Microroc Production

From Linear_Collider_LAPP

(Redirected from MicrorocProd)
Jump to: navigation, search

Before beeing cabled on ASU, each channel of each MICROROC chip is individually tested, to ensure minimum performance of final detector in term of dead channel, reliability, noise and dynamic range.



  • Have an account on lappsl5
  • Have Jean's framework installed and properly compiled [help]
  • Have followed a demonstration of ASIC and test socket proper handling in respect of electrostatic hazard, by Cyril, Guillaume or Renaud


  • MICROROC test board, each one with its power supply
  • A pulse generator, with tf=2ns, dV=-100mV, width=5us
  • Labview 2010 and acquisition v 5.0b from Cyril installed and launched on each computer

Each testbench is marked with a color, RED or BLUE !

MICROROC Qualification

How MICROROC are tested ?

First, the ASIC are stimulated with 4 kinds of signals:

  • Check configurability of the ASIC
  • Check input bonding of the 64 channels (except channel 8 on board 1)
  • Perform 100 acquisitions for each DAC0 (low threshold) setting without any injection
  • Perform 100 acquisitions for each DAC0 setting with 50fC signal injection

Then, data are transferred to lappsl5 and analyzed with framework

  • Reconstruction of bonding, pedestal and data
  • Analyse bonding, pedestal and data
  • Compute and save gain of each channel

The following checks are performed

  • Verify that all input bonding are connected (see paragraph known problems)
  • Verify that the data are well formatted
  • Verify that all channels flip, and that they are not noisy beyond flipping dac value
  • Verify that all gain are ok

How to launch qualification tests ?

The first time only:

  • Run the Labview application
  • On the first tab, clic on "Envoi slow-control" this should give OK
  • On the calibration tab
    • choose "piedestaux" and clic on "transfert configuration piédestaux (expert)"
    • choose "calibration sur 1 chip" and clic on "transfert configuration calibration (expert)"

Now, each time, on the qualification MICROROC tab:

  1. Choose the number of the chip to be tested (ensure that run is "1")
  2. Clic on "Start qualification"
  3. When qualification is done, just follow instruction on Labview computer screen (and goto 2)

Do not forget to push back the pressure pad of the test socket !

After each qualification, the data must be transferred on lappsl5 in the /lapp_data/LC/Detecteur/Micromegas/data/MICROROC/PROD_2/LABVIEW directory.

How to check data ?

Log in lappsl5, as lcdet or do not forget to change your group to LC (newgrp LC).
Change current directory to the trunk directory of your framework.
Do not forget to source env.csh before launching qualification script.
In your framework directory, just type:

./Batch/qualification_microroc.sh xxx

where xxx is the number of the chip to be tested

How to interpret results ?

This has been tested with version 1570 of the framework.
This is an exemple of a good chip, without any doubt:

161:(14:07) ~/FRAMEWORK/trunk$ ./Batch/qualification_microroc.sh 372
src/analyse/Renaud/BATCH/change_chip_number.sh: edit microroc_test_bonding_2.xml file
src/analyse/Renaud/BATCH/change_chip_number.sh: edit microroc_test_pedestal_2.xml file
src/analyse/Renaud/BATCH/change_chip_number.sh: edit microroc_test_data_2.xml file
src/analyse/Renaud/BATCH/change_chip_number.sh: edit test_existing_files.sh
src/analyse/Renaud/BATCH/change_chip_number.sh: edit test_microroc_3.sh file
src/analyse/Renaud/BATCH/test_existing_files.sh: files exists
src/analyse/Renaud/BATCH/test_microroc_3.sh:reconstruction for bonding test
src/analyse/Renaud/BATCH/test_microroc_3.sh:reconstruction of pedestals (this may take a moment)
src/analyse/Renaud/BATCH/test_microroc_3.sh:reconstruction of data (this may take a moment)
src/analyse/Renaud/BATCH/test_microroc_3.sh:analyze bonding
Chip 372 on test board 1 has bonding OK !!!
src/analyse/Renaud/BATCH/test_microroc_3.sh:analyze pedestal
src/analyse/Renaud/BATCH/test_microroc_3.sh:analyze data
src/analyse/Renaud/BATCH/test_microroc_3.sh:analyze gain
   OO:::::::::OO   K:::::::K    K:::::K 
 OO:::::::::::::OO K:::::::K    K:::::K 
O:::::::OOO:::::::OK:::::::K   K::::::K 
O::::::O   O::::::OKK::::::K  K:::::KKK 
O:::::O     O:::::O  K:::::K K:::::K    
O:::::O     O:::::O  K::::::K:::::K     
O:::::O     O:::::O  K:::::::::::K      
O:::::O     O:::::O  K:::::::::::K      
O:::::O     O:::::O  K::::::K:::::K     
O:::::O     O:::::O  K:::::K K:::::K    
O::::::O   O::::::OKK::::::K  K:::::KKK 
O:::::::OOO:::::::OK:::::::K   K::::::K 
 OO:::::::::::::OO K:::::::K    K:::::K 
   OO:::::::::OO   K:::::::K    K:::::K 
162:(14:12) ~/FRAMEWORK/trunk$

