Thermal Performance of the GLAST LAT Tracker

1
Thermal Performance of the GLAST LAT Tracker

• Silvia Rainò
• INFN-Bari
• For the Italian GLAST Tracker Collaboration

2
Outline

• The GLAST (Gamma Ray Large Area Space Telescope)
  Instrument
• Environmental Tests on the LAT (Large Area
  Telescope) Tracker Thermal-Vacuum Test Results
• Dependence of the Silicon Detectors Performance
  on Temperature

3
The GLAST Observatory
Launch Vehicle Delta II 2920-10H Launch
Location Kennedy Space Center Orbit Altitude
575 Km Orbit Inclination 28.5
degrees Orbit Period 95 Minutes Orientation
X to the Sun
4
The LAT Instruments
Tracker
GLAST-LAT Performance
Tracker
Aeff 8000 cm2 (E gt 100 MeV) Angular Res. lt
3.50 _at_ E100 MeV lt 0.150 _at_ Egt10
GeV F.O.V. 2.4 sr Source Loc. lt
0.5 Energy Resolution lt 10 Dead time lt
100 ?s Time Res. 2 ?s Power
500 W
Grid
Thermal Blanket
ACD
DAQ Electronics
Calorimeter
Systems work together to identify and measure the
flux of cosmic gamma rays with energy 20 MeV
-gt300 GeV.
5
The GLAST LAT Tracker

• 16 towers in a 4 ? 4 array
• 19 stacked carbon fiber panels trays
• 18 x,y alternated planes/tower
• W converter (12? 3 r.l., 4? 18 r.l.)
• 2mm gap between x/y oriented SSDs

• 37cm ? 37cm of active cross section
• 83 m2 of Si
• 11500 SSD, 1M channels

6
TKR Pictures
7
Tracker detector construction work-flow
8
Environmental Tests on TKR
Aim of the Environmental Tests on the GLAST TKR
Towers is mainly to verify the capability to
withstand the conditions expected during launch
phases (mechanical tests) and on orbit
(thermal/thermal-vacuum tests)
Trays
Vibration tests thermal cycles
Vibration tests
Towers
Thermal vacuum cycles thermal balance
This presentation focuses on the results of the
THERMAL BALANCE TEST on the EM Tower and on the
expected thermal performance of the GLAST Tracker
9
EM-Tower Thermal-balance test
Temperature measurements Thermocouples ? 100
placed throughout the tower and MGSE Thermistors
16 placed two per each flex cable
Action measure the temperature distribution in
the tower under predicted orbital operating
conditions, by using a combination of 3 tracker
power levels (8,10,12 W) and 3 controlled base
plate temperatures (20, 0, -15C).
Goal Correlate GLAST EM TKR Tower internal
temperatures with those measured by the Tower
cable thermistors and also with those predicted
by the Tower thermal math model.
10
Test Set-up
Guard shield over the the tower MLI blanket
system to prevent heat losses
Thermocouples on the side-walls of the tower
MLI blanket covers the tower
11
in the thermal-vacuum chamber
Thermal vacuum chamber in Alenia AIT - Rome
12
Thermal Balance Test Profile
13
TB Test Results -1
The temperature distribution along the tower has
been studied dividing it into six nodes. This
division allows to create a correlation of the
experimental results with the math model.
Balance condition lt 0.2C/hr Balance duration
?4hr
14
TB Test Results - 2
Thermistors vs. thermocouples
Thermocouples (continuous) and thermistors
(dashed)
15
TB Test Results -3
Average Temperature for each node at the
different balance cases
?T ? 5 ?? 6.5C between top and bottom trays of
the tower
-15 C
0 C
20 C
16
GLAST TKR Silicon Detectors Performance

• Signal/Noise dependence on temperature

17
SSDs performance simulation
A Monte Carlo Code, developed to simulate all the
physical processes that occur in Silicon Strip
detectors, has been used to evaluate the total
number of electron-hole pairs in silicon produced
by a MIP, as a function of temperature.
The number of e-h pairs has been estimated as a
function of temperature. Since Egap slightly
decreases as the temperature increases
the total number of charge carriers in silicon
slightly increases with the temperature.
(more details in F.Loparcos talk yesterday)
18
Pair Production vs Temperature
e-h pairs distributions have been fitted with a
Landau function
19
Noise sources
Thermal noise due to the feedback
resistor i2nf4KT/Rf
S H A P E R
Electronic noise due to the 1st stage of the
pre-amplifier i2na 0 v2na 2.7KT/gmAf/f
Thermal noise due to the bias resistor i2nb4KT/R
b
20
RMS Noise Voltage
At the shaper output, the rms noise voltage is
H2Vpre/V HshaperVout/Vpre
H1Vpre/Id HshaperVout/Vpre
21
Noise temperature dependence
ENC evaluated from the rms noise voltage assuming
a front-end overall gain of 115 mV/fC
22
Signal/Noise vs Temperature
The signal to noise ratio (S/N) has been
evaluated as the ratio between the most
probable/average number of electron-hole pairs
and ENC.
S/N ? 21.5 _at_ 300 K
MIPs crossing 400 ?m thick Si
S/N ? 19 _at_ 300 K
23
Conclusions

• A detailed study of the mechanical and thermal
  performance of the GLAST TKR has been done on an
  Engineering Model Tower before the beginning of
  the flight production started in April 2004.

• The results of the Thermal Balance Test have
  been presented an average temperature gradient
  of 6C between the top and bottom of the tower
  has been found at a nominal tower power level of
  10 W confirming the results of the thermal math
  model

• The dependence of the detectors performance on
  the temperature has been studied, by using a
  Monte Carlo code developed to fully simulate the
  silicon detector performance.

• In order to understand the overall performance
  of the TKR, a complete simulation taking into
  account the strip geometry, the charge sharing,
  the impact angles of particles, . is needed (see
  M.Brigida talk)

24
2007
25
Tracker Trays
26
End of Mission expectations
At the end of mission a total dose of 5 krad (a
factor 5? engineering margin is considered)
the leakage current per strip increases to Il94
nA at T0298 K.
ENC ? 2000 e- _at_ 300K
The S/N ratio decreases from 20 at 250 K to about
4 at 350 K.
27
Simulation Parameters

• Some Parameters used for the SSDs simulation
• Il1.5nA/strip _at_ 296K
• Cd47nF
• Rb40M?
• Rf10G?
• Af3?10-12V2
• gm0.7mS

28
Front-end electronics
Preamplifier
Shaper
Detector