Measurements of temperature on LHC thermal models

Measurements of temperatureon LHC thermal models

Christine Darve1, Juan Casas2, Moyses Kuchnir1

1: Fermi National Accelerator Laboratory, Batavia, IL, USA2: CERN, European Laboratory for Particle Physics, Geneva, CH

Ch. D. 12/05/00 Measurements of temperature on LHC thermal models 2

Measurements of temperatureon LHC thermal models

Z Thermal measurements on modelsZ Presentation of both projectsZ Sensors implementationZ Results

Z Use of cryogenic thermometers at Fermilab and at CERN

Z Uncertainty evaluations


Introduction to LHC thermal models

Inner Triplet heat exchanger test unit (US-IT-HXTU)y Validation of the Inner Triplet cooling scheme

ü Heat transfer based on the exchange between the two-phase saturated He IIused to extract heat loads generated in the stagnant pressurized He II bath.

ü Extreme heat loads due to the detector proximity : nominal Q’=7 W/m.

y Investigation in a linear model-based predictive controlü Reduce requirements on temperature sensor, cryogenic sys. performanceü Self-regulating process

y Related project: Small scale heat exchanger testü Material property measurements: Kapitza resistance.ü Results used to estimate the wetted area of the full-scale model.


From the LHC Inner Triplet to the Inner Triplet heatexchanger test unit

IT to Thermal modelIT to Thermal model

••Four modulesFour modules

••Magnet simulatorsMagnet simulators

••Full helium capacityFull helium capacity


Inner triplet heat exchanger test unitModules

8 cryogenic thermometers

2 heaters


Inner triplet heat exchanger test unitPiping system

Heat exchanger tube

Shield cooling pipes Sat He II supplyMagnet simulator

Pressurized He circuit


Inner triplet heat exchanger test unit

Instrumentation port flange

Feed-box----

– ----Turnaround


Inner triplet heat exchanger test unit


Thermal measurements on the US-IT-HXTU

j Why do we perform thermal measurements?

j To measure the temperature rise along the He II heat load path, for differentLHC heat load scenarios-> maximum: heat load, Tsat.

j To understand the behavior of He II (co-current two-phase flow, wetted area).j To calibrate and measure the performance of the HX tube.j To validate the theoretical model->ultimate LHC condition (472W).

ò Function of temperature measurements

ò Display the transport of the heat load from pressurized to saturated He II.ò Indicator to check the evolution towards steady-state conditions.ò Variable parameter, Tsat, for various measurements of the HX tube

performance.



ñ How are they implemented ?

Glued to printed circuit board (PCB) card.

Thermometers immersed in pressurized He II bath along the HX tube (pressurized side) and magnet simulator pipes.

Thermometer wires routed through a 3 m long feedthrough from the pressurized He II bath to the room temperature.

Measurement of the saturated temperature provided from the pressure measurement.

To reduce air leaks through the instrumentation feedthrough(subatmospheric circuit).

PressurizedPressurizedHe IIHe IItemperaturetemperature

SaturatedSaturatedHe IIHe IItemperaturetemperature



Pressurized He IIPressurized He II

Connector @Connector @Room temperatureRoom temperature

Outer shell of the HX

HXSaturated He IISaturated He II



From JT valve

Tsat9214

TT4121 TT4151 TT5121

TT4121B TT4151B TT5121B RB

TT4221 TT4251 TT5221

TT4221B TT4251B TT5221B

Y4203

DT1

DT3

DT4

DT2

DT5DT6

DT1: from the Module thermal center, to the module end within the pressurized He II

DT2: within the connecting pipe

DT3: between connecting pipe and He II HX

DT4: within the pressurized He II side of He II HX

DT5: across the He II heat exchanger wall

DT6: due to the vapor pressure drop.Vapor V= 448 cm/sec


0

100

200

300

400

500

600

1.5 1.7 1.9 2.1Temperature [K]

X(T

c) -

X(T

w)

[(W

.cm

-2)3.

