Too small number of avalanche electrons


I’m trying to simulate the multi-wire proportional counter with TrackSRIM and AvalancheMicroscopic classes. I succeeded to use the TrackSRIM, and then use AvalancheElectron at each cluster points. What I anticipate is the large numbers of avalanche electrons near the wires, but only a few (almost 2~3) electrons are generated near the wires. I checked that the number of electrons increases as I increase the anode voltage, but It was also smaller than I expected.
Because the wire spaces, voltages, and other conditions are same as one of the real experiment, I think there should be avalanches if I don’t miss something.

MWPC_Proto.C (11.3 KB)
isobutane_gas.txt (25.3 KB)
Xenon_in_isobutane.txt (2.3 KB)

I guess @hschindl can help.

I’m afraid I can’t judge offhand what’s the expected gain for pure isobutane at this pressure and for this particular field configuration. To understand/debug the issue, I would suggest to start with a simplified program and then gradually add complexity.

  • As a starting point, you could perhaps approximate your chamber by a 2D geometry (e. g. a row of wires between parallel plates) which you can model using ComponentAnalyticField?
  • Make sure that the gas file you are using covers the entire range of electric fields encountered in your chamber.
  • Then simulate a single electron drift line using DriftLineRKF and calculate the average gain.
  • As a cross-check you can use AvalancheMicroscopic instead; the average of the avalanche size distribution should be consistent with the average gain you got using DriftLineRKF.

If the results are in the right ballpark, I would then move on to the 3D geometry (i. e. neBEM) and subsequently to simulating drift lines from a track.

Let me know if you encounter any technical problems…

In general, TrackSrim and AvalancheMicroscopic are not really a good match. TrackSrim is normally used for projectiles that deposit a large number of primary electron-ion pairs along their track. AvalancheMicroscopic on the other hand is typically used for problems where fluctuations are important. If you are dealing with hundreds or thousands of primary electrons than these fluctuations are largely washed out, and it’s really an overkill to simulate them all on a microscopic level.
Instead, I would recommend to use DriftLineRKF (with the “ion tail” option switched on, and the “scaling factor” set the to the number of electrons in a cluster).

Thank you for the explanation.
I should check the gas file again and use DriftLineRKF starting with more simplified geometry.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.