Reception tests

(J.Roloff, D.Boye)

Second batch from BNL (March 2022)

IV tested @ BNL

  • 6x DC-LGAD(20um, W3070)
  • 10x DC-LGAD(50um, W3046)
  • 3x AC-LGADs (various pitches, W2)


The characteristics of the different sensors can be found here: *Excel spreadsheet

IV characterization

Setting used for the IV measurement

CV characterization

Setting used for the CV measurement

TimePix3 Integration

(G.D'Amen, J.Roloff)

Katherine Readout for Timepix3 — overall view and panels.

TimePix3 chipboards


  • Board: B12
  • Type: 3x AC-LGADs
  • Wafer: W2001
  • Size: 2mm x 2mm # 14, 15 and 16
  • V_depl: ~ -150V
  • V_BD: ~ -200V (??? to check)
  • Pixel geometry: pitch 55um, gap 30um
  • Strip geometry: pitch 200um, gap 120um
  • Central AC-LGAD active area (x, y): (105,100)(145, 100)(145, 140)(105, 140)
  • Bottom left AC-LGAD active area (x, y): (20,0)(70, 0)(70, 50)(20, 50)
  • Bump bond locations on strips: (24,5), (28,5), (32,5), (34,5), (38,5), (42,5)


  • 500 um Silicon (?)
  • Given by Lukas for calibration
  • V_depl: ~70V
  • V_op: 150-300V
  • detector has been mistreated a bit and has a quite high leakage current (which mostly is generated at the dicing edges and goes in the guard ring)
  • Generally speaking putting more than 1-2V/um sensor thickness will not buy you much apart from increasing the leakage since velocity saturation kicks in

Ethernet configurations to connect to Katherine TimePix3

Open a new wired connection using the following settings:

  • IPv4 Method: Manual
  • Address:
  • Netmask:
  • Gateway:

How to connect to Katherine TimePix3 setup

  1. Configure Ethernet connection on your PC (see previous step)
  2. Connect Ethernet cable from Katherine to PC
  3. (optional) Connect Lemo cable for bias from Katherine to chipboard
  4. Connect chipboard to Katherine using VHDCI cable/port
  5. Connect 5V DC (@4A) power supply to turn ON Katherine. Setup should look like this:
    TimePix3 + Katherine setup at CERN.
  6. Wait for led close to Ethernet port to blink orange
  7. Open MMTrackLab. Katherine should appear under "Live detectors". If it doesn't, click on "Add manually..." and "Auto-scan for Katherine devices". You should see something like this:
    MMTrackLab window for available Katherine readout systems.
  8. Once you are connected to the Katherine board, you should see a windows like this one listing the board IP address ( and the Timepix3 chip name (H3-W00053):
    MMTrackLab window when connected to a properly configured Katherine board attached to a TimePix3 chipboard (H3-W00053).

Katherine LED color code

  • blinking red = looking for chipboard. If none is found, it keeps blinking red
  • solid red = ??? (observed with B4)
  • orange light + red light = chip found
  • blinking orange = issue with chip communication
  • Fully functioning pattern is: blinking red -> stop -> blinking red -> solid red + blinking orange -> solid red + solid orange (observed with B12)

UniGe collaboration

(G.Barone, D.Boye)


  • 2022/03/14: second batch of sensors shipped from BNL

ToDo List


DEBUG info

  • LEMO connector is qualified for up to 100V (even though it happily supports much more) and the TimePix PCB design is not really HV compatible


MMTrackLab connects to Katherine board, but the program crashes when I try to run an acquisition

The TP3 chipset must be installed and recognized by the board to start an acquisition. See if a Timepix3 chip is listed on the left column under the Katherine menu. If the chipset is connected but does not appear in the left column, reboot the board.

(Linux Only) MMTrackLab crashes when I try to save a figure/scan result/configuration file

Run MMTrackLab as sudoer.

What's the difference between the threshold set by the Threshold Scan function and the one(s) obtained from the Threshold Equalization function? (response by Petr Mánek)

You can view the Threshold Scan procedure as a series of acquisitions where the threshold (the global THL of the Timepix3 chip) is systematically increased, and the sum of all pixels' values (or the derivative of the sum, depending on the settings) is recorded at each step. This is useful for instance when you wish to estimate the specific amount of deposited energy (in eV) associated with threshold levels that you can configure. If your Timepix3 chip observes a radiation source that emits particles of known energy, you can expect to see a peak for that energy in the generated histogram. If you repeat this several times for different energies and note the center of each peak, you can construct a characteristic curve that maps thresholds to energies and use it for lookup. As you suspected, the Threshold Equalization procedure evaluates the response pedestal for each pixel independently and then calculates a 4-bit adjustment offset (also known as trim or local threshold) for each pixel in order to compensate against variances between pixels due to manufacturing irregularities. If successful, this makes the chip behave more homogeneously or equally (hence the name of the procedure), meaning that from that point onward only global threshold of the entire chip needs to be tuned. At the end of the procedure, a settings preset is generated (its name is adjustable but defaults to "Equalized [YYYY-MM-dd]") so that the calculated settings can be recalled later.

