Loop Heat Pipe Startup BehaviorsPipe Startup BehaviorsSchematic of a Loop Heat Pipe Primary Wi k Secondary Wick Wi kWick Vapor Channel Pump Core Vapor Channel Bayonet Reservoir Evaporator

Loop Heat Pipe Startup BehaviorsLoop Heat Pipe Startup Behaviors

Jentung KuNASA Goddard Space Flight Centerp g

Greenbelt, Maryland, USA301-286-3130

[email protected]

46th International Conference on Environmental SystemsVienna Austria July 10 14 2016Vienna, Austria, July 10-14, 2016

Outline• Introduction• Introduction• LHP Operation – Background• LHP Startup Scenarios• Fluid Distribution Between Evaporator and Reservoir• Enhancing Start-up Success• Other Startup Issues• Summary and Conclusions

LHP Startup Behaviors – 2016 Ku 2

Introduction/Background

• LHPs have been used for instrument thermal control on many orbiting spacecraft.

• An LHP must start successfully before it can commence its service.

• The way an LHP starts may affect its subsequent operations.

• The startup of the LHP is one of the most complex transient phenomena in LHP operation.

• This presentation focuses on the issues related to the startup of a single-evaporator LHP servicing a heat source all by itself.


Schematic of a Loop Heat PipePrimary

Wi kSecondary

Wi kWick WickVapor Channel Pump Core

Vapor Channel

Bayonet

Reservoir Evaporator

Vapor LineLiquid Line

Condenser

• The reservoir forms an integral part of the evaporator assembly.• The primary wick with fine pore sizes provides the pumping force.

LHP Startup Behaviors – 2016 Ku

• The secondary wick connects the reservoir and evaporator, supplying liquid.

4

LHP Startup Scenarios

• Temperature Overshoot and Undershoot during Start-up

• Four Start up Scenarios• Four Start-up Scenarios

• Start-up Success

• Effect of Heat Load on Start-up Success

• Flow Reversal


Temperature Overshoot During Start-up

TemperatureOvershoot

No TemperatureOvershoot

Final Tcc

Initial Tcc Initial Tcc

Final Tcc

TemperatureOvershoot

No TemperatureOvershootInitial Tcc

Initial Tcc

Final TccFinal Tcc


Temperature undershoot can be defined in a similar manner.

Four Start-up Scenarios for LHP

• Vapor grooves– Liquid filled: T

empe

ratu

re

TeTccStart-up

Tamb

Tem

pera

ture

TeT

Tamb

St tq

superheat is required for nucleate boilingV

Time

p

Time

TccStart-up

– Vapor presence: instant evaporation

(a) Situation 1 (c) Situation 3

• Liquid core– Liquid filled:

low heat leak

Tem

pera

ture

TeTccStart-up

Tamb

Tem

pera

ture

TeTcc

Tamb

Start-uplow heat leak– Vapor presence:

high heat leakTime Time

Start up


(b) Situation 2 (d) Situation 4

7

Start-up Success

• The beginning of liquid evaporation or nucleate boiling in vapor grooves is characterized by the rise of the vapor line temperature to near the reservoir saturation temperature and th d f th li id li t tthe drop of the liquid line temperature.

• A successful start-up is characterized by: – The vapor line temperature is the same as or close to the reservoir

temperature;– The evaporator temperature is higher than the reservoir

t t b t d t i d b th h t l d d thtemperature by an amount determined by the heat load and the evaporator thermal conductance;

– The liquid line temperature is lower than the reservoir temperature; Temperatures of the reservoir evaporator vapor line and liquid– Temperatures of the reservoir, evaporator, vapor line and liquid line approach their respective steady state temperatures asymptotically.


High and Low Power Start-up

• With high power to the evaporator, liquid in the vapor grooves can be vaporized quickly regardless of the initial two-phase status in the grooves and the evaporator core.status in the grooves and the evaporator core.– The required superheat, if any, can be achieved in a short time.– Within the short time, the total heat leak is small.

