Will precision of gas table affect the simulation result?

Dear experts,

I used two different gas files for 1 track simulation in a drift chamber, they are:
20 steps of grid between 100-400000 V/cm and nColl = 10
100 steps of grid between 10-500000 V/cm and nColl = 12

The question is: will the result be different if I used more precise gas table? I didn’t see difference in the output file…

@hschindl that’s for you :slight_smile:

In general yes, the results will be (slightly) different. The nColl parameter controls the statistical error of the Magboltz calculation at each of the electric field values (higher nColl means better precision). I wouldn’t expect a bit difference between nColl == 10 and nColl == 12 though.
Increasing the number of steps makes sense if the transport properties (drift velocity, Townsend coefficient etc.) exhibit features that you can only resolve by increasing the granularity of the electric field grid. 100 steps is (usually) a bit of an overkill. The lower and upper limit of the electric field grid should be set such that it corresponds to the range of fields present your chamber.

See Gas files | Garfield++ for some more pointers on this topic…

Thank you for the explanation, I looked at the page you sent, it says “if I use DriftLineRKF and AvalancheMC”, does it mean gas file is not used if I use AvalancheMicroscopic to simulate the drift and avalanche?
And just another non-relative question… is there a function to get number of cluster in TrackHeed? I didn’t find so far…

Correct, AvalancheMicroscopic does not use the gas file, it uses directly the microscopic electron-molecule scattering cross-sections.

Do you mean the average number of clusters per cm? You can use the function GetClusterDensity for that. Or the number of clusters on a specific track? To get that number you need to loop over the clusters and count them.

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