Restrict tracking volume to mag field volume

[tvami]

I am updating ldmx-sw, here are the details.

What are the issues that this addresses?

Resolves https://github.com/LDMX-Software/ldmx-sw/issues/1562 Alternative solution to https://github.com/LDMX-Software/ldmx-sw/pull/1632

Check List

• [x] I successfully compiled ldmx-sw with my developments.
• [x] I read, understood and follow the coding rules.

• [x] I ran my developments and the following shows that they are successful.

[tvami]

Interesting this didnt seem to have solved the issue

in https://github.com/LDMX-Software/ldmx-sw/actions/runs/14349763985/job/40226890572?pr=1685

  < 22:35:18    PROP           ERROR     Step failed with MagneticFieldError:1: Interpolation out of bounds was requested
  ---
  > 06:28:17    PROP           ERROR     Step failed with MagneticFieldError:1: Interpolation out of bounds was requested

@bloodyyugo @EBerzin do you have any other ideas? Maybe it's still worth to merge this as the HCAL part works, and the tracker volume restriction is probably towards the solution.

[bloodyyugo]

Interesting this didnt seem to have solved the issue

in https://github.com/LDMX-Software/ldmx-sw/actions/runs/14349763985/job/40226890572?pr=1685

  < 22:35:18    PROP           ERROR     Step failed with MagneticFieldError:1: Interpolation out of bounds was requested
  ---
  > 06:28:17    PROP           ERROR     Step failed with MagneticFieldError:1: Interpolation out of bounds was requested

@bloodyyugo @EBerzin do you have any other ideas? Maybe it's still worth to merge this as the HCAL part works, and the tracker volume restriction is probably towards the solution.

Do you think these errors are coming during the extrapolation to the HCal but not in the track finding/fitting itself?

[bloodyyugo]

Interesting this didnt seem to have solved the issue in https://github.com/LDMX-Software/ldmx-sw/actions/runs/14349763985/job/40226890572?pr=1685

  < 22:35:18    PROP           ERROR     Step failed with MagneticFieldError:1: Interpolation out of bounds was requested
  ---
  > 06:28:17    PROP           ERROR     Step failed with MagneticFieldError:1: Interpolation out of bounds was requested

@bloodyyugo @EBerzin do you have any other ideas? Maybe it's still worth to merge this as the HCAL part works, and the tracker volume restriction is probably towards the solution.

Do you think these errors are coming during the extrapolation to the HCal but not in the track finding/fitting itself?

...if so probably what we need to do is use ACTs to propagate to the edge of B-Field and then do linear extrapolation (by hand, probably) to where ever we want.

[tvami]

Do you think these errors are coming during the extrapolation to the HCal but not in the track finding/fitting itself?

Not the HCAL, I removed the HCAL extrapolation, so whatever msg is left is just for extrapolating in the tracker volume

[tvami]

Non of the distributions change tho, are we sure that this part of the code is not overwritten somehow later on from the GMDL maybe @bloodyyugo ?

[bloodyyugo]

Non of the distributions change tho, are we sure that this part of the code is not overwritten somehow later on from the GMDL maybe @bloodyyugo ?

I thought I tracked it all the way down to what you changed...I can look again.

[tvami]

...if so probably what we need to do is use ACTs to propagate to the edge of B-Field and then do linear extrapolation (by hand, probably) to where ever we want.

So this is what this PR does now. And still the mag field error is shown.

There are two issues really:

• HCAL extrapolation being too much in Z --> this is now resolved
• Having low pT tracks going too much in Y --> I was hoping to resolve this by restricting the tracker volume (as we discussed a few weeks ago), but the error msg is still there

[tvami]

I'll factor out the HCAL change in https://github.com/LDMX-Software/ldmx-sw/pull/1687

[tvami]

@dependabot rebase

[tvami]

hey @bloodyyugo I came back to this (now that the DR writing is less crazy) and have the scope of the PR simplified. This leads to absolutely no effect on the msgs we wanted to get rid off, like:

  < 22:35:18    PROP           ERROR     Step failed with MagneticFieldError:1: Interpolation out of bounds was requested
  ---
  > 06:28:17    PROP           ERROR     Step failed with MagneticFieldError:1: Interpolation out of bounds was requested

like really not a single case got fixed.

[tvami]

if you look at the log here https://github.com/LDMX-Software/ldmx-sw/actions/runs/16161339701/job/45614175339?pr=1685 I think things actually got worse, or stayed the same... Given that trend, I could go in the other direction and make the sizes bigger... let's see

[tvami]

OK so bigger makes absolutely no changes in the logs

[tvami]

So I tried a few things... in the end what I learned is that this is already there at the CKF level, it was just hidden, now I brought the ACTS level to be the same between GSF and CKF

[tvami]

I'm closing this again, let's see if @bloodyyugo has any other ideas and then we can open it again