Straw Tracker

The Straw Tracker is being developed for the 2015 BL4S competition. The detector formed part of the STT in the ZEUS experiment. The STT project page, with some photos of the detector, is at:


Date Action
2015-05-15 Integrate UDP recv into DataChannel
Tested correction to error handling code
Received data from udp_sender_receiver test program, recoding it to disk
2015-05-13 Integrate UDP socket code into rcd_trb Module
Looked at RAW event format
Worked on error handling code
Loaded module on BL4S DAQ hardware and tested error handling
2015-04-24 Discussion of Readout architecture with Markus and Jorgen
Jorgen proposes writing a RCD_empty package into which we can add TRB specific code
A TRBscope program should be written to capture data from the TRB
The TRBscope can explore the data format and understand handling of UNIX sockets
A TRBemulator could be written to send arbitrary data at arbitrary times, or replay a sequence of packets from the TRB
Ultimately the TRBscope code should be integrated into the RCD_empty package to form an RCD_TRB package that can run in the BL4S data-taking partition
2015-04-08 Working with Elena Zarkh
Set up DAQ PC and electronics
Dual timer is tricky to start up as a pulse generator
Used switch box to view test pulses from front end electronics
Found that the TRB needs a LVDS trigger, when disconnected it free runs to around 40kHz
The trigger box (based around one of the level converter cards) accepts TTL level signals and outputs LVDS across all channels
Some channels seem to be flipped or offset - check trigger box outputs when setting up
Outputs are on a pin header so one can use the switch box to probe the channels, connecting to a scope over LEMO cables
The discriminator for one of the channels had a very narrow pulse width, killing the coincidence rate - we adjusted this up to ~100ns
We measured the coincidence rate of the scintilators to be just below 10Hz
A small test run did not collect any meaningful statistics
Overnight run started ~21:50
2015-04-07 Meeting with Elena Zarkh
Started HV for overnight conditioning
Found DAQ PC( had a dead power supply
Borrowed PSU from bl4sdaq1 for the moment
2015-04-02 Meeting with Konstantine Zhukov
Started gas flow to condition straws
2015-03-13 Visit SR1 to see CERN's STYX module
Check the gas system in SR1
Mix the gasses (Ar/CO2)

List of Issues

  • Trigger does not work properly: Signal is too low?

  • One channel with trigger signal rate goes up to only 1 evt/sec?

  • Pulse width of one of the PMT's is too narrow ?---Need to be investigate!
    • This is in fact design. The larger PMT has a wider signal after the discriminator and the threshold for it is set higher. The small PMT has a lower threshold (almost in the noise) and should have a narrow signal. This, according to Steffen, is the standard way of making such coincidences. He was also surprised at the low rate though. Maybe one really has to scan the scintillator HV again, adjust the thresholds, check the timing etc. more carefully.
  • 2 out of 6 FE boards with not enough events ?
    • It has been seen in the past that this can be caused by the gas not being on long enough. Given that you turned on the gas the week before Elena came, it is probably not the cause, but it is something one can check by taking data at different times after the gas has been turned on.
  • Connection cables between FE boards and Driver boards could be broken?

  • The reason of the wiring crossover : connection from FE boards to STYX box?
    • This is due to the fact that the TRB can take 31 signals on each connector and the 32nd channel is for the trigger. It is best if the outermost straws are not rad out. However, these straws are one something like the 4th readout channel - this can be seen in the mapping file. Steffen calls this the “fudge cable”!


The StrawTracker modules are used in the Straw Tube Young student eXperiment (STYX). Lots of information on their setup is available on their project pages.


The STYX analysis software is held in a git repository at Bonn.
git URL
Use the credentials listed on the SysAdmin page.


On the styxbn machine, in ~styx/bin are two scripts: and Both of these call the perl script to interact with the TRB. This script starts the daq_evtbuild executable from ~styx/event_builder/hadessoft. It's console output is into a new xterm and it writes raw data files into ~styx/data/etraxp114 (this is the TRB processor name).

The daq_evtbuild program is part of the Hades DAQ software. It programs the TRBs FPGA and TDCs, configures the triggering and readout to onboard memory and transmits packets over UDP [1].

udp_sender_receiver.cpp test program is under src/test directory in rcd_trb repository :

Resources for "git tutorial"


To use GCC 4.8 and ROOT 5.34, execute the following:
source /afs/
export CC=$(which gcc)
export CXX=$(which g++)
source /afs/
This allows cmake to find the requisite libraries.


After preparing the environment as above, run cmake with the CMakeLists file, then run make:
cmake CMakeLists.txt


The STYX setup is located in SR1. Access is controlled under the ATL_SR1 zone.


Front end

16 channel front end cards are available with ASDQ chips. These perform discrimination against an input threshold voltage (~<1V). They output multiplexed timing signals, 200ns per straw, for six straws giving a total readout period of 1.2μs.


Digitisation is done with the TRB[2] system. The TRB has 4 HPTDC chips[3]. The TRB configuration has a resolution of 100ps and a readout period of up to several microseconds. The data is zero suppressed and forwarded to a PC over a network. The PC hostname is and the user credentials are listed on the SysAdmin page.

The PC acts as DHCP server for a private network (192.168.0.*) on which the TRB resides. The PC also hosts the filesystem for the TRB from it's /var/diskless/ directory. Control and readout scripts are located in ~styx/bin/.


HPTDC manual:

Adaptation for BL4S

For BL4S, we might use the front end cards and digitise the signals with standard VME TDCs. Alternatively we could use the TRB and build events by embedding the TRB packets. We will need to understand the data volume of the system in designing the ReadOutSystem.

Alternatively we may be able to read in data from the TRB over ethernet.


NIM Modules
  • Coincidence unit,
  • Discriminator,
  • Pulse Generator,
  • Power supply +-%5V,
  • Thershold Volatge 2V,


ASDQ FE board

  • 2 Long FE Boards ->96 Straw connected.

  • 1 Short FE Boards->72 Straw connected.

ASDQ FE chip


The detector requires ~ 2kV power supplies.

The front-end cards will require low voltage supplies.


A standard ArCO2 mix is used. This is the same as is used in the ATLAS TRT. It is available pre-mixed.


The detector element at CERN is mounted horizontally to collect data from cosmic rays. For installation into the BeamLine, we will need a vertical mounting solution.




-- TimBrooks - 2015-03-01

Topic attachments
I Attachment History Action Size Date Who Comment
JPEGjpg Straw_electronics.jpg r1 manage 4108.8 K 2015-04-01 - 18:28 CandanDozen Straw Electronics
Edit | Attach | Watch | Print version | History: r22 < r21 < r20 < r19 < r18 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r22 - 2015-06-04 - TimBrooks
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    BL4S All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright &© 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback