2004 SAE AUTOMOTIVE
15-17
June 29-July
1
EFFECT OF FLOW OBSTRUCTIONS
and engine compartment recirculated air on heat exchanger and system
performance as simulated in the test bench
With great contributions by:
X. Li
W. Atkinson, H. Fernquist, W. Hill, S. Lepper, J. Wertenbach
Pega Hrnjak
We will:
• Define the problem;
• Identify differences between reality
and experiments;
• Look at the experimental results:
– Effect of flow obstruction (air side
maldistribution);
• Discuss design with obstructions and
maldistibution in mind.
A typical situation at the front
Effect of obstructions
• AR CRP – check the sensitivity of four
systems on barriers in most difficult
situations – low rpm and flow, high T;
• Baseline – no obstructions;
• Distance of barriers:
– 51 and 108 mm upstream (bumper)
– 41 and 172 mm downstream (engine)
• Effect of system: Q and COP.
Inlet flow obstruction
Flow restriction at the outlet
In condenser wind tunnel
Barriers far from HX
Air out
A
Cross-Section A-A
303 mm
13 mm
Rear Flow
Restriction
Thermocouple
grids
Blocked area
Gas cooler/condenser
Bottom
172 mm
108 mm
100 mm
A
Front Flow
Restriction
49.5 mm
Side Wall
Air in
Side Wall
152 mm
187 mm
509.5 mm
559 mm
Top
Close to HX
A
Air out
Cross-Section A-A
Top
Bottom
Thermocouple
grids
13 mm
303 mm
Rear Flow
Restriction
41 mm
Blocked area
51 mm
100 mm
Gas cooler/
condenser
A
Front Flow
Restriction
49.5 mm
Side Wall
Air in
Side Wall
152 mm
187 mm
509.5 mm
559 mm
Problem in the lab:
Simulate unknown fan
Maintained constant DP.
Lab air flow is not sensitive to pressure drop (“stiff” fan).
Evaporator Chamber Wall
Condenser Chamber Wall
DP
Evaporator
EXV
Condenser
SG
Air in
Air in
TM
Comp
Compressor Chamber
Motor
MF
Short test matrix
Basic Test Point Matrix
Evaporator Air
Cond/Gas
Cooler Air
Comp
T (oC)
Flow
(l/s)
rpm
109
70
425
900
27
109
60
425
900
45
27
109
45
425
900
LSH45
45
27
130
45
850
1500
HSH45
45
27
130
45
1320
2500
Test
name
T
(oC)
RH
(%)
Flow
(l/s)
ISH70
45
27
ISH60
45
ISH45
Effect of proximity
Condenser air flow rate
Example
1.8
1.6
without restriction
with restriction at 52/41 mm
with restriction 108/172 mm
without restriction (ReRUN)
1.583
1.57
Air flow rate [kg/s]
1.43
1.37
1.4
1.2
1.021
1.00
1.0
0.94
0.88
0.8
0.6
0.50
0.5093
0.46 0.43
0.50 0.47
0.5126
0.44
0.50 0.47
0.5116
0.44
0.4
0.2
0.0
ISH70
ISH60
ISH45
Test conditions
LSH45
HSH45
Effect of proximity
Cooling capacity
Example
Evaporator capacity Qe [kW]
10
9
without restriction
with restriction at 52/41 mm
with restriction 108/172 mm
without restriction (ReRUN)
8
7.54 7.52
7.377.398
6.74 6.67 6.596.609
7
6
4.86 4.79 4.794.944
5
4
4.20 4.19 4.18 4.2
3.79 3.73 3.713.791
3
2
1
0
ISH70
ISH60
ISH45
Test conditions
LSH45
HSH45
Effect of proximity
Efficiency
Example
4.0
without restriction
with restriction at 52/41 mm
3.5
2.96
COP [-]
3.0
with restriction 108/172 mm
without restriction (ReRUN)
2.968
2.84 2.83
2.49
2.5
2.41 2.392.423
2.16 2.15 2.132.162
2.0
1.80 1.78 1.761.782
1.78 1.74 1.721.779
1.5
1.0
0.5
0.0
ISH70
ISH60
ISH45
Test conditions
LSH45
HSH45
What about various systems?
Air flow reduction when barriers are very close (50mm)
25%
R134a Baseline
R744
R134a Enhanced
R290 secondary loop
Air flow reduction
20%
15%
10%
5%
0%
ISH70
ISH60
ISH45
Test condition
LSH45
HSH45
Why difference?
1. Shape matters!
R134a baseline/R290
Barrier
R134a enhanced
R744
Why difference?
2. DP and test conditions matter!
Condenser/Gas cooler airside pressure drop
Airside pressure drop DPca [Pa]
120
100
80
60
40
20
0
0.0
0.5
1.0
1.5
2.0
2.5
Air flow rate AFRc [kg/s]
R134a baseline
R744
R134 enhanced
R290 secondary loop
We have assumed constant DP
Consequence: lower DP HX is more affected
DP
n2
DP barrier
DP barrier
n1
Both HXs have same flow
V
DP
In reality fan/HX interactions
are slightly different
Functions of
the fan
sensitivity
V
Capacity reduction due to barriers:
R744 more sensitive at high temperatures less at low
System capacity reduction
16%
R134a Baseline
R134a Enhanced
14%
12%
R744
R290 secondary loop
Very close barriers (50mm)
10%
R744 air flow rates are the lowest
8%
6%
4%
2%
0%
-2%
ISH70
ISH60
ISH45
Test condition
LSH45
HSH45
COP reduction due to barriers:
R744 more sensitive at high temperatures less at low
12%
R134a Baseline
R744
R134a Enhanced
R290 secondary loop
System COP reduction
10%
Very close barriers (50mm)
8%
R744 air flow rates are the lowest
6%
4%
2%
0%
ISH70
ISH60
ISH45
Test condition
LSH45
HSH45
How to design HX to be less
affected by barriers and
maldistribution?
• First understand:
– air side maldistribution;
– refrigerant side heat transfer;
– refrigerant side maldistribution;
– operation of the heat exchanger and
system effects;
• Then be creative.
Arranging refrigerant flow orientation is one way to improve
performance of R744 when barriers obstruct air flow:
Expose gas cooler exit to lower temp. –
sacrifice less important area!
velocity profile
v
v
v
v
w/o barrier
with barrier
Shorter headers – less costly
System effects
Reducing approach TD benefits R744 very much
o
R134a
80
o
P=10M Pa
60
31 C
o
Te mpe rature [ C]
100
120 C
120
o
t=38 C
40
Air
20
R744
0
-250
-200
-150
-100
-50
0
50
Enthalpy [kJ/kg]
From: Yin, Bullard, Hrnjak, 2002
R744 is very sensitive to gas
cooler exit conditions
120
4.0
η m=f(P r)
R744
3.5
W =W s /η m
28
COP: 11%
DP:0.5MPa
32
80
Gas cooler
CO P
Te mpe rature [C]
100
30
60
40
20
Comp
IHX
3.0
34
36
2.5
38
3.9 C Ev ap
0
-300 -250 -200 -150 -100
2.0
IHX
ε =0.8
o
-50
Enthalpy [kJ/kg]
0
50
40
1.5
7500
8500
42
9500
45
50
10500
11500
Discharge pressure [kPa]
From: Yin, Bullard, Hrnjak, 2002
But, most of the heat in gas
cooler is exchanged in inlet zone
Refrigerant temperature drop in a
single pass gas cooler
Enthalpy change as a function of
refrigerant temperature
130
130
120
T
su rfa ce
Test #3
120
o
p re d ic te d
T c a i= 4 3 .1 C
o
110
T c ri= 1 2 2 .6 C
100
T c ro = 4 6 .2 C
o
T c ro
90
o
p re d
= 4 4 .6 C
mr=27.14g/s
80
DP
70
DP
60
cal
m eas
=2 1kPa
= 1 0 9 .1 k P a
AFR:787scfm
50
0.2
0.4
0.6
0.8
Length [m]
1
110
Predicted
Measured
100
90
80
70
60
50
40
0
Temperature [oC]
o
Temperature [ C]
Tr,predicted
1.2
40
-150
-100
-50
0
Enthalpy [kJ/kg]
From: Yin, Bullard, Hrnjak, 2002
50
Same approach with R134a?
Summary
• Barriers reduce capacity and COP;
• Effect is perhaps less than expected;
• Results should be carefully interpreted due
to simulation of unknown fan – R744 had
less air even having lower DP;
• R744 system examined was more sensitive
to recirculation than R134a at higher
temperatures and less at lower temps;
• Effects of air side maldistribution could be
reduced by good design.