extend/fix track filtering after MR2135.

Incorporating modifications by @se.gorbunov in track fit. Fix and extend track filtering. Add new vertex histos. Try suggestion by @se.gorbunov to tune STS pulls by modifying

fFitter.SetDefaultMomentumForMs(1.);

added QA alignment mCBM labels

requested review from @se.gorbunov

assigned to @se.gorbunov

Dear @f.uhlig, @p.-a.loizeau, @v.friese,

you have been identified as code owner of at least one file which was changed with this merge request.

Please check the changes and approve them or request changes.

added CodeOwners label

@se.gorbunov remember our meeting last week in GSI. Here is the problem:

The pulls in x for e.g. STS U0 are as attached:

while for the y direction they are :

So y error propagation is correct while x is only 1/3 of the correct value (Sigma == 1). You suggested to modify the momentum for MS as in

fFitter.SetDefaultMomentumForMs(1.);

Correct ? Why should be x and y different if MS affects them both ?

added 1 commit

• 9add4310 - fix MS momentum for mCBM25 to 1.5 GeV/c

Compare with previous version

added 4 commits

• 9add4310...268286ba - 2 commits from branch computing:master
• 17800ccb - extend/fix track filtering after MR2135.
• a9bd245c - fix MS momentum for mCBM25 to 1.5 GeV/c

Compare with previous version

marked this merge request as ready

resolved all threads

@a.bercuci, Do I understand it right - you have changed the default momentum for MS from 0.5 GeV to 1.5 GeV and the puls changed like this:

Pull X 0.36 -> 1.19
Pull Y 1.09 -> 1.40

And the question is, why does the multiple scattering in the track fit have a different effect on X and Y?

I would say it is because the resolution in X is better than in Y. The extrapolated track error is a sum of the measurement error and the MS. In X, the measurements are more precise; therefore, the error of the extrapolated track is dominated by the Multiple Scattering effect. In Y, the resolution is worse and MS doesn't add that much to the extrapolated track error.

you have changed the default momentum for MS from 0.5 GeV to 1.5 GeV and the puls changed like this:

yes.

And the question is, why does the multiple scattering in the track fit have a different effect on X and Y?

yes.

Now the question is : If I tune only MS through p, I can fix either X or Y, but not both in the same time. Obviously one has to tune also the error parameterization of STS in order to get the pulls right, for both X and Y in the same time. Do you have this experience in simulations. How to proceed here ?

Nevertheless, for the current MR, I don't think we have to fix such issues. So I would say that from my side, the MR is done !

I think we should distinguish between the intrinsic STS error and the alignment uncertainty. If the difference in XY comes from the alignment (as I suspect), we should probably not correct for it in the track fit.

I resolved the thread in order to merge the MR.

added 29 commits

• a9bd245c...e1d71f8f - 27 commits from branch computing:master
• 837a4c03 - extend/fix track filtering after MR2135.
• 9d3580d2 - fix MS momentum for mCBM25 to 1.5 GeV/c

Compare with previous version

resolved all threads

enabled an automatic merge when all merge checks for 9d3580d2 pass

resolved all threads

merged