16th Thermal and ECLS Workshop - ESTEC 1

Thermal Analysis of Planck HFI

16th European Workshop on Thermal and ECLS SoftwareESTEC, 22-23 October 2002

Jayne FeredayRuther ford Appleton Laboratory

j .fereday@rl.ac.uk

16th Thermal and ECLS Workshop - ESTEC 2

SUMMARY

• Planck mission overview

• Thermal design concept

• Systematic effects

• Global thermal model

• Radiative transfer model

• Thermal analysis issues

16th Thermal and ECLS Workshop - ESTEC 3

3rd medium-sized mission of ESA’s Horizon 2000 scientific programmeTesting theories of the early universe and the origin of cosmic structure Measuring anisotropies in the Cosmic Microwave Background - ∆T/T ~ 2 x 10-6

PLANCK MISSION OVERVIEW

LFI (30 - 100GHz)

Tuned radio receiver array at 20K

HFI (100 - 850GHz)

Bolometer arrays at 100mK

16th Thermal and ECLS Workshop - ESTEC 4

Passive cooling to 50/60K• 3 v-groove shields to maximise radiation to space

JPL sorption cooler to ~20K• cools LFI and HFI precooling stage• thermally sunk to v-groove shields

RAL 4K JT cooler• precooling at 18K

Dilution refr igerator to 100mK• precooling at 1.6K

THERMAL DESIGN CONCEPT (1)

16th Thermal and ECLS Workshop - ESTEC 5

0.1K stage

1.6K stage

4K stage

18K plate

GFRP struts

THERMAL DESIGN CONCEPT (2)

LFI

HFI

16th Thermal and ECLS Workshop - ESTEC 6

Work done as part of ESA’s Systematic Effects Working Group (SEWG)

Observations must be corrected for the effects of detector noise and layers of intervening material

Bolometer performance highly dependent on absolute temperature and temperature stability

SYSTEMATIC EFFECTS

⇒ Stringent stability requirements (e.g. bolometers < 19 nK.Hz-0.5)

⇒ System level analysis needed – detailedunderstanding of fluctuations

⇒ Interaction between instruments, spacecraft and coolers is complex

⇒ End to end modelling vital

⇒ Global thermal model created

⇒ Used to assess stability issues by providing full feedback loops to all spacecraft elements

16th Thermal and ECLS Workshop - ESTEC 7

GLOBAL THERMAL MODEL

SVM (Alcatel)

LFI (LFI)

Waveguides and struts

HFI (RAL)

Harness

Solar array

Cooler interfaces

Shields

Space loads

16th Thermal and ECLS Workshop - ESTEC 8

Operational environment• L2 position and orbit - scanning/spin

Coolers• Sorption cycling• JT instabilities• Effect of JT getter temperature being raised

General• Electronics switching • Influence of 3rd V-groove temperature on LFI• Influence of LFI temperature on HFI

Combined effects

Stability of sorption cooler identified as largest source

SOURCES OF INSTABILITY

16th Thermal and ECLS Workshop - ESTEC 9

SORPTION SIMULATIONS RESULTS

17.48

17.5

17.52

17.54

17.56

17.58

17.6

10000 11000 12000 13000

Time, s

Tem

p, K

4.5882

4.58825

4.5883

4.58835

4.5884

4.58845

4.5885

10000 11000 12000 13000

Time, s

Tem

p H

FI,

K

50.3739

50.374

50.3741

50.3742

50.3743

50.3744

50.3745

Tem

p s

hie

ld, K

4.5882

4.58825

4.5883

4.58835

4.5884

4.58845

4.5885

10000 11000 12000 13000

Time, s

Tem

p H

FI,

K

50.3739

50.374

50.3741

50.3742

50.3743

50.3744

50.3745

Tem

p s

hie

ld, K

4K

Telescope

Sorption cooler cold end

16th Thermal and ECLS Workshop - ESTEC 10

AIM: To produce a modelled detector output, including noise,to be used for the “ removal of systematic effects”

INPUTS: Simulated sky Time Ordered Data (TOD)TOD of the thermal variations of critical nodes in the HFI.

Model received showing bolometer performance against thermal behaviour• supplied in MathCAD format• produces bolometer output power, taking into account system noise• originally took constant temperature as an input• modified to accept arrays of thermal fluctuation data

Excel used originally• MathCAD add-in• macros used to feed thermal model output data into model• run time and file size became restrictive very quickly

MathCAD model re-written in C++C++ command line model now used that enables sky data and thermal data files to be input in batches

RADIATIVE TRANSFER MODEL

16th Thermal and ECLS Workshop - ESTEC 11

TRANSFER MODEL OUTPUT

Detected Model Sky at 100GHz with Stable4K Power on Detector from Temp Fluctuationsand ‘Constant’ Sky

16th Thermal and ECLS Workshop - ESTEC 12

Detected Model Sky at 100GHz with TemperatureFluctuations

TRANSFER MODEL OUTPUT

16th Thermal and ECLS Workshop - ESTEC 13

Run time• Sorption cooler data sets 40,000s • 1/3s data output required• Convergence criteria must be high• Constants material properties used where possible• Nodes set to boundary temperatures where possible

Format compatibility• Thermal model in ESATAN – output to .csv files• Sky TOD in FITS format• Bolometer model in MathCAD

Configuration control• Several different models required – cooldown, cooler failure cases, stability cases• $INCLUDE function used whenever possible to ensure model update is easy – ‘plug in’ modules

THERMAL ANALYSIS ISSUES