Heat Transfer Calculations for Cooling System 2011-26-0087

Published on
Performance Prediction and Experimental 19th-21st January
Validation 2011 SIAT, India
Manvendra M Umekar and D Govindaraj
Ashok Leyland Technical Center, India

Copyright © 2011 SAE International and Copyright © 2011 SIAT, India

ABSTRACT INTRODUCTION
The two major areas of engine development, the automobile Internal combustion engine generates large amount of
industry is concentrating and putting its major product heat during combustion, part of which is converted into
development efforts on are emission control and improved mechanical energy. The difference is the waste energy and
fuel efficiency. These developments either directly or indirectly it is carried away by hot exhaust gases, intake air and
impact the cooling system requirements significantly. A small engine cooling system. Engine cooling system is the explicit
change in cooling system requires the vehicle to undergo cooling of engine components, which removes the heat by
cooling tests before commercial release to make sure it convective heat transfer mechanism. The combustion chamber
performs as per requirements. Normally, on an average, 4-5 temperature is sometimes even more than melting temperature
trials are conducted before clearing the vehicle for production. of engine material and hot enough to evaporate the lubricants.
This involves a lot of time, efforts and resources. The role of engine cooling system is to remove heat fast
enough to keep the engine material temperature well below
The objective of this study is to develop a methodology its melting point and increase durability of engine.
to predict the cooling system performance analytically and
validate the results with experimental data. This study will According to the type of cooling system, internal combustion
help in reducing the number of trials thereby saving cost, engines are classified as air cooled engines and water cooled
time and efforts involved. Basic components of cooling engines. Air cooled engines carry fins on the outer surface
system include radiator, charge air cooler and fan. The most of the engine body (cylinder), which is directly exposed
important parameters used to validate the analytical results to atmospheric air. Unlike air cooling, when water is used
are the Limiting Ambient Temperature (LAT) and charge as a heat transfer medium it is called as water cooled
air cooler outlet temperature. LAT is the maximum ambient engine. Water is preferred as heat transfer medium over
temperature at which the vehicle cooling system can work air because of its higher specific heat capacity, density and
without failure. thermal conductivity. This allows water to transfer heat over
greater distance with much less volumetric flow and reduced
The Steady-state analytical calculations have been performed temperature difference. This property of water allows it to
for engine cooling system (Radiator – Charge Air Cooler carry away heat from the primary heat source to a secondary
(CAC) – Fan) combinations using lumped parameter approach cooling system, which can be designed more optimally.
and the methodology is validated experimentally. The Generally, engine cooling system consists of radiator, charge
deviation between the proposed analytical method and test is air cooler and fan (Refer Fig. 1).
reasonable. FAN laws are used to evaluate fan performance
for different speeds of engine.

Keywords : Heat Transfer, Cooling System, Performance,

Heat Exchanger

1
Symposium on International Automotive Technology 2011

ANALYTICAL METHODOLOGY
(CALCULATION)
The component performance characteristic are used in
the methodology, to evaluate the effect of cooling system
performance parameters. The smooth component characteristic
curves are plotted using component test data. The characteristic
curves are extrapolated or interpolated to predict component
behaviour at the given condition.

The Number of Transfer Units (NTU) method is commonly

used to study heat exchanger performance,

where, Cmin is the minimum value of mCp on the cold or

hot side and U is overall heart transfer coefficient.

Heat exchanger effectiveness (e) is defined as the ratio of the

actual amount of heat transferred to the maximum possible
heat transfer that can be transferred with an infinitely large
area.