• With low power to the evaporator, start-up could be problematic.– Under situation 4, the required superheat for nucleate boiling may

never be achieved.never be achieved.– A reverse flow may occur prior to nucleate boiling.– After the loop starts, a steady state may not be established within

the allowable temperature limit at low powers due to a high heat leak p p gfrom evaporator to CC if the core contains vapor.


Unsuccessful Startup Under Situation 4

CC

Evap

T

mpe

ratu

rep

Loop does not start

Tem

• The vapor grooves are filled with liquid, and a superheat is needed to initiate nucleate boiling

TimeTime

needed to initiate nucleate boiling.• Because of the high heat leak from the evaporator to reservoir,

the required superheated was never attained.


Schematic of LHP–A and Thermocouple Locations

1 2 3 4 5

6 7 8

9 10 11

6 7 8

6,7,8 9,10,1112

13 134

35

36 ReservoirEvaporator

Lin

e

14

13 15

16

1731

32

33

34

1

Vapor Line

Liqu

id

18

19

20222324252628

29

30

Con

dens

er

Con

dens

er 3

212223242526

27

Condenser 2


High Power Startup of LHP-A

• Successful startup with 50W to evaporator 300

301

– Loop started 3 minutes after power application

– 4.5K superheat for nucleate boiling 297

298

299

)

Pump (TC9)

boiling

• This was a situation 3 startup.Th i t t

295

296

297

Tem

pera

ture

(C

Reservoir (TC5)

– The reservoir temperature rose with evaporator because of a heat leak due to heat conduction instead of heat

292

293

294Vapor line (TC15)

pipe effect.291

292

10:15 10:20 10:25 10:30

Time (hr)

Pump liquid inlet (TC36)


Low Power Startup of LHP-A

• Successful startup with 5W to evaporator– It took 45 minutes to initiate

nucleate boiling– 2.5 K superheat for nucleate

boilingboiling– 4K temperature overshoot

Thi it ti 4 t t• This was a situation 4 startup.– Reservoir temperature rose

with evaporator (due to heat pipe effect) prior to nucleatepipe effect) prior to nucleate boiling.


Ad l ti

Low Power Startup of LHP-A

• Adverse elevation -evaporator and reservoir were 690mm above the condenser

P

Thermal Vacuum Test, Chiller @-10°C, -27” Elevation14060333

• Startup with 10W to evaporator

Pump

Rure

(C)

(W)

80

100

12050

40

30

323

303

313

(K

)

– It took 85 minutes to initiate nucleate boiling

– 2.5 K superheat for nucleate boiling

Res.

Vapor Line

Tem

pera

tu

Pow

er

40

60

8030

20

10283

293

Te

mp

era

ture

boiling– 20K temperature overshoot

• This was a situation 4 startup

Liquid Line

Time(Hours)

20:0019:0018:0017:0016:0015:000

200

-10263

273

• This was a situation 4 startup.– Reservoir temperature rose

with evaporator (due to heat pipe effect) prior to nucleate

Figure 2.9 – NRL Nickel Wick LHP Start-Up


pipe effect) prior to nucleate boiling.

14

Flow Reversal during Startup Transient

• Flow reversal during startup typically occurs under Situation 4.– Liquid evaporation takes place at the core of the evaporator.– Vapor flow via the liquid line to the condenserVapor flow via the liquid line to the condenser.

• Flow reversal can last from seconds with high power startup to hours with low power startuphours with low power startup.

• After nucleate boiling, forward flow will be established, and LHP will begin its normal operationLHP will begin its normal operation.


Schematic of LHP-B

16 26 2827

1511

TC 9,10,11TC 6,7,8

29251714

9

10

8

7

EVAPORATOR

13

312319

18 24 30

12

9

6

5

VAPOR LINE

DP

35

37

21

20 22 32 38

423

136

CONDENSER

LIQUID LINECOMPENSATIONCHAMBER

TC 4TC2

TC 1,3,5

AP34

3533

36


Flow Reversal During LHP-B Startup(100 grams/ +6.35mm/ 5W/ 290K)

• Situation 4 startup under an adverse tilt• Flow reversal lasted for 4+ hours with 5W without startup.• Loop started with 100W, then operated at 5W.

