LOW VOLTAGE SOLUTION FOR CBM MUCH By Vinod Singh Negi Under the - - PowerPoint PPT Presentation

▶
low voltage solution for cbm much
SMART_READER_LITE
LIVE PREVIEW

LOW VOLTAGE SOLUTION FOR CBM MUCH By Vinod Singh Negi Under the - - PowerPoint PPT Presentation

Aug 03, 2023 170 likes •1.01k views

LOW VOLTAGE SOLUTION FOR CBM MUCH By Vinod Singh Negi Under the supervision of Jogender saini and Dr. Subhasish Chattopadhyay 1 O UT L INE OF WORK DONE Radiation testing of Low voltage components. Neutron Irradiation * Damage Study *

slide-1
SLIDE 1

LOW VOLTAGE SOLUTION FOR CBM MUCH

By Vinod Singh Negi Under the supervision

  • f

Jogender saini and Dr.Subhasish Chattopadhyay

1

slide-2
SLIDE 2

OUTLINE OF WORK DONE

 Radiation testing of Low voltage components.

Neutron Irradiation

* Damage Study * Single Event upset analysis for control board Gamma Irradiation * TID tolrrence

 Fault tolrrence LV Board Design

* Hard On and Stable OFF

 Implementation of reliable Ethernet communication .

* UDP Implemented successfully

* TCP/IP Imlementation , yet to start

 GUI Development for handling huge number of channels

* 30 Channel control *60 parameter monitored

2

2

slide-3
SLIDE 3

RADIATION TESTING OF LOW VOLTAGE COMPONENTS

NEUTRON IRRADIATION * DAMAGE STUDY

* SINGLE EVENT UPSET ANALYSIS FOR CONTROL BOARD

GAMMA IRRADIATION

* TID TOLERANCE

3

slide-4
SLIDE 4

CONDITIONS

 No annealing , Continuous Radiation

 Accelerated life test (Acceleration factor

1000)

 No Cooling With greater SEU cross section and

Damage factor

4

slide-5
SLIDE 5

NEUTRON IRRADIATION TESTING OF LVDB

COMPONENTS

5

slide-6
SLIDE 6

NIEL FACTOR VS PARTICLE TYPE, ENERGY

10-10 10-9 10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102 103 104

particle energy [MeV]

10-5 10-4 10-3 10-2 10-1 100 101 102 103 104

D(E) / (95 MeV mb)

neutrons neutrons pions pions protons protons electrons electrons

100 101 102 103 104 0.4 0.6 0.8 1 2 4

neutrons neutrons pions pions protons protons

NIEL factor of Silicon

for comparison, normalized to 1 MeV neutron

slide-7
SLIDE 7

7

Damage factor

slide-8
SLIDE 8

Reference https://rd50.web.cern.ch/rd50/NIEL/default.html

ENERGY DAMAGE ENERGY DAMAGE

slide-9
SLIDE 9
slide-10
SLIDE 10

Neutron = 1 x 1012 Neutron = 5 x 1011

slide-11
SLIDE 11

NEUTRON IRRADIATION

Exactly same set up were used

 14 Mev mono-energetic neutron generator which works on

D-T reaction with the threshold of 300kev.

 Yield = 7.7 × 108N14/s  Flux = ( 6.132× 107 )N14 /cm

2 /s

R

2

* Damage Factor = 2 Converter, multiplexer switch and NOT gate sustained dose

  • f 4.46 × 1012Neq/cm2

Observations

 Slight increment in peak to peak ripples.  No effect on dc value of devices.  Single event transient and upsets were observed 11

slide-12
SLIDE 12

12

Converters Current sensing IC SPI Digital POT(LATCH) Multiplexers . All the components were closely packed so that maximum beam exposure

slide-13
SLIDE 13

13

Converters Current sensing IC SPI Digital POT(LATCH) Multiplexers ALL COMPONENTS ARE PLACED AT < 1.3cm

slide-14
SLIDE 14

14

FPGA at < 1.5cm SPI FLASH < 2.5 cm

slide-15
SLIDE 15

COMPUTER

12V

Non Isolated DC to DC converter 2 ERROR RESILENT SPI CONTROL POT Non Isolated DC to DC converter 1 ADC 1