4 ]


ñ Heat transfer into and through pressurized He II (DT 1 - 4)

One-dimension model In steady-state, heat flux Q’ is given by:

( ) ( ) 29.0

'

−

⋅=l

TwXTcXSQ

( )( )5.2)16.2(31520)( TeTX −⋅−−⋅=

He II

l

Tw Tc

Q

S


US-IT-HXTU- Small scale HX

z Kapitza Resistance and DT 5

Rkapitzaα2

Ckapitza S.

βe

S Ccu.

Rth=2.Rkapitza+Rcu=α(1/Tpres3)+β

Rth=(Tpres-Tsat)/Qelec


0

10

20

30

40

50

60

70

0 2 4 6 8 10Q (w att)

Tpr

es -

Tsa

t (m

K)

1.75K

1.80K

1.85K

1.90K

1.95K

2.00K

Thermal measurements on the small-scale heatexchanger test unit - results

2

4

6

8

10

12

0.12 0.14 0.16 0.18 0.20Tsat -3 (K-3)

Rth

(10

-3 K

. W-1)

#1

#2

#3

OFHC-US-IT-HXTU

Bronze (95%Cu-5% Sn)

OFHC+ HCl

#1 - OFHC #2 – OFHC + HCl #3 - BronzeCharacteristics

OD/ ID (mm) 97/86 97/86 123/101Wall thickness (mm) 0.7 0.7 0.5Corrugation depth (mm) 5 5 11Corrugation pitch (mm) 12.4 12.4 11.7Surface (cm2) for one side 416 416 978Shape of the corrugated pipe Helical Helical BellowsSurface treatment None Hydrochloric acid None

ResultsCKapitza (W ⋅ K-4 ⋅ m-2) 893 1138 565Kapitza conductance @ 1.85 K (W ⋅ K-1 ⋅ cm-2) 0.565 0.72 0.357Thermal conductivity @ 1.85 K (W ⋅ K-1 ⋅ m-1) 88 88 2.4Relative performance Ref. 27% -37%


US-IT-HXTU - Nominal Condition: 248 W


US-IT-HXTU - Ultimate Condition: 315 W


8

18

28

38

48

58

68

4 5 6 7 8 9 10 11Power applied (W/m)

T-Ts

at (m

K)

HX side @ middle of the module

HX side @connecting pipe

Magnet side @ connecting pipe

Magnet simulator pipe @ middle

US-IT-HXTU - Some results

C/C: The difference of temperature on the heat exchanger interface < 50 mK.The thermal gradient within the heat exchanger pipe is still to be measured in december at CERN.

Ultimate condition @ 315 W

P=24.1 mbar

Tsat Tsat ~ 1.90 K~ 1.90 K

P~21 mbarP~21 mbar



0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

90.0

DT1 DT2 DT3 DT4 DT5 DT6 TotalDT

Predicted Measured

Example for Tsat= 1.915 K , Q’tot= 315 W

Heat loads on the US-IT-HXTU extrapolated from the heat load in the Inner Triplet at IP5 or IP1,by Tom Peterson.

Equations:

dT/dx=-f(T) qm

=>∆T=q3L/1200


Cryostat thermal model (CTM)

z Measurements of the LHC cryostat dipole performance+ Heat loads to the actively cooled shield and to the dummy cold mass.+ Components performance:

õ Multi-Layer Insulation,õ support posts,õ beam screen,õ cryogenic thermometers.

z Conditions under investigationõ Steady-state and transient modes for the LHC conditionsõ Insulation vacuum degradation

z Adoption of an actively cooled screen @ 5-10 K (CTM3)?