35000

40000

310

312NRL LHP 01/08/2001

01/08/2001

20000

25000

30000

304

306

308

op (P

a)

re (K

)

Evap (7)

Liq Line (34)

5000

10000

15000

298

300

302

Pre

ssur

e D

ro

Tem

pera

tur

CC (3)

-5000

0

5000

292

294

296

8:30 9:30 10:30 11:30 12:30 13:30 14:30 15:30100 W

DP

Vap Line (15)

CC Inlet (36)

5 W 5 W


Time (hours)

17

V id f ti i th t t l ff t th

Fluid Distribution in Evaporator and Reservoir

• Vapor void fraction in the evaporator core strongly affects the LHP startup and low power operation.

• Vapor void fraction depends upon the fluid distribution in the evaporator and reservoir.

• Factors affecting the fluid distribution– Fluid inventory– Pre-conditioning of the loop prior to startup– Body forces

• Evaporator/reservoir assembly design• Tilt between evaporator and reservoirp• Elevation between evaporator and condenser

• Startup is affected by combinations of factors.


p y

18

Evaporator Assembly and Gravity Effect on Fluid Distribution

LiquidVapor

Liquid

fromCondenser Liquid

Evaporator

Gravity

Vapor

Reservoir

Liquid

from

Vapor

LiquidGravity

Condenser Reservoir Evaporator


LHP-B Startup Tests

• Fluid inventories: 83 grams, 100 grams, and 113 grams• Tilts: +6.35mm, 0 mm, and -6.35mm (evaporator end to reservoir end)• Successful startups with 100W or higher under all conditions

St t hi hl d d tilt d i t ith 100W• Startup highly depends on tilts and inventory with

LHP-B Startup - Reverse Flow(100 grams/ 0mm/ 50W/ 270K)

• Situation 4 startup• Flow reversal lasted for 15 minutes with 50W!• 20K temperature overshoot

310

315NRL LHP 01/18/2001

01/18/01

295

300

305

K)

Evap (7)

Vap Line (15)

280

285

290

Tem

pera

ture

(K

CC (3)

CC Inlet (36)

Liq Line (34)

265

270

275

280

Liq Line (33)

50 WATTS0W

q ( )


2657:30 8:00 8:30 9:00 9:30 10:00

Time (hours)

50 WATTS0W

21

Enhancing Start-up Success

• Superheat is required for nucleate boiling for Situation 3 and Situation 4 startups– Situation 3: Loop will start, but may take a long time with low powers.Situation 3: Loop will start, but may take a long time with low powers.– Situation 4: Loop may not start with low powers.

• Methods to enhance startup successMethods to enhance startup success– Start-up heater– Thermoelectric converter (TEC)


U t t d h t l li d th

Startup Heaters

• Use a concentrated heat source over a localized area on the evaporator.

• The high heat flux will quickly raise the temperature of liquid in the vicinity of the heater while minimizing the heat leak to the reservoir.

• Once nucleate boiling starts and first bubbles are generated, no superheat is required for liquid evaporation.

• The startup heater has proven to be very effective in enhancing the startup success.p

• Many LHPs in flight applications employ such a device because of its simplicity in design and ease in implementation.


of its simplicity in design and ease in implementation.

23

TEC and Thermal Strap

Q

QTEC H

QTEC, L

QTEC, AppTECTTEC, L

Thermal StrapQTEC L

QTEC, H

QTEC, AppTEC

TTEC, H

Thermal Strap

-Qsub Qleak

QTEC, H

CC, Tset Evap, TE-Qsub Qleak

TEC, L

CC, Tset Evap, TE

• Heat Flow When TEC Is Cooling the Reservoir

• Heat Flow When TEC Is Heating the Reservoir


Enhancing LHP Startup Using Thermoelectric Converter (TEC)Situation 4 startup

• Without TEC (Figure A)– CC temperature rises with evaporator temperature due to heat leaks.– Required superheat may never be attained at low powers.

