I'm submitting a ...

• bug report
• feature request

Checklist

• I've added at least 1 label under Labels
• Assigned a reviewer under Assignees

Copied from slack message.

Hi folks! Rachel has