Introduction to LHC thermal models


Thermal measurements on the CTM



View of the radiation screen View of the dummy cold mass


960

MLI

Radiation

cold mass

Vacum vessel

Thermal shield

Support post

890

Carbon compositespacer

supply line 50- 75 K

5- 10 K supply line

carbon composite spacer

50-75 KReturn line

5-10 KReturn line

Simulatedcold mass

screen


LHC accelerator cross-section CTM3 cross section




z Performance of the thermal shield (50-75K) and radiation screen(5-10K)The temperature is measured at each extremity of the helium pipes

z Heat load to the dummy cold mass Heat load to the dummy cold mass is measured with the boil-off method


m’ = mass-flow, ∆H = enthalpy difference

Q’ = m’· ∆H

m’ = mass-flow, L = latent heat of vaporization

Q1.9 K’ = m’ · L


Implementation of sensors on the CTM

On the cold mass

On a pipe

On the shield


Implementation of sensors on the CTM

In the MLI

Thermalization of theinstrumentation wires


Some results - CTM2

Total heat load measured:

0

10

20

30

40

50

50 70 90 110 130 150Temperature at the rear section of the Thermal Shield [K]

Q t

s [

W]

Line E

Line F

Total

@ 50-75 K / Line E+Ferror: ± 5%

@ 5-20 K / Line C+Derror: ± 4.7%

@ 1.9 Kerror: ± 2.5%

Heat inleak[W/m]

measured calculated measured calculated measured calculatedCTM1 4.78 4.58 0.23 0.24 0.18 0.19CTM2 4.32 4.12 0.48 0.33 0.16 (1) 0.12

The contribution of the feed box on the measured He in-leak at 1.9 K is estimated to 390 mW.

@5-10K@5-10K


Influence of the insulation vacuum degradation

0

10

20

30

40

50

60

70

80

90

100

1.E-03 1.E-02 1.E-01

Pressure [Pa]

CTM1

CTM2

CTM3

CTM3 - Some results

= CTM3, dummy cold mass experienced He leaks -> results difficult to interpret.

= MLI qualification on an horizontal cryostat was performed. = Choice of the shield assembly: welded to the extruded pipe

C/C: Since the cost of a radiation screen material is dominant and even if a better performancewith a 5K cryostat was measured, the LHC cryostat will only consider to wrap 10 layers of MLIaround the cold mass. No rigid radiation screen (5-10 K) will be used in the LHC accelerator.

Averaged heat loads (Watts): Insulation vacuum 7.5⋅10-5 Pa

Component Average Temperatures K 50-75 K 5-10 K 1.9 K

Supply and return boxes 0.300Thermal shield 68 43.2Radiation screen 8.7 3.1Cold-mass 1.8 0.441Support System 14.9 1.1 0.129Total CTM Cryostat 28.3 2.0 0.012

w/o actively cooled shield @5-10K

w/ partial cooled shield @5-10K

w/ actively cooled shield @5-10K


Some cryogenic thermometers used atFermilab - US-IT-HXTU

è Thermometer immersed into He II bathè Commercial sensorsè Printed circuit boardè Calibration facilityè Chebychev polynom

k Pro and cons+ Cryogen true value- Implementation- Feedthrough&Connector


Some cryogenic thermometers used atFermilab

Allen-Bradley

Cernox+copperCernox on cards


Calibration of the thermometer

z Fermilab facility


è Under vacuum: Industrial-type cryogenic thermometers with built-inheat interceptè Sensor implementationè Thermometric blocè Accessories

Copper blocsRadiation protectionThermalization foil

k Pro and cons+ Easy-to-use- Thermal anchoring

Some cryogenic thermometers used atCERN - CTM


z Layout of the cryogenic thermometer

Some cryogenic thermometers used at CERN


Some cryogenic thermometers used at CERN

// foil


Uncertainty evaluations

õõ Measurement stringMeasurement stringõ Cryogenic thermometerõ wire, conditionerõ control system and acquisitionõ calibration fit

åå Environmental factorEnvironmental factorå Thermo-cyclingå Moistureå Magnetic fieldå Irradiation

Error± 5 mK at 1.8 K

Systematic errorSystematic error•Calibration: fit

•Effect of the liquid level gauge

Statistical errorStatistical error•Stability of the bath temperature

•Stability of (Tpres-Tsat)


Overheating

Check US-IT-HXTU thermal measurements:õ Thermometers calibrated with a 0.2 µA current but used with a 1 µA.õ Implementation on the PCB

ò High resistance at low temperature (40 kΩ at 1.8 K)

ò Large dispersion of resistance (40-12 kΩ at 1.8K)


Overheating - Resistance influence

To = Temperature out of the calibration fit relative to a current of 0.2 µA used for the calibration data.

If : R=32 KΩ @ 1.8 K I=1 µ A then Error = 0.1%

If : R=12 KΩ @ 1.8 K I=1 µ A

then Error = 0.02%

Influence of the self-heating R=39kohm @1.8K, I=0.2microA

0

2

4

6

8

10

12

14

16

18

0 1 2 3 4 5 6 7 8 9 10

Current [microAmp]

Err

or

(T-T

o/T

o) [

%]

T1.5

T1.6

T1.7

T1.8

T1.9

T2

T2.1

T2.2

T2.3

T3

T4.1

T4.3

Influence of the self-heating R=12kohm @1.8K, I=0.2microA

0

1

2

3

4

5

6

7

8

0 1 2 3 4 5 6 7 8 9 10

Current [microAmp]

Err

or

(T-T

o/T

o)

[%]

T1.4

T1.5

T1.6

T1.7

T1.8

T1.9

T2

T2.1

T2.2

T2.3

T3

T4.1

T4.3


T=1.

5K

T=1.

6K

T=1.

7K

T=1.

8K

T=1.

9K

T=2K

T=2.

1K

T=2.

2K

T=2.

3K

0.5uA

10uA0

4

8

12

16

(T-T

o/T

o)

[%]

Fit = simple polynomial

T=1

.4 K

T=1

.5 K

T=1

.6 K

T=1

.7 K

T=1

.8 K

T=1

.9 K

T=2

.0 K

T=2

.1 K

T=2

.2 K

T=2

.3 K

0.5microAmp

10microAmp0

4

8

12

16

(T-T

o)/T

[%]

Fit = Chebychev polynomial 0.5microAmp

1microAmp

2microAmp

5microAmp

10microAmp

Overheating - fitting expression

T= Σ Ai·Xi

Z= log (R)

X=((Z-Zl)-(Zu-Z))/(Zu-Zl)

T=Σ Ai · COS (i· ARCCOS(X))


To=1

.4 K

To=1

.6 K

To=1

.8 K

To=2

.0 K

To=2

.2 K

To=

2.4

KTo

=2.6

KTo

=2.8

K

To=3

.0 K

To=3

.2 K

0.5 microAmp

20 microAmp0

100

200

300

400

500

600(T

-To)

[m

K]

Temperature drift vs. current

0.5 microAmp

1 microAmp

2 microAmp

5 microAmp

10 microAmp

20 microAmp

Overheating - Current influence

Calibration fit:Calibration fit:

w/ R@ 0.2 w/ R@ 0.2 microAmpmicroAmp

T=T=CherbCherb(R0.2)(R0.2)

T: R w/ 0.5- 20 microAmp

To: R w/ 0.2 microAmp


To=1

.4 K

To=1

.6 K

To=1

.8 K

To=2

.0 K

To=2

.2 K

To=

2.4

K

To=2

.6 K

To=2

.8 K

To=3

.0 K

To=3

.2 K

0.5 microAmp

5 microAmp

0

20

40

60

80

100

120

140

(T-T

o)

[mK

]

Drift of temperature vs. current

0.5 microAmp

1 microAmp

2 microAmp

5 microAmp

Overheating

Drift =1.3 Drift =1.3 mKmK

for To=1.8 Kfor To=1.8 K

I=1 I=1 microAmpmicroAmp


Conclusion

í Measurement of thermal model performancesí Temperatures rise in He II => US-IT-HXTUí Heat loads => CTM

í Improvement of techniques for measurements of temperature

í Better reliability

Documents

Measurements of temperature on LHC thermal models