• With TEC (Figures B and C)– TEC can maintain a constant CC temperature to achieve the required

superheat, resulting in a successful start-up.TEC l l th CC t t th i d h t– TEC can also cool the CC to create the required superheat.

– Startup heaters can be eliminated.

e Evape

CC

EvapTem

pera

ture

Tem

pera

ture

T

Tem

pera

ture

CC

Evap

TCC

EvapTem

pera

ture

T

Time

Evap

Loop does not start

T

Loop does not start

T

Time

Loop starts

Time

Loop starts

Fi B Fi CFigure A Figure B Figure C

25

LHP-C with TEC and Startup Heaters

• TEC was installed on evaporator and connected to CC via a thermal strap.

• An electric heater was also installed on evaporator to serve asAn electric heater was also installed on evaporator to serve as the startup heater

101112141516 27123

713

Vapor Line CCEvaporator

32

Fill TubeAluminum Saddle

Compensation

Chamber

28

1729

18

192030

456

89

36 Ambient33 34 35

Thermal Mass Vapor Line

Evaporator

Copper Strap

TEC Saddle

Slot for TEC

21 22 23 24 25 26

Liquid Line31

Liquid Line


LHP-C Startup with TEC and Startup Heater

• TEC provided required heating and cooling during startup to maintain the CC set point.

• By comparison, a higher temperature overshoot when an electric heater was used for CC temperature control.

12/14/2009; 253K sink; 350 g mass;TEC set point@303K on the CC - TC2; PID 12/16/2009; 253K sink; 350 g mass;EH set point@303K on the CC - TC02; PID

25

30

35

300

305

310

)

12/14/2009; 253K sink; 350 g mass;TEC set point@303K on the CC TC2; PID

Evap (5)CC (2)

40

50

300

310

320

)

Evap (5)

CC (2)

5

10

15

20

280

285

290

295

Po

wer

(W)

Tem

per

atu

re (K

)

Evap Power

CC In (27)

Vap Line (11)

20

30

280

290

300

Pow

er (W

)

Tem

pera

ture

(K)

E P

CC In (27)

Vap Line (11)

-5

0

5

270

275

280

9:45 10:00 10:15 10:30

Time (HH:MM)

TEC Power

0

10

260

270

14:00 14:15 14:30 14:45 15:00

Time (HH:MM)

Evap Power

EH Power


( ) ( )

• No heat load to the evaporator TEC was used to cool the reservoir

Using TEC to Enhance Start-up SuccessNo heat load to the evaporator. TEC was used to cool the reservoir.

• The loop started at 8:17 with a superheat of 2K.• The heat input to the evaporators came from the power applied to the

TECs, and the heat pumped out of the reservoirs.TECs, and the heat pumped out of the reservoirs.• Once started, the loop continued to operate. Additional heat came from the

sensible heat released from the thermal mass attached to the evaporatorStart-up Test August 12, 2009

50

60

276

278

280

Start up Test August 12, 2009

Mass

30

40

272

274

276

Pow

er (

W)

Tem

pera

ture

(K

)

Vapor Line

0

10

20

266

268

270

CC

Evaporator

Liquid Line


02668:00 8:15 8:30 8:45 9:00 9:15

Time (HH:MM)

Other Start-up Issues

• Pressure Spike

• Pressure Surge• Pressure Surge

• Reservoir Temperature Undershoot

• Repeated Cycles of Loop Start-up and Shutdown


Pressure Spike

• The required superheat for nucleate boiling can be higher than 10K.

• Right after nucleate boiling, the vapor bubble will absorb the sensible heat stored in the superheated liquid and grow rapidly.

• The growth of the vapor bubble is similar to an explosion. • Experimental data shows that the pressure differential across the p p

evaporator can be as high as 45 kPa. • Such a high pressure drop may exceed the capillary limit of the

primary wick and cause the vapor to penetrate the wick to reach p y p pthe evaporator core.

• However, the high pressure drop only lasted for fractions of a second.

• Because of the short duration of the pressure spike and the ability of the LHP to tolerate a vapor bubble in the evaporator core, no LHP deprime due to the pressure spike has been observed.


p p p

30

Aft th b ili i i i li id i th li i t i t

Pressure Surge

• After the boiling incipience, liquid in the vapor line is swept into the condenser.

• Liquid moves toward the reservoir at the same volumetric flow t th i b i t d i thrate as the vapor is being generated in the vapor grooves.

• The liquid mass flow rate along the condenser and liquid line can be two orders of magnitude higher than its steady state value at th h t l d d i th l ti f th LHPthe same heat load during the normal operation of the LHP.

• A high flow rate induces a surge of the pressure drop that is imposed on the primary wick until vapor reaches the condenser.

• The magnitude and duration of the pressure surge depend on the working fluid, saturation temperature, heat load, volume of the vapor line and vapor grooves, and initial vapor line temperature.

• The pressure surge is more severe at a low reservoir temperature.• An LHP can usually sustain the pressure surge without any

problem due to its high capillary capability.


Repeated Cycles of Loop Startup and Shutdown

• When a severe reservoir temperature undershoot happens, the reservoir control heater will be turned on.

• If the heater power is so large that it raises the reservoir temperature faster than the evaporator can catch up, the loop will be flooded with liquid again by the time the reservoirwill be flooded with liquid again by the time the reservoir reaches its set point temperature.

• The re start will follow the same process as the previous• The re-start will follow the same process as the previous startup. In some cases, this leads to repeated startup and shutdown cycles.


C t l l d th th l t i t i it t i t

Repeated Startup/Shutdown Cycles in LHP-C

• Control sensor was placed on the thermal mass to maintain its set point at 313K.

• Control heater (electrical) was attached to the CC.• Repeated startup/shutdown cycles with 10W and 20W to thermal mass• Repeated startup/shutdown cycles with 10W and 20W to thermal mass.• Successful startup with 40W to thermal mass.

140330

12/18/2009; 253K sink; 350 g mass; Pre-heating CC; EH set point@313K on the TM - TC33; PID

100

120

310

320

Evap (5)TM (33)

40

60

80

290

300

Po

wer

(W)

Tem

pera

ture

(K)

CC (2)

CC In (27)

Vap Line (11)

0

20

270

280

Evap Power

EH Power


-202608:30 8:45 9:00 9:15 9:30 9:45 10:00 10:15 10:30 10:45

Time (HH:MM)

• LHP startup is one of most complex transient phenomena

Summary and Conclusions• LHP startup is one of most complex transient phenomena.• There are four possible startup scenarios, which are determined by

the initial fluid distribution between evaporator and CC.Se eral factors affect fl id distrib tion bet een e aporator and CC• Several factors affect fluid distribution between evaporator and CC.– Fluid inventory– Pre-conditioning

B d f it– Body forces, e.g. gravity• Evaporator/CC assembly design• Tilt between evaporator and reservoir

El ti b t t d d• Elevation between evaporator and condenser• Startup success is a function of startup scenario, power to

evaporator, and how the CC temperature is controlled.• Using a startup heater or a thermoelectric converter can greatly

enhance startup success.• Repeated startup and shutdown cycles can happen. This can be


avoided or mitigated by using a smaller increments for reservoir temperature rise.

34

Loop Heat Pipe Startup BehaviorsPipe Startup BehaviorsSchematic of a Loop Heat Pipe Primary Wi k Secondary Wick Wi kWick Vapor Channel Pump Core Vapor Channel Bayonet Reservoir Evaporator

Documents