The analytical methodology is based on the following

justifiable assumptions,

- The heat exchanger operates under steady state

conditions.
- No phase change occurs: both fluids are single phase
Figure 1. Engine Cooling System and are unmixed.
- Heat losses are negligible.
Modern product development efforts are greatly relying
on Computer Aided engineering (CAE) as it reduces time - The temperature in the fluid streams is uniform over the
to transforms a concept into reality. This paper discusses flow cross section.
the approach and guidelines to find the appropriate cooling - The fluids have constant specific heats.
system analytically for the given engine, saving the cost and
time incurred in testing. The CAE involves detailed analysis - The fluids are incompressible.
of all the components of a system and the time required - Constant radiator inlet temperature.
increases with complexity of problem and expected level of
accuracy. But it is necessary to utilize CAE efficiently to - Uniform air flow distribution.
obtain reliable solutions to various real life problems. This - Effect of turbulence is negligible.
paper explains simple and quick analytical approach to predict
the behaviour of engine cooling system (radiator + CAC + The combined behaviour of all cooling system components
fan). Cooling system performance is predicted using lumped (i.e. radiator, CAC and fan) and the effect of change in
parameter approach along with the approximate representation performance parameters for individual on the whole system
of cooling system components. The engine test bed data is evaluated (studied) using operating point concept.
along with component characteristic behaviour curves are
used to predict engine cooling system performance. The Operating Point
component characteristic curves are developed by conducting
It is defined as the point of intersection of component
tests at supplier end. Also as the lumped parameter approach
characteristic behaviour curves. It is the only point at which
used in the methodology is highly sensitive to the inputs, a
two components can perform together (as shown in Fig. 2
separate analytical study has been done to verify the given
and 3). The operating point for given set of radiator-fan
component characteristics [1].
2
 Symposium on International Automotive Technology 2011

combination is the function of fan speed. As the fan speed speed Vs flow rate). The heat removed by cooling system
increase from N1 to N3, operating point shifts from “A” is calculated using radiator heat transfer characteristics for
and to “C” as shown in Fig. 2. In this case total pressure evaluated coolant (inner fluid) flow rate and air (outer fluid)
drop across cooling system is due to two heat exchangers flow rate.
(i.e. radiator and CAC). The effect of radiator and CAC on
air flow is modelled with the help of their pressure drop
characteristics. The total pressure drop across cooling system
is the combined effect of radiator and CAC as shown in Fig.
3. The mass flow rate and pressure drop at the operating
point is calculated using radiator, CAC and fan characteristic.

Figure 4. Radiator (Heat Exchanger) Characteristic

Heat exchanger effectiveness is one of the important

performance parameters of heat exchanger which allows
one to predict the performance of heat exchanger under
different conditions. This paper discusses the comparison
between analytical and physical performance of radiator (heat
exchanger), hence radiator analytical effectiveness (eA) must
be equal to the one measured through test ( i.e. Experimental
effectiveness, (eE) under same working conditions.
Figure 2. Operating Point for Radiator - Fan System
(1)

The analytical effectiveness is defined as the ratio of

heat transfer at operating point to maximum heat transfer
at component testing condition. QOpt pt is calculated from
component characteristic data, hence Q Max under same
conditions is used to calculate analytical effectiveness.

(2)

Maximum heat transfer is given by,

(3)

Where, inlet temperatures for inner and outer fluids are

obtained from component test data.
Figure 3. Operating Point for Engine Cooling System
(Combined Behaviour of Radiator and CAC) Radiator actual (test) effectiveness is defined as the ratio
of heat carried away by the coolant or the heat required to
Heat exchanger characteristic can be defined as the rate of be remove using engine cooling system to the maximum
heat transfer over a wide range of flow rates. In other words possible heat transfer.
heat transfer from inner fluid to outer fluid for different inner
fluid flow rates and visa-versa as shown in Fig 4. Coolant
(4)
pump is mechanically (directly) coupled with engine speed,
hence at given speed of engine respective coolant flow rate
is calculated using coolant pump characteristic. (i.e. pump
3
Symposium on International Automotive Technology 2011

Maximum heat transfer is given by, exchanger core size, fin structure and fan size are used in the
testing to understand the effect of variation in performance
(5) parameters and components on the end results. The details
of cooling system combinations and components are as
Hence from equation 1, 2, 4 and 5 mentioned in Table 1.

Test bed setup is modified taking into account the assumptions

(6) and limitations of algorithm. The general practice of using
fresh coolant from tank (open loop) for testing is avoided and
an attempt has been made to ensure that the test conditions
Radiator water inlet temperature is derived using equation 6, resembles the actual vehicle operating conditions (closed
system) by installing fan, radiator and CAC on test bed
with engine. This enables recording of relative (cumulative)
temperature rise of coolant with engine speed. Analytical
(7) calculations are performed assuming that there is no coolant
recirculation i.e. pump outlet flow rate is nearly equal
to radiator inlet coolant flow rate. To replicate the same
during testing, jack opened thermostat is used during testing.
Hence, radiator water outlet temperature is, Similarly as discussed before the algorithm assumes that the
fan is directly coupled with engine (i.e. constant speed ratio),
hence to ensure the same during testing hydraulic clutch of
(8) fan is locked to avoid any slip or relative motion due to
viscous effect.

FAN laws are used to evaluate fan performance for different The parameters of interest during testing are Engine Water
speeds of engine, which makes the algorithm independent of Inlet Temperature (EWIT), Engine Water Outlet Temperature
engine speed and fan speed [Appendix A]. Similarly use of (EWOT), pressure drop across radiator, coolant flow rate,
pump characteristic data makes the algorithm independent radiator air inlet velocity and temperature. Resistance
of coolant flow rate. Temperature Detector (RTD) and U type manometers are
used to measure temperature and pressure across radiator
ENGINE TESTING respectively. Coolant flow rate is measured using Rotameter,
which is installed in the coolant path between radiator
The objective of engine testing is to validate the simulation outlet and pump inlet. The pressure drop due to restrictions
tool developed for cooling system performance prediction. offered by rotameter can be avoided with the use of pump
Three different cooling system combinations in make, heat characteristic (Speed Vs flow rate) to get approximate coolant

Table 1. Combinations Used for Cooling System Trials

4
 Symposium on International Automotive Technology 2011

flow rate. The radiator frontal area is virtually divided into

12 equal squares to measure radiator air inlet temperature
and velocity as shown in Fig. 5. At each section, a RTD
sensor was placed using bolted bracket to measure air inlet
temperature and anemometer was placed as shown in Fig. 6
to measure air velocity. The size of each section is chosen
equal to the diameter of anemometer to minimize air flow
disturbance.

Figure 6. Radiator Air Inlet Velocity Measurement,

Sealing for Cowl Side Leakage, Aeration Check

RESULTS AND DISCUSSIONS

The focus of this paper is to validate the analytical
methodology for performance prediction of cooling system
with experiments. Knowing the fact that algorithm is very
sensitive to inputs, the analytical check has been performed
for all component characteristics to ensure the accuracy of
the algorithm. Due to the incomplete inputs from the supplier
for the third combination, results were compared for only first
two combinations. Some parameters measured during engine
testing like coolant flow rate, heat carried away by coolant
and air radiator inlet temperature (ambient temperature) are
used as inputs for the analytical calculation and remaining
parameters (i.e. EWOT, EWIT and air velocity) are used to
validate the outcome of the algorithm. During testing it is
observed that the radiator air inlet temperature and velocity
are not constant across radiator front area, minor variations
were observed between all 12 points. Hence arithmetic
mean temperature is used as input for analytical calculation
and similarly arithmetic mean of velocity is used for the
validation.

The lumped parameter approach is used to predict the engine

cooling system performance analytically. As shown in Fig. 7,
Figure 5. Setup for Measurement of Radiator 9, 11 and 12, EWOT and EWIT compared reasonably well
Inlet Temperature between analytical and test results for both the combinations.
For the combination 1, Maximum deviation for EWOT and
EWIT is 7.8% and 9.1% respectively at 2000 rpm. The
Engine cooling trials are performed under full throttle
reason for this difference can be attributed to the deviation
condition (FTP) over a wide range of engine speeds. Test
in Inlet Temperature Difference (ITD) as shown in the
data is collected for every 200 rpm increment from 1000 rpm
Fig. 8 and 10. EWOT and EWITs are dependent strongly
to 2400 rpm. Initially engine was run for 15 - 20 minutes
on ITD which is the difference between the coolant radiator
for reaching to a stable state before measuring the data. As
inlet temperature and air radiator inlet temperature. Similarly,
it has been a time consuming exercise to measure radiator
for the combination 2, identical trends can be observed for
inlet velocity, randomly two engine speeds are selected for
EWOT and EWIT as seen in the Fig 11 and Fig. 12. ITD is
comparison with simulation tool. Radiator inlet velocity is
a function of air inlet temperature, which again is dependent
measured only for 2400 rpm (Max power) and 1400 rpm
on room temperature. Increase in room temperature is due
(Max torque).
5
Symposium on International Automotive Technology 2011

to closed engine test bed conditions. Engine generates large (v) Approximation of component performance curve to
amount of heat and a part of which is loss to the surrounding evaluate component behaviour under different conditions.
air through radiation. The heat transfer increases with engine
speed and hence the temperature of surrounding air. (vi) Air flow due to leakage is not captured in the fan laws.