How do we get (automatically) the optimal threshold value from the threshold scan? (response by Petr Mánek)

That depends on what you consider to be "optimal" in this case. What exactly are you optimizing for? If you are looking for a threshold value that is the mean of all pixels in the chip, you will definitely want to review the outputs of Threshold Equalization. In particular, have a look in the "Equalized" settings preset that is produced at the end. In addition to containing all per-pixel adjustments, the preset contains (a global) threshold that is set to a value 10*standard_deviation above the mean observed noise level.

We are trying to take data from a small area at the center of the chip and have been able to set the proper area mask from the Advanced menu under Chipset. However when we close MMTrack we lose the mask setting. We have noticed that there is an option to read the mask from file, but we don't know how to save it on file in the first place. It looks like it doesn't get saved in the chipset preset. (response by Petr Mánek)

MMTrackLab has the capability to fully save and recall settings for Katherine readouts and Timepix3 chips, and label them with arbitrary names (that make sense to you as the user). Your use case looks like it may benefit from this feature. If you look in the left pane of the Data Source Control window, where you would normally configure chip settings, you will see a "Preset:" drop-down box in the upper part of each settings widget. This box contains all currently available presets that you can switch between (selecting an option in this box will activate the preset). Whenever you deviate from the preset in any way by changing some parameters in the UI (e.g. adjust "Threshold" or open the "Advanced" window and configure a custom mask), the "Preset:" box will transform from a drop-down box into a conventional text box, so as to indicate that the preset is currently being modified and that whatever changes were made, have not yet been saved. To the right of the box, you will find two buttons: one with a floppy disk icon that saves (overwrites) the preset with the modified settings (potentially renaming it whatever name is written in the text box), and another that discards the modifications, returning to the previously saved state. There is a third option revealed if you long-click the floppy disk button, which saves the modified settings as a new preset (again, with the name taken from the text box). This is really useful if you want to duplicate a settings preset entirely, but change a small subset of parameters (e.g. impose a custom mask on settings produced by Threshold Equalization). To keep track of all your settings, there is a "Manage configurations" window accessible from the main window of MMTrackLab via "Tools > Manage configurations...". In this window, you can view all available presets and potentially reorganize them. The software also has capability to download presets via the Internet for known chips and devices (our research group maintains a curated public repository of configurations) and import presets from settings files used by other DAQ software.

How to check if TP3 chipboard is recognized (response by Petr Burian)

So, serial number of katherine device is stored in Katherine, chip ID of chip is stored in Timepix3 asic. For low level diagnostics, it is possible to use led diodes. After power up, red led is blinking...booting system, then red led is off for short time and start to blink again, in this phase Katherine is looking for a chip on the board. When chip is found, orange led starts to means the chip was found and digital test (test of communication interface is tested) is in progress. When test is OK the orange led should be ON. So, when chipboards etc is ok, then red and orange leds are ON. When orange is blinking and then is OFF, it means there is problem on communication lines with chip, but it is possible to detect chip on board. You also check diagnostic report on: http://ip_of_your_katherine/log.txt

Useful Links

-- GabrieleDAmen - 2022-03-15

Topic attachments
I Attachment History Action Size Date Who Comment
PNGpng CV-test.PNG r1 manage 119.7 K 2022-04-20 - 16:27 DialloBoye  
PNGpng IV-bnl-00.PNG r1 manage 108.9 K 2022-04-20 - 16:27 DialloBoye  
JPEGjpeg localsetup.JPEG r1 manage 172.3 K 2022-03-17 - 15:46 GabrieleDAmen  
PNGpng mmtrack00.png r1 manage 114.8 K 2022-03-17 - 15:26 GabrieleDAmen MMTrackLab instructions
PNGpng mmtrack00b.png r1 manage 52.2 K 2022-03-17 - 15:26 GabrieleDAmen MMTrackLab instructions
PNGpng mmtrack01.png r1 manage 183.3 K 2022-03-17 - 15:26 GabrieleDAmen MMTrackLab instructions
PDFpdf test.pdf r1 manage 21.7 K 2022-04-04 - 11:21 DialloBoye  
PNGpng timepix3.png r1 manage 611.4 K 2022-03-16 - 14:56 GabrieleDAmen timepix3 readout ports
Edit | Attach | Watch | Print version | History: r21 < r20 < r19 < r18 < r17 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r21 - 2023-02-07 - GabrieleDAmen
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Sandbox All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright &© 2008-2023 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback