight of a large number of small particlesbetween two very small illuminated spots, c
26、alled probe volumes in a plane normal to the optical path. This typeof equipment is able to measure two-dimensional of velocity vector. The L2F head can be nearby themeasuring volume, or as far away as 10 ft, and it is possible as well to sample inside highly confined spaces,such as between the radi
27、ator and the AC condenser cores. There did not appear to be a requirement forspecial seeding techniques since the normal engine bay environment was found to be adequately seeded.The L2F measuring technique does have drawbacks, such as high original equipment costs and the long timeneeded to survey o
28、ne complete measuring plane.In view of experience with different related problems, turbulence and time averaged velocity profilenonuniformity are thought to be of importance to effective heat transfer. At present it is, however, not possibleto predict how a particular installation/configuration will
29、 perform as a function of these variables.4.2 Mercedes-Benz (MB)Cooling flow is determined through integration of the velocity field calculated frompressure survey data as follows.A small pitot probe combined with a static pressure probe (special MB design) is mounted on a traverse gear.The probe re
30、aches into the engine bay through a small gap in the hood. This gap is sealed by two flexiblerubber lips in order to avoid influencing the engine bay air flow. The probe is traversed over a 30 x 40 cm (12 x16 in) grid either inside the small space between the AC condenser and the radiator or, if the
31、re is nocondenser, in a plane just downstream of the radiator.The experimental data are recorded by a data acquisition system and processed by the wind tunnel facilitycomputer. Final output is in the form of contour plots of constant dynamic pressure or of constant air speed.Further reduction of the
32、 measured data is possible by transferring the data to the host computer in the dataprocessing center. 3D-plots of the radiator and the complete flow field are available.4.3 Motor Industry Research Association (MIRA)Cooling system heat dissipation rates on both fullscale claymodels and prototype veh
33、icles are measured by means of an external hot water system. The method is notcomplicated: only the water flow rate and the temperature difference across the radiator have to be measured.The present 60 kW water system has a capacity of 228 L (50 gal) and is inadequate for some vehicles, e.g.,trucks
34、and high-performance cars. A new 200 kW system is under construction. The method givessatisfactory results, but does have the drawback of having to route hot water lines between the metric andnonmetric surfaces of the wind tunnel balance, so that aerodynamic force measurements cannot be madesimultan
35、eously.For cooling flow measurement MIRA pioneered, and still uses, the technique based on a fixed array of 20 to 30pitot and reverse-pitot tubes imbedded inside the radiator core between the fins. Each so-instrumentedsample radiator requires calibration inside an airflow test stand before use for c
36、ooling flow measurement.Despite the nonuniformities usually revealed in the observed pressure distributions, the technique yieldsreasonably accurate cooling volume flow rates. At very low air speeds measuring accuracy is, however, nolonger acceptable and propeller anemometers are used.4.4 VolvoCooli
37、ng flow measurement is achieved through integration of velocities calculated at up to 30 gridpoints in the radiator inlet face from measured pressure differentials of specially built micro-orifices. The micro-orifices (8 to 9 mm high x 4 mm wide x 30 mm long, or 0.31 to 0.35 x 0.2 x 1.2 in) are pres
38、sed into the finnedradiator core. The difference between the throat pressure of each micro-orifice and the approach stream totalis sensed by a transducer inside a Scanivalve. The depression in pressure level generated by the throatamplifies the actual pitot static pressure difference by a factor of
39、4. This enables air speed measurement downto only 2 m/s (7.8 ft/s). The long aspect ratio design featured by the micro-orifice minimizes errors due to flowangularity effects. Each pressure is integrated over 1 to 2 s.SAE J2082 Issued JUN92-22-The measuring accuracy is satisfactory, but the method ha
40、s drawbacks. For example, measuring problems areencountered if there are upstream obstructions, such as an AC condenser, because the total pressure may nolonger be uniform. The need to custom-fit each micro-orifice into the sample radiator, the requirement tocalibrate each assembly as a unit, the le
41、ngth of time necessary to sample the whole flow field, and the inabilityto detect reversing flow are cited as disadvantages.Volvos experience with one type of European made (10 mm or 0.4 in) propeller anemometer has beendisappointing. In general, such anemometers tended to generate large errors and
42、were incapable of sensingreversed flow direction. Anemometers were used in traversing the exit plane of the radiator, but wakeinterference effects were never satisfactorily resolved. Loss of accuracy due to flow angularity and undetectedback flow associated with fan shroud proximity mitigated agains
43、t further use of propeller anemometers.The first experience with LDA was not encouraging: data acquisition was too slow, an excessively large numberof test data points were needed in order to determine cooling flow, and problems were encountered in finding asuitable place to calibrate the equipment
44、under similar conditions.5. Key WordsAir, Calibration, Condenser, Cooling, Fan, Flow, Heat, Measurement, Pressure, Radiator,Anemometer, Road Vehicles.PREPARED BY THE SAE COOLING FLOW MEASUREMENT STANDARDS COMMITTEESAE J2082 Issued JUN92-23-APPENDIX AWORK STATEMENT #1 RAM AIR EFFECT ON COOLING FLOW M
45、EASUREMENTA.1 ObjectiveTo compare and calibrate cooling airflow measuring devices with a standard orifice flow meterunder yawed and unyawed ram air conditions.A.2 Discussion of ProblemIntake efficiency and cooling flow measurement at high forward speed has not beenconsidered to be very important for
46、 standard road vehicles because maximum cooling requirements wereusually met at low forward speeds. This explains the common use of plenum type intakes such ascharacterized by the sudden expansion in Figure A1. However, as the drag coefficient continues to bereduced, the penalty of excess cooling ai
47、r and low pressure recovery becomes more apparent, and it willbecome necessary to build more efficient air paths and to measure cooling flow more accurately.At zero forward speed the captured airflow contracts upon entry, as depicted schematically in Figure A2 andproduces the highest velocities at t
48、he walls. Figure A3 shows how a theoretical pressure distribution on theinternal surface of an engine intake changes as the approach flow goes from zero to higher Mach numbers(i.e., nondimensional velocities). The suction peak just inside the lip highlight can cause lip separation, asFigure A4 depic
49、ts for an approach flow Mach number of 0.21.As the forward speed increases, the internal high speed flow gradually shifts away from the wall to mid-passage, as shown in Figure A3. As the captured flow velocity ratio V/VD increases beyond 1 (see FigureA2), the flow has to slow down as it expands to fill the radiator inlet face.Figure A5 maps typical loss characteristics for low speed intakes. Note that losses are high for low and veryhigh velocity ratios. Loss is a minimum when the intake and approach flow velocities are matched.Although automobiles suffer far greater fl
