Embed Size (px)
Citation preview
Comparison of Measured and Analytical
Performance of Shell-and-tube Heat Exchangers
Cooling and Heating Supercritical Carbon
Dioxide
Bechtel Marine Propulsion CorporationBettis Atomic Power Laboratory, West Mifflin, PAKnolls Atomic Power Laboratory, Niskayuna, NY
Page 1Supercritical CO2 Power Cycles Symposium September 9-10, 2014
Dioxide
Timothy Cox
Patrick Fourspring
Purpose
• Evaluate whether commonly used design tools
are applicable to Supercritical Carbon Dioxide
(S-CO2)
– Conventional shell and tube heat exchangers are
Page 2Supercritical CO2 Power Cycles Symposium September 9-10, 2014
– Conventional shell and tube heat exchangers are
well understood
– First operational data available with S-CO2
• Shell side and tube side
• Heating and cooling applications
• 100 kWe Brayton Cycle, 1MW Heat Source
• Two Shell and Tube Heat Exchangers
IST Overview
TurbineTurbine CompressorPrecooler
Generator
Chilled
Water
Hot
Oil
Page 3Supercritical CO2 Power Cycles Symposium September 9-10, 2014
IHX
Recuperator
Precooler
Shell 1
Precooler
Shell 2
Bypass
Intermediate Heat Exchanger
• Mineral oil on the shell (hot) side
• S-CO2 on the tube (cold) side
• Design duty 857 kW
• Geometry
6.0650 inch 6.0650 inch
Page 4Supercritical CO2 Power Cycles Symposium September 9-10, 2014
• Geometry
– Overall Length: 17 ft.
– Tubes: 230 – 10 inch diameter
– Baffles: 23 shell
Precooler• Chilled water in the tubes
• S-CO2 on the shell side
• Design duty 936 kW
• Geometry – Two identical units
– Length: 19 ft. each – 10 inch diameter
Page 5Supercritical CO2 Power Cycles Symposium September 9-10, 2014
– Length: 19 ft. each – 10 inch diameter
– Tubes: 77 each shell
– Baffles: 39 each
Heat Exchanger Analysis
• Xist® Shell and Tube Heat Exchanger Design
Software from the Heat Transfer Research
Institute (HTRI)
• Nodalized Method – important for the widely
Page 6Supercritical CO2 Power Cycles Symposium September 9-10, 2014
• Nodalized Method – important for the widely
varying fluid properties of S-CO2
• Fluid Properties:
– VMGThermo included with Xist®
– Can be linked with NIST’s REFPROP
– Oil properties entered via grid input
Steady-State Operating Data Points
IHX Precooler (Series)
Case
ṁ
S-CO2 (lbm/s)
Tin
S-CO2 (°F)
ṁ
Oil (lbm/s)
Tin
Oil (°F)
ṁ
S-CO2 (lbm/s)
Tout
S-CO2 (°F)
ṁ
Water (lbm/s)
Tin
Water (°F)
Cold
Idle 3.5 129.2 16.7 176.5 5.6 101.2 6.9 89.5
300°F
Page 7Supercritical CO2 Power Cycles Symposium September 9-10, 2014
300°F Hold
4.8 201.6 16.7 299.5 7.2 101.0 9.5 89.5
Hot Idle
4.8 429.3 50 571.3 7.6 96.8 7.1 81.5
Full
Power 8.9 361.6 50 548.0 11.2 97.1 15.8 81.0
IHX Heat Transfer
300
400
500
600
Me
asu
red
He
at
Tra
nsf
er
(kW
)
HTRI w/REFPROP
HTRI w/VMGThermo
Dittus-Boelter
Measured = Predicted
Page 8Supercritical CO2 Power Cycles Symposium September 9-10, 2014
0
100
200
300
0 100 200 300 400 500 600
Me
asu
red
He
at
Tra
nsf
er
(kW
)
Predicted Heat Transfer (kW)
IHX Pressure Drop
1.5
2
2.5
3
Me
asu
red
Pre
ssu
re D
rop
(p
si)
HTRI w/REFPROP
HTRI w/VMGThermo
Friction Factor Analysis
Measured = Predicted
Page 9Supercritical CO2 Power Cycles Symposium September 9-10, 2014
0
0.5
1
1.5
0 0.5 1 1.5 2 2.5 3
Me
asu
red
Pre
ssu
re D
rop
(p
si)
Predicted Pressure Drop (psi)
Precooler Heat Transfer
Page 10Supercritical CO2 Power Cycles Symposium September 9-10, 2014
Precooler Temperature Profile
Page 11Supercritical CO2 Power Cycles Symposium September 9-10, 2014
Precooler Pressure Drop
Page 12Supercritical CO2 Power Cycles Symposium September 9-10, 2014
Difference Between Single Node LMTD Method Predicted
Heat Transfer and Measured Test Data using both Cold
Idle and Full Power Cases as Baselines
IHX Precooler (Series)
Case Cold Idle UA
Full Power
UA
Cold Idle UA
Full Power
UA
Q=UA∆TLM
Page 13Supercritical CO2 Power Cycles Symposium September 9-10, 2014
UA UA UA UA
Cold Idle
----- 77.3% ----- 73.3%
300°F Hold
3.8% 83.9% -29.0% 23.1%
Hot Idle
-35.4% 14.6% -23.0% 33.5%
Full Power
77.5% ----- -42.3% -----
Summary
• HTRI’s Xist® Software does a good job at
predicting the heat transfer and pressure drop
performance of S-CO2 in shell and tube heat
exchangers.
Page 14Supercritical CO2 Power Cycles Symposium September 9-10, 2014
exchangers.
• Methods using average properties such as
LMTD should be avoided, especially at
temperatures and pressures near the critical
point.