As shown in Table 2, a good comparison has been observed

between analytical and test results for radiator air inlet
velocity. The deviation of 5.5% is witnessed at 2400 rpm
for both the combinations, where as the maximum deviation
is 9.5% and 13 % for combination 1 and combination
2 respectively at 1400 rpm. Air velocity is higher for
Combination 1 than Combination 2 because as it can be seen
in Table 1, core of second combination (fin pitch = 2.5 fpi)
is much denser than combination 1 (fin pitch = 1.8 fpi) and
son the increase in restriction which causes reduction in air
flow. Similar trend has been in theoretical method too. For
all these cases, air velocity deviation is less than 15% which
is reasonable. As discussed earlier, during experimentation
Figure 7. Combination 1 - Comparison between
a varying radiator air inlet temperature is observed before
Analytical and Experimental Results for EWOT
radiator, measured at all 12 locations. While a constant
temperature assumption is made during analytical calculation
and arithmetic mean of experimental data is used in the
algorithm. This could be one of the reasons for deviations.

Apart from the above reasons, the other probable causes for
deviation are,

(i) Incompressible fluid assumption during analytical

calculation.

(ii) Effect of turbulence which is not captured in the

analytical calculation.

(iii) Uniform air flow assumption.

(iv) Sensitivity (error) of the sensors and human error.

Figure 8. Combination 1 - Comparison between
Analytical and Experimental Results for ITD

Table 2. Radiator Air Inlet Velocity

6
 Symposium on International Automotive Technology 2011

Figure 9. Combination 1 - Comparison between Figure 12. Combination 2 - Comparison between

Analytical and Experimental Results for EWIT Analytical and Experimental Results for EWIT

CONCLUSIONS
The Steady-state analytical calculations have been performed
for a cooling system (Radiator – CAC – Fan) combination
using lumped parameter approach and maximum deviation
observed between analytical results and test is nearly ±15%
over a wide range of engine speeds. The radiator inlet
velocity shows good match between test and analytical
methodology for both combinations with maximum deviation
of 13%. The accuracy of the proposed analytical method is
depend on the accuracy of the component characteristic in the
cooling system. The combination with operating point falling
in the range of component characteristic data points shows
better comparison with mathematical model. The proposed
Figure 10. Combination 2 - Comparison between methodology could be very useful to give quick first cut idea
Analytical and Experimental Results for ITD of required cooling system for the designers. The results of
all combinations are of different nature and this study will
aid in development of wide range of correlations between the
test and analytical results apart from minimizing resources
and efforts involved in common practice.

REFERENCES
1. Yu-Juei Changa, Kuei-Chang Hsub, Yur-Tsai Linb, Chi-
Chuan Wanga. “A Generalized Friction Correlation For
Louver Fin Geometry”, International Journal of Heat
and Mass Transfer 43, 2237-2243, 2000

2. Incropera F P and De Witt D P, “Fundamentals of Heat

and Mass Transfer”, 3rd edition, John Wiley and Sons,
New York, 1990

3. Dohoy Jung and Dennis N Assanis, “Numerical

Figure 11. Combination 2 - Comparison between Modeling of Cross Flow Compact Heat Exchanger with
Analytical and Experimental Results for EWOT Louvered Fins using Thermal Resistance Concept”, SAE
Paper No. 2006-01-0726, 2006

