component test h4irrad 15 th november 2011 g. spiezia, p. peronnard, g. foucard, s. danzeca, p....
TRANSCRIPT
COMPONENT TESTH4IRRAD15TH NOVEMBER 2011
G. Spiezia, P. Peronnard, G. Foucard, S. Danzeca, P. Gander, E. Fadakis (EN/STI/ECE)
Outline
Component description Goal of the test Setup Radiation conditions and dosimetry Results Conclusions
Components under test
DUT Type-Tec
Voltage Parameters Nb
partsTemperature ºC
AD822 Jfet-Amp +/-5 V
Current bias/OffsetLatch-up
2 Room
AD8220 DiffAmp +/-5 V OffsetLatch-up 2 Room
LTC2052 OpAmp-CMOS +/-5 V Offset
Latch-up 2 Room
XC95108 CPLD +5 V Latch-up 3 70
XC95108 CPLD +5 V Latch-up 3 Room
MAX11046 ADC +5 V Latch-up 3 75
PCM1702k DAC +/-5 V Latch-up 3 75
Motivation Compare test at H4IRRAD and PSI for a
better understanding of the ELDRS effect and destructive events
Some devices taken from the TRAD report AD822
Known to be sensitive to ELDRS To be tested at PSI at higher dose rate
AD8220 Tested at PSI and failed at 40 Gy Comparison with H4IRRAD
LTC2052 LET threshold for latch-up at <10 MeV.cm2/mg.,
sth=10-6 cm2 ssat=10-4
Test at H4IRRAD and PSI to verify if latch-up are induced
Motivation CPLD XC95108
Tested at PSI HI and showed latch-up Comparison with H4IRRAD. Test at room
temperature and heated Implementation of a shift registers
MAX11046-16 bit ADC Tested at PSI. No Latch-up Comparison with H4IRRAD. Test at room
temperature and heated PCM1702K- 20 bit DAC
Possible use of the component Test at H4IRRAD to verify destructive events.
Test at room temperature and heated Test at PSI to verify the TID and the cross
section for Soft and destructive SEE
SETUP Monitoring of the parameter drift
Amplifier parameters, ADC reference, and current consumption measured with a multimeter. Slow acquisition
Latch-up test Separate anti latch-up circuit for each positive and
negative Voltage supply Unique anti latch-up circuit for the 3 ADC MAX11046
and the amplifiers 16 channels acquired at 1 MS/s for the latch-up detection
H4IRRAD – Beam conditions H4IRRAD Internal location Mixed field and high energy hadrons
±5 cm
±10 cm
@ M. Calviani
H4IRRAD – Beam conditions H4IRRAD Internal location
1.2e9 pot per cycle Cycle ~ 45 s Bunch length 5 s TID/Pot ~ 3e-12 [Gy/pot] TID per cycle ~ 3.6 mGy TID during extraction (5s)=2.6Gy/h Average Dose rate (3.6 mGy over 45 sec)
= 0.3 Gy/h Recommended dose rate to evaluate ELDR
H4IRRAD – Dosimetry FLUKA calculation (TID in air)
High gradient on the target line
TID 35 Gy, HEH 4-5x1010 cm-2
H4IRRAD – Dosimetry for TID FLUKA calculation (TID in air)
High gradient on the target line Score position to be reviewed
Radfets on the RADMON (2 oxide Thickness)
FLUKA [Gy Air ]±50% 400nm [Gy -Si] 100nm [Gy -Si] TID ±50%
19/06-27/06 70 30 46
19/06-25/07 160 70 100 100±50
19/06-21/07 130 60 85 85±40
SLOT 319/10-7/11 32 25(TO be Verified) 35 35±17
H4IRRAD – Dosimetry for TID FLUKA calculation (TID in air)
High gradient on the target line Score position to be reviewed
Radfets on the RADMON (2 oxide Thickness) HEH fluence is ~5x1010cm-2
POT= 1.12e+013 Radmon 1 Radmon 2 Radmon 3 Radmon 4 Radmon 5
Fluka Fluence HEH [cm-2] 5,62E+09 1,20E+10 2,62E+10 6,12E+10 8,35E+09
Fluence with SEU/TID ±11% 3,50E+09 1,07E+10 2,59E+10 4,78E+10 5,66E+09
Fluence with SEU/POT ±16% 0,00E+00 0,00E+00 3,20E+10 5,07E+10 0,00E+00
% Difference Fluka vs SEU/TID
38 11 1 22 32
% Difference Fluka vs SEU/POT
- - -22 17 -
Results
DUT TypeTechnology
PSI, p+, >250Gy/h *
H4IRRAD0.3 Gy/h *
H4IRRADHEH [cm-
2]**
Temperature
AD822 Jfet-Amp No tested 35-ok 0 latch Room
AD8220 DiffAmp 40 Gy - broken 35-ok 0 latch Room
LTC2052 OpAmp-CMOS
Not tested yet 35-ok 0 latch Room
XC95108 CPLD 35 – ok 0 latch Room
XC95108 CPLD 35 - ok 2 latch 70 C
MAX11046
ADC- BiCMOS
200 Gy-Broken 35 – ok 0 latch 80 C
PCM1702k DAC 0 latch 70 C
•*Max TID without failure or TID at which measurements are out of spec•**Fluence/ nb latch
Results Notes
Signals are noisy due to long cables Decimation is applied for analysisA
mplit
ude [
V]
October November
Noise
Expected level
Results – AD822
H4IRRAD 3 Gy/h(per
extraction) 0.3Gy/h (average) 35 Gy Within
specification but TID is low
PSI test to be done
NSREC2009: Ibias degradation at about 50-70 Gy at 0.01 rad/s0.3 Gy/h
~35 Gy (100nm radfet measurement)
Ibias current
Results – AD8220 PSI
250 Gy/h 230 MeV p+ 30 Gy –failure Ibias very high as
soon as the beam is on-1nA
H4IRRAD 3 Gy/h(per
extraction) 0.3Gy/h
(average)
Within specification
Differential amplifier; output signal
October November
Results – CPLD
H4IRRAD 5x1010cm2
2 Latch-up events on the heated CPLD
Power ON
Power OFF
Current increase to 500 mA
First event
Results – CPLD
H4IRRAD 5x1010cm2
2 Latch-up events on the heated CPLD
Power ON
Power OFF
Current increase to 500 mA
Conclusions AD822
almost at the limit of specifications;
Seem to agree with literature
PSI test to be done
CPLD
Other components within specifications
2 Latch-up on the CPLD
No Latch-up on MAX11046 and PCM
Note
Signals are noisy
Detailed tests on mixed signal devices require a protected zone where a mother board can be placed
Comparison H4IRRAD (low dose rate and mixed field) – PSI (p+ 230 MeV, high flux, high dose rate)
Total TID 35 Gy; HEH fluence 5x1010cm-2
Less Fluence than expected
Difficult to compare
How to explain
H4IRRAD spectrum Spectrum in the target line is dominated
by high energy hadrons (Needed to verify destructive events)
Dose enhancement effect seems to appear mainly in case of Co60 irradiation
Microelectronics may degrade less in a Hadron environment with respect to pure gamma irradiation
Open point: depending on the technology, the Hadron energy (literature focused on proton), and the percentage composition of our spectrum
Results-slot 1 &2DUT
Type-Technology
PSI, p+, >250Gy/h *
H4IRRAD0.3 Gy/h *
Comment
MAX410 OpAmp-Bip 100 Gy-ok 85-ok
OPA2227 OpAmp 200 Gy -ok 85-ok
TL072 OpAmp 200 Gy-ok -
TL431 Voltage Ref 200 Gy-ok 85 – ok
TL432 Voltage Ref 200 Gy-ok 85 - ok
LM4041 Voltage Ref 200 Gy-Ok 85 – ok
INA141 DiffAmp 130Gy-Out of spec 85 - ok Limit of
spec
MAX6341 Voltage Ref (ADC)
25 Gy-Out of spec
25 Gy-Out of spec.
3 mV drift over 100 Gy* Max TID without failure
or TID at which measurements are out of spec