Gas gain measurement setup

DSC03121.JPG DSC03117_r.JPG This page contains information on operating the gas gain measurement setup in H037.

General information

This setup is used to measure the gas gain in a GridPix detector by reading out simultaneously the detector via the Pixelman or RelaxDAQ software and the induced signal on the grid by an oscilloscope.

Pressurised test vessel

  • Cylindrical vessel with CF150 flanges
  • 4 CF40 ports on the cylinder wall:
    • blind flange to hold the Fe-55 source on the inside
    • gas input with pressure sensor
    • gas output with needle valve
    • connection to vacuum pump (clap with yellow valve), a Pfeiffer TCP 015 turbo pump achieving pressures down to 10^-8 mbar
  • 4 CF40 ports on the top flange with feedthroughs for:
    • HV lines for grid voltage, guard voltage and cathode voltage; SHV connector for external test input for preamplifier calibration (needs BNC->SHV adapter)
    • preamplifier power supply and signal output
    • flat data cable
    • residual gas analyzer (RGA) to measure impurities under ultra-high vacuum ( P < 10^-6 mbar); don't switch on before achieving this good vacuum!, this could damage the device
  • With gas in- and outlet closed the vessel is tight and can be evacuated (to accelerate the flushing procedure after opening) or operated at pressures P > 1 bar.

GridPix inside

DSC03113_r.JPG

  • The drift volume with chip, guard electrode and cathode mesh.
  • Currently the drift volume is not equipped with field shaping electrodes. We neglect field deforming effects because the area of the chip is far enough from the edges.
  • Drift length is 6 cm. We apply a potential difference of 1800 V between guard and cathode for a drift field of 300 V/cm.
  • The distance between guard and grid is about 3 mm, so to maintain the drift field, the difference between guard and grid voltage has to be 100 V.
  • Typical grid voltages for different gases:
    • Ar/iC_4H_10 : 380 V (1 bar), 400 V (1.5 bar)
    • Ar 4.7:
  • Honeycomb preamplifier connected to the grid's HV input pad on PCB backplane to read out the (positive) signal induced by the moving charge of the electron avalanche.
    • include drawing of preamp with all connections
    • has 4 inputs:
      • signal from grid (IN), decoupled from HV input pad via 47 pF capacitor
      • calibration test pulse (CAL)
      • +5/-5 V power supply
    • has 2 outputs:
      • fast signal for triggering the acquisition
      • slow signal to record on oscilloscope for analysis

Trigger setup

P1010145.JPG

  • to synchronise the acquisition of both the readout software (Pixelman or RelaxDAQ) and the oscilloscope
  • It consists of several timers, some used as gates, others as delays. Here's the flow diagram of the trigger logic.
  • We substituted some of the NIM modules, mainly gates and NIM <-> TTL converters, by a Wiener NIMbox. It can be programmed with the help of D:\labview\GainSetup_nimbox.vi. The LabVIEW block diagram of the NIMbox shows its current functionality. (An alternative layout uses the shutter signal instead of the trigger to reset the flip-flops for the Relaxd and oscilloscope states.)

Preparing the setup

Testing a GridPix before closing the vessel

Sometimes it's necessary to open the vessel for repairs: a new chip, fix broken connections, remove impurities, etc. In this stage it's crucial not to forget things to check before closing the vessel and bringing it from from the cleanroom back to the cryolab (every error means another cleanroom visit and everytime you need new gaskets for any opened port...). These are the things that have to be verifiedbefore closing:

  • Are all the lines connected? (HV, preamp power and signal, GND lines) test with multimeter
  • Where do the HV and ground lines go? prevent them from touching/inducing
  • read out chip with Relaxd next to probing station
  • also test under HV (bring HV supply and gas bottle to cleanroom)
    • in air (grid voltage < 400 V!)
    • with Ar flow through tube close to the chip to check for spark hotspots

Calibrate the preamplifier

Use well-known test pulse to induce charge in preamplifier to calibrate its response. This can be done by both the amplitude and the integral of the recorded waveform. Since the peak height (in certain limits) is proportional to area under the curve both methods are equivalent. Fill with more info and details

Operating the setup

The operation of the gas gain measurement set-up includes several hazards:

  • radioactivity: Fe-55 gamma source, 5.9 keV,
  • high voltage up to 2.5 kV,
  • of pressurised (flammable) gas.
To protect the set-up as well as the experimentalist's health from harm we created this list as a sort of check list, also to assist the experimentalist in case of (his/her) memory leak.

Please follow the order of the actions to take! Only skip parts if really unnecessary (i.e. if the PC is already switched on).

Safety rules

DSC03128_arrow.JPG

  • Never open the gas exhaust valve when HV is on! You may easily get a field higher than the breakdown limit at lower pressure!
  • Don't switch on the power supply for the Relaxd board when the vessel is evacuated. This may damage the chip because the resulting heat won't be dissipated.
  • Use only the power button (see picture) to switch on/off the turbo-pump. At first a pre-pump will evacuate and the turbo-stage will start rotating slowly before taking over the pumping when its rotation speed is high enough.