7
Symposium on International Automotive Technology 2011

4. Tsai B J and Wu C L, “Investigation of a Miniature CONTACT

Centrifugal Fan”, Applied Thermal Engineering, Vol 27,
229-239, 2007 Manvendra M. Umekar,
Senior Engineer - Advanced Engineering,
5. Jeff Stein and Mark M Hydeman, “Development and Ashok Leyland Ltd.,
Testing of the Characteristic Curve Fan Model”, AN- Technical Centre, Vellivoyal Chavadi,
04-3-1/ 347-356, ASHRAE, 2004 Chennai - 600103, India
Phone: +91 44 25398155 Fax: +91 4425398003
6. Stinnes W H and Von Backstrom T W, “Effect of Cross- E-mail: [email protected]
Flow on the Performance of Air-cooled Heat Exchanger
Fans”, Applied Thermal Engineering, vol 22, 1403-1415, Govindaraj D.
2002 Senior Engineer - Advanced Engineering,
Ashok Leyland Ltd.,
7. Hongmin Li, “Flow Driven by a Stamped Metal Cooling
Technical Centre, Vellivoyal Chavadi,
Fan - Numerical Model and Validation”, Experimental
Chennai - 600103, India
Thermal and Fluid Science, vol 33, 683-694, 2009
Phone: +91 44 25398122 Fax: +91 4425398003
8. Szu Hsien Liu, Rong Fung Huang and Chuang An E-mail: [email protected]
Lin, “Computational and Experimental Investigation
of Performance Curve of an Axial Flow Fan Using DEFINITIONS/ABBREVIATIONS
Downstream Flow Resistance Method”, Experimental
Cp = Specific heat capacity,
Thermal and Fluid Science, vol 34, 827-837, 2010
e = Effectiveness.
9. Gilies Philip T, “Investigating the Use of Alluvial Fan EWOT = Engine water outlet temperature, deg C
Volume to Represent Fan Size in Morphometric Studies”,
Geomorphology, vol 121, 317-328, 2010 EWIT = Engine water outlet temperature, deg C
FTP = Full throttle condition
10. Wei-Mon Yan and Pay-Jen Sheen, “Heat Transfer
Q = Heat load on the system, J
and Friction Characteristic of Fin-and-Tube Heat
Exchangers”, International Journal of Heat and Mass rpm = Revolutions per minute
Transfer 43, 1651-1659, 2000 RTD = Resistance temperature detector

11. El-Hawat S M, Heikal M R and Sazhin S S, “An T = Temperature, deg C

Improved Three-Dimensional Numerical Model of
Flow and Heat Transfer Over Louvered Fin Arrays”, SUBSCRIPT
International Journal of Heat Exchangers 1524-5608/ Opt pt = operating point
Vol II, 3-10, 2001
Max = Maximum
Comp = component
ACKNOWLEDGMENTS
E = Experimental or Test.
We w i s h t o e x p r e s s o u r s i n c e r e t h a n k s t o D r.
A = Analytical
Venkatasubramanian S. H. for his invaluable guidance
throughout the project. We also wish to thank Mr. Prakash min = minimum
C. and Dr. Sathya Prasad M. for their valuable suggestions i = inlet
to guide us in the right direction. We specially thank Mr. J. o = outlet
Narayanan, Test Engineer for his constant support throughout
c = coolant
the testing phase and for ensuring that our requirements
are met. We also wish to express our gratitude to Mr. Gopa SUPERSCRIPT
Kishor Gummadi, Mr. Kothandam, Mr. R.K.Kannan, and
Mr. Vasudevan for providing us the resources with their h = hot fluid
active involvement. We also like to appreciate the support c = cold fluid
of Mr. Venketasan, Mr. Manikandan, Mr. John and Mr. i = inner fluid
Sundaravadivel during experimentation exercise without
o = outer fluid
which the validation of algorithm could have been difficult.

8
 Symposium on International Automotive Technology 2011

APPENDIX

The Technical Paper Review Committee (TPRC) SIAT 2011 has approved this paper for publication. Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of
This paper is reviewed by a minimum of three (3) subject experts and follows SAE guidelines. SIAT 2011. The author is solely responsible for the content of the paper.
SAE Customer Service:
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or Tel: 877-606-7323 (inside USA and Canada)
transmitted, in any form by any means, electronic, mechanical photocopying, recording, or otherwise,
Tel: 724-776-4970 (outside USA)
without the prior written permission of SIAT 2011.
Fax: 724-776-0790
Email: [email protected]
SAE Web Address: http://www.sae.org
Printed in USA