This is an example of a chip with one bad channel:

exemple here

This is an example of a chip with bad bonding:

exemple here

This message after pedestal or data testing:

Warning in <Fit>: Fit data is empty

****** bin/root/test_microroc_2: Warning! Chip 542 Channel 55, fit may be not correct ! Inflexion: 150 Dac min: 150 Dac max: 220

To check manually a channel, (ie channel 2 of chip 535) just type:

root './src/analyse/Renaud/show.C(535,2)'


.x ./src/analyse/Renaud/show.C(535,2)

Known bugs and limitation

  • Inputs 8 and 42 are not tested on testboard 1
  • Inputs 8 and 42 and 62 are not tested on testboard 2
  • Calibration on channel 62 and 63 is wrong because of a problem inside the chip

Reference Informations

The power consumption should be around 288 mA @ 6.5 V at the beginning of the run and about 293 mA @ 6.5 V at the end of the run.

Labview default value

  • Chip parameters
    • en_otaq: OFF
    • en_gbst: ON
    • en_trig_out: OFF
    • LG and HG gain: 3
    • RAZ channel external only
  • Configuration tab : tempo_sc: 2 ms
  • Calibration auto tab
    • Pedestal: 150 -> 220
    • Calibration: 480 -> 590
    • Test pattern: 7 with offset 1 (first is channels 0, 8, 16 ... 56)
    • Tempo SC: 50 ms
    • Timeout receiver: 50 ms

Microroc nominal performances

Preamp + Shaper

Shaper setting Gain (measured) Gain (simulated) Tp (measured) Tp (simulated)
00 7.6 mV/fC 10.5 mV/fC 76.3 ns 30 ns
01 9.7 mV/fC 11.0 mV/fC 114.2 ns 100 ns
10 9.8 mV/fC 11.1 mV/fC 157.4 ns 150 ns
11 10.2 mV/fC 11.2 mV/fC 192.3 ns 200 ns


Measured DAC slope: 1.38 mV / DACu or 0.72 DACu / mV

Expected result from qualification

Injected charge 50 fC @ shaper setting of "11" -> 10.2 [mv/fC] * 50 [fC] = 510 [mV]
Conversion ratio: 510 [mV] * 0.72 [DACu/mV] = 369.4 [DACu]
This correspond to a conversion ratio of: 7.38 [DACu/fC] or 0.13 [fC/DACu] (with a 20% tolerance on chip manufacturing, inter-channel matching of few %)

Threshold DAC: q_{thr}=1.39 [mV] +/- 10 [µV]
Offset DAC: q_{off}=2.8 [mV] +/- 100 [µV]
ratio: 1.39/2.8=0.496

Courbe: (221-196)/15 = 1.66 [THR/OFF]


Result from calibration

Mean of gain: 0.14 [fC/DACu] or 7.14 [DACu/fC]

Personal tools