MAX 4373

Rsense FEE 2 FEE 1 10kohms

ADC 2

M U X 5V

UDP

RADIATION ZONE

slide-16
SLIDE 16

16

PCB exposed to radiation environment. ADCs digitized and fetch different voltage and current of devices under test to Spartan 6 Lx9 FPGA boards. FPGA send data to computer via UDP protocol Online real time parameter monitoring and data logging were done in Matlab.

slide-17
SLIDE 17

17

slide-18
SLIDE 18

18

Converter 1 Converter 2

Digital Pot

Current Sensing IC

MUX CHANNEL NUMBER

slide-19
SLIDE 19

19

slide-20
SLIDE 20

20

  • 10% data plotted
  • 2.23 X 10e12 N14 /cm2 = 4.46 X 10e12 N1 Mev /cm2
  • No change in DC Value
  • Ripple increases at the end
slide-21
SLIDE 21

21

40mV pk-pk 60mV pk-pk 3.21 x 1012 Neq/ cm2

slide-22
SLIDE 22

22 1 2 3 4

X 1012

Normalized Gain

slide-23
SLIDE 23

23

MUX Response

continue

SET

slide-24
SLIDE 24

CONCLUSION

Various component from different companies were irradiated and after multiple failures we have some components which can tolerate desire radiation level.

All components were tested at very high dose rate

24

S.No Gamma Neutron Size Converter1 600Gy 4.46 ×1012 Neq/cm2 Converter2 600Gy 4.46 ×1012 Neq/cm2 Mux 60Gy 4.46 ×1012 Neq/cm2 Digital Pot 300 Gy 4.46 ×1012 Neq/cm2 Current sensing IC 260 Gy 3.73 ×1012 Neq/cm2 Switch >700Gy 4.46 ×1012 Neq/cm2 Not Gate >700Gy 4.46 ×1012 Neq/cm2

slide-25
SLIDE 25

25

RS-232 via USB-UART Bridge 10/100 Ethernet PHY 24 GPIOS 128MB SPI FLASH Spartan 6 lx9 board

  • Automatic rebooting of Bit- file in FPGA with

desirable Refresh rates

  • SPI Flash refreshed with the help of Diligent

Software.

  • Read-back of bit file can be done via impact.

CONTROL BOARD

slide-26
SLIDE 26

SINGLE EVENT UPSETS(BIT-FILE)

26

Size = 2722512bits Dimensions = 2cm X 2cm Flux= 2.725 x 10 7 Neq /cm2/s

slide-27
SLIDE 27

27

REFRESH RATE 1 HOUR

13935 13656 13985

slide-28
SLIDE 28

28

REFRESH RATE 10 MINUTES

slide-29
SLIDE 29

29

REFRESH RATE 10 MINUTES . Bin Size = 5 Total entities = 140 Max Occurrence = 398 UPSETS IN BIT FILE

slide-30
SLIDE 30

30

REFRESH RATE 2 MINUTES Bin Size = 1 Total entities = 240 Max Occurrence = 42 UPSETS IN BIT FILE

slide-31
SLIDE 31

31

UPSETS IN FLASH MEMORY

  • 128Mb
  • Size 10 X 7.10 mm2
  • FLUX= 107 Neq/cm2/s
  • Read-back by Diglent software
  • File format .mcs (Hex-File)
slide-32
SLIDE 32

32

UPSETS IN FLASH MEMORY

  • 128Mb
  • Size 10 X 7.10 mm2
  • FLUX= 107 Neq/cm2/s
  • Read-back by Diglent software
  • File format .mcs (Hex-File)
slide-33
SLIDE 33

33

UPSETS IN FLASH MEMORY

  • 128Mb
  • Size 10 X 7.10 mm2
  • FLUX= 107 Neq/cm2/s
  • Read-back by Diglent software
  • File format .mcs (Hex-File)
slide-34
SLIDE 34

34

UPSETS IN FLASH MEMORY Bin Size = 1 Total entities = 240 Max Occurrence = 38

slide-35
SLIDE 35

35

UPSETS IN FLASH MEMORY

REFRESH RATE 3 MINS Bin Size = 1 Total entities = 345 Max Occurrence = 12

slide-36
SLIDE 36

DIGITAL POT

Size 8 bit Dimensions =1.63mm x 2.9mm Flux = 3.62 x 107 N14 /cm2 /s

36

2.73Kohm= 00000111

5V ERROR RESILENT SPI CONTROL POT 10kohms

slide-37
SLIDE 37

37

REFRESH RATE / DURATION BEFORE VALUE NUMBER OF UPSETS AFTER VALUE 2MINS (2)Hrs 00000111 Nil 00000111 10 MINS(2Hrs) 00000111 Nil 00000111 25MINS (2Hrs) 00000111 Nil 00000111 60 MINS(2Hrs) 00000111 14 00001010 120 MINS(2Hrs) 00000111 31 10010110 240 MINS(6Hrs) 00000111 33 11011011

slide-38
SLIDE 38

LVDB BLOCK DIAGRAM

Commercial DC Power Supply

MAX 4373

D S G

Rsense FEE 1

MAX 4373

D S G

Rsense FEE2

MAX 4373

D S G

Rsense Load15

Analog Multiplexer Controller 12V

1.2-3.3 V

Selection Line

Non Isolated DC to DC converter 15

3

Non Isolated DC to DC converter 1

ERROR RESILENT SPI CONTROL ERROR RESILENT SPI CONTROL ERROR RESILENT SPI CONTROL

Non Isolated DC to DC converter 2

vcc Analog MUX

12V

CS1 CS2 CS15

slide-39
SLIDE 39

GAMMA IRRADIATION TEST

 Converter, Multiplexer, Switches , NOT gate were tested  Gamma Irradiation with cobalt 60 source(1.17Mev)  Dose rate 240Krad/hr  Resolution of ADC 0.8mV  Sampling rate 5Msps  Fan cooling  On line data acquisition, data logging .  Offline data segregation , data merging.  Analysis of segregated data on mat-lab.

39

slide-40
SLIDE 40

40

Irradiation test set ups

  • Device under test exposed to radiation environment
  • ADCs digitized and fetch different voltage and current of device under

test to Spartan 6 Lx9 FPGA boards

  • Online real time parameter monitoring(Chip-scope pro)and data logging
  • f device under test .
  • Analysis of stored data for understanding the effect of accumulated dose

General Set up

slide-41
SLIDE 41

Gamma Chamber FAN Cooling Online data monitoring and logging FPGA Converter Experimental Set Up

41

continue

slide-42
SLIDE 42

TEST RESULTS WITHOUT RADIATION

Time in minutes A D C c

  • u

n t s Converter output voltage

42

continue

slide-43
SLIDE 43

DC RESPONSE OF CONVERTER1

Dose in rad *104 A D C c

  • u

n t s Converter Start Failing 2.625 Attempt to recover Dose rate is too high in comparison to annealing rate of converter

43

continue

slide-44
SLIDE 44

Time in minutes Ripples without radiation Ripples with radiation A D C c

  • u

n t s Dose in rad *104 Average Value Instantaneous Value Average Value Instantaneous Value

44

continue

slide-45
SLIDE 45

FFT WITHOUT RADIATION

Switching Frequency 620kHz

45

continue

slide-46
SLIDE 46

FFT WITHOUT RADIATION

2307 Dc gain Zoom View at F=0 DC value = 2325

46

continue

slide-47
SLIDE 47

FFT WITH RADIATION

47

continue

slide-48
SLIDE 48

48

600 400 200 800

DC response of Converter2

slide-49
SLIDE 49

Frequency (MHz)

P

  • w

e r S p e c t r a l D e n s i t y 49

continue

slide-50
SLIDE 50

50

RIPPLES WITH RADIATION

200 400 600

slide-51
SLIDE 51

51

Without Radiation With Radiation SWITCHING FREQENCY

slide-52
SLIDE 52

OBSERVATIONS

 Average peak to peak value of ripple is more or less

same before and after irradiation.

 Converter start failing after 26.25krad.  Upto 26.25krad Dc gain is more less same before and

after irradiation so is the efficiency.

 Due to gamma radiation ion hole pair are generated in

subtract which alter the number of charge particle in current thus shot noise is introduced.

 FFT spectrum(without radiation) shows peaks other

with switching frequency and folded version of high frequency noise.

52

slide-53
SLIDE 53

Ch#0 ---Gnd Ch#1 --- 1.4V Ch#2 ---Gnd Ch#3 --- 1.4V Ch#4 ---Gnd Ch#5 ---1.4V Ch#6 ---Gnd Ch#7 ---1.4V

1 2 3 4 5 6 7 0 1 2 3 4

Irradiation set up for multiplexers

Expected Output

  • f Multiplexer

53

slide-54
SLIDE 54

54

MUX RESPONSE

continue

slide-55
SLIDE 55

55

600 400 200 800

slide-56
SLIDE 56

56

Error Resilient control system

slide-57
SLIDE 57

LVDB BLOCK DIAGRAM

Commercial DC Power Supply

MAX 4373

D S G

Rsense FEE 1

MAX 4373

D S G

Rsense FEE2

MAX 4373

D S G

Rsense Load15

Analog Multiplexer Controller 12V

1.2-3.3 V

Selection Line

Non Isolated DC to DC converter 15

3

Non Isolated DC to DC converter 1

ERROR RESILENT SPI CONTROL ERROR RESILENT SPI CONTROL ERROR RESILENT SPI CONTROL

Non Isolated DC to DC converter 2

vcc Analog MUX

12V

CS1 CS2 CS15

slide-58
SLIDE 58

Single CMOS output

Neutron

Head on collision with potential neutron can upset the stored output with very high probability If single CMOS logic is used then control system will be very unreliable.

slide-59
SLIDE 59

59

Reliable approach

  • SPI control resistor based control system with Hard ON

and Stable OFF

  • 100Kohm resistance had been divided into 256 division
slide-60
SLIDE 60

60

  • Base current of npn transistor is controlled by SPI control resistor such

that on binary count of 00000000 converter is OFF and 11111111 make it ON

  • Incoming neutron should change status of all eight bit then only output of

controller is effected ON OFF

continue Inside Cut off region threshold

slide-61
SLIDE 61

61

Reliable approach

  • Neutron Flux upsetting all six or seven bit simultaneously
  • Chances are one in million
slide-62
SLIDE 62

62

Vdd R1 R2 X 100-X . Vcc R3 R4 R5 R6 Digital pot To Inhibit Pin T1 T2

FAULT TOLERANT CONTROL WITH SPI POT

LOGIC(HARD ON STABLE OFF) T1 , T2 Both in deep cut off-------------ON Either one of them ON--------------------OFF Ib1 Ib2

slide-63
SLIDE 63

63

R1 + Xon R2 + 100- Xon DEEP INTO CUT OFF REGION R1 + Xoff R2 + 100- Xoff JUST ABOVE CUT OFF REGION AND R1 + Xoff =R2 + 100- Xoff Xoff = R2-R1 +100 2

slide-64
SLIDE 64

64

Rb1=R1 + Xon Rb2=R2 + 100 -Xon Rb1=Rb2=R1 + Xoff =R2 + 100- Xoff UPSET CAUSE BY RADIATION Xon= 00000000 Xoff= 11111000

Floating enable= ON Disable voltage >0.8-3.3 Enable voltage < 0.8

ON OFF . Hard ON And Stable off ZOOM

slide-65
SLIDE 65

65

Rb1=R1 + Xon Rb2=R2 + 100 -Xon Rb1=Rb2=R1 + Xoff =R2 + 100- Xoff UPSET CAUSE BY RADIATION

Xon= 00000000 Xoff= 11111000

Floating enable= ON Disable voltage >0.8-3.3 Enable voltage < 0.8

ON OFF .

Hard ON and Stable OFF

slide-66
SLIDE 66

First prototype of LVDB Board

  • 4 layer PCB
  • Voltage and Current monitoring points
  • On- Off control
  • Radiation hard components
slide-67
SLIDE 67

Current monitoring MUX Control MUX Voltage monitoring MUX ADC Input 12V Control Ports

slide-68
SLIDE 68

Output Connector

Input Filter capacitor Output Filter capacitor Sense Resistor Control resistor Inductors DC-DC Convertor Z O O M

slide-69
SLIDE 69

69

4:1 design A D C SPI control register is keep on refreshing with 1Khz rate via multiplexer Current and voltage monitoring of each channel

slide-70
SLIDE 70

Controller Scrubbing Error Resilient design of board

slide-71
SLIDE 71

71

  • Maximum four board can be interface with Spartan six lx9 board
  • In our design we had connected two boards
  • UART/ Ethernet
  • GUI on mat lab

Computer U D P

slide-72
SLIDE 72

72

Front Panel Board 1 Board 2

Monitoring and control GUI on mat-lab

slide-73
SLIDE 73

HEIP FROM COLLABORATION

73

EPICS ????

slide-74
SLIDE 74

Thank you

For more information regarding the radiation hard components, please contact me at vnegi@vecc,gov.in

74

slide-75
SLIDE 75

COMPUTER

12V

Non Isolated DC to DC converter 2 ERROR RESILENT SPI CONTROL POT Non Isolated DC to DC converter 1 ADC 1

MAX 4373

Rsense FEE 2 FEE 1 10kohms

ADC 2

M U X 5V

UDP

slide-76
SLIDE 76

76

slide-77
SLIDE 77

77

Vdd R1 R2 X 100-X . Vcc R3 R4 R5 R6 Digital pot To Inhibit Pin T1 T2

FAULT TOLERANT CONTROL WITH SPI POT

LOGIC(HARD ON STABLE OFF) T1 , T2 Both in deep cut off-------------ON Either one of them ON--------------------OFF Ib1 Ib2

slide-78
SLIDE 78

78

R1 + Xon R2 + 100- Xon DEEP INTO CUT OFF REGION R1 + Xoff R2 + 100- Xoff JUST ABOVE CUT OFF REGION AND R1 + Xoff =R2 + 100- Xoff Xoff = R2-R1 +100 2

slide-79
SLIDE 79

79

Damage factor

slide-80
SLIDE 80

80

Rb1=R1 + Xon Rb2=R2 + 100 -Xon Rb1=Rb2=R1 + Xoff =R2 + 100- Xoff UPSET CAUSE BY RADIATION Xon= 00000000 Xoff= 11111000

Floating enable= ON Disable voltage >0.8-3.3 Enable voltage < 0.8

ON OFF . Hard ON And Stable off ZOOM

slide-81
SLIDE 81

81

Rb1=R1 + Xon Rb2=R2 + 100 -Xon Rb1=Rb2=R1 + Xoff =R2 + 100- Xoff UPSET CAUSE BY RADIATION

Xon= 00000000 Xoff= 11111000

Floating enable= ON Disable voltage >0.8-3.3 Enable voltage < 0.8

ON OFF .

Hard ON and Stable OFF

slide-82
SLIDE 82

82

slide-83
SLIDE 83

83

R1 + Xon R2 + 100- Xon DEEP INTO CUT OFF REGION R1 + Xoff R2 + 100- Xoff JUST ABOVE CUT OFF REGION AND R1 + Xoff =R2 + 100- Xoff Xoff = R2-R1 +100 2