Start up the system

Switch on the hardware

  1. Computer
    • By default it should be already on.
    • If logged out (for instance due to automatic security updates\dots) log in with the user name admmedipix. Note that this login has administrative rights.
  2. Oscilloscope LeCroy WaveRunner
    1. Check if the oscilloscope is connected via Ethernet to the computer.
    2. Boot the Windows system on the oscilloscope.
    3. To connect to the shared folder on the PC open the text file Network connection sharing settings.txt. Execute the command in the command prompt and enter the password for admmedipix.
    4. The shared folder should now be at drive F:\.
  3. Oscilloscope Tektronix
    • This oscilloscope allows to monitor the correct timing of signals.
    • Since it doesn't alter the system and/or the timing of signals there are no special precautions to be taken.
  4. NIM crate
    • Switch on the crate and plug in the fan's power cord (on its back).
  5. Relaxd board
    1. Disconnect the 2 cables for the SHUTTER and the BUSY signal. (Due to some bug in Pixelman it doesn't connect correctly to the Relaxd board and thus won't find the chip if the two are connected on start-up).
    2. Switch on the power supply. It should deliver 10 V and around 0.4 - 0.6 A, depending on the number of hit pixels (high power consumption for an almost full matrix).
    3. Don't reconnect the cables for SHUTTER and BUSY before getting to the Pixelman step.
  6. Preamplifier of the grid
    • Switch on the power supply for the preamp.
    • Verify that the noise on the LeCroy oscilloscope is as low as expected (lower than 10-15 mV).
relaxd_board.png

Start the software

  1. Command prompt
    • To monitor the connection between the Relaxd board and the PC type the following in the command prompt: ping -t 192.168.33.175
  2. Pixelman
    1. Start JPixelman from the QuickStart menu next to the Windows start button.
    2. Check if the Chip ID makes sense.
    3. Options / Device Settings / Interface specific settings: set Medipix digital voltage to 90.
    4. Run it for a few, i.e. O(10), acquisitions (don't abort!).
    5. Reconnect the 2 cables for SHUTTER and BUSY to the Relaxd board (see sketch).
    6. Run again for a few acquisitions, don't abort (otherwise RelaxDAQ starts up with initialization errors).
  3. RelaxDAQ
    1. Start it with the QuickStart symbol next to the Windows Start button.
    2. Set mode to "Use pixel configuration" and load the respective binary pixel configuration file ( *.bpc; e.g. D:\Rolf\121213_D06-W0056_edge_optimised.bpc). Note that the bpc file automatically sets the mode (counting, ToT or ToA mode; ToA in the above example).
    3. Set the trigger to "Software" and run RelaxDAQ during increase of the voltage (see next step).
  4. LabVIEW
    1. Make sure the ISEG HV supplies are switched on.
    2. Launch Set and check HV from the directory D:\Rolf\.
    3. Run the LabView programme (black arrow top left).
    4. Set the current limit for a software trip to 160 nA.
    5. Switch the red toggles on the power supplies to HV on.
    6. Slowly increase the voltage by keeping V(cathode) = V(guard) + 1800 V = V(grid) + 1900 V.
    7. Monitor the chip's behaviour via Pixelman or RelaxDAQ, especially when you get into the voltage domain where you expect signals.
    8. Be very careful with increasing the voltage when entering the discharge domain. Avoid staying in that domain for a long time.
-- RolfS - 2013-03-15
Topic attachments
I Attachment Action Size Date Who Comment
Unknown file formatJPG DSC03113_r.JPG manage 2154.1 K 2013-01-09 - 15:21 UnknownUser The drift volume with PCB-A on support disk, (invisible) guard electrode and cathode mesh
Unknown file formatJPG DSC03117_r.JPG manage 2063.2 K 2013-01-09 - 14:57 UnknownUser The top flange with 4 ports for feedthroughs
Unknown file formatJPG DSC03121.JPG manage 2214.3 K 2013-01-09 - 14:52 UnknownUser The gain measurement setup as of end 2012.
Unknown file formatJPG DSC03128_arrow.JPG manage 733.5 K 2013-01-09 - 15:15 UnknownUser Power button on the turbo pump
Unknown file formatJPG Nimbox_back__panel.JPG manage 216.8 K 2013-01-11 - 08:40 UnknownUser Block diagram/back panel of the Labview programmable NIMbox
Unknown file formatJPG P1010006.JPG manage 838.2 K 2011-12-01 - 12:37 UnknownUser The drift volume with chip, guard electrode and cathode mesh
Unknown file formatJPG P1010014_r.JPG manage 836.2 K 2011-12-01 - 13:04 UnknownUser The pressurised test vessel
Unknown file formatJPG P1010015.JPG manage 774.8 K 2011-12-01 - 12:32 UnknownUser The gas gain measurement setup in H037.
Unknown file formatJPG P1010145.JPG manage 786.4 K 2013-01-11 - 14:24 UnknownUser Picture of the slimmed-down trigger synchronisation with the NIMbox (right)
Unknown file formatJPG nimbox_back_panel.JPG manage 226.6 K 2013-01-30 - 10:10 UnknownUser Block diagram/back panel of the Labview programmable NIMbox
Unknown file formatJPG nimbox_v2_back_panel.JPG manage 226.6 K 2013-01-29 - 16:06 UnknownUser Block diagram/back panel of the Labview programmable NIMbox (alternative layout!)
PNGpng relaxd_board.png manage 26.6 K 2013-02-11 - 11:03 UnknownUser Sketch of the Relaxd board and its connectors
PDFpdf trigsys.pdf manage 33.3 K 2013-01-09 - 15:01 UnknownUser Flow diagram of the trigger logic
Topic revision: r13 - 2013-03-15 - rolfs_40nikhef_2enl
 
This site is powered by the TWiki collaboration platform Powered by Perl This site is powered by the TWiki collaboration platformCopyright © 2008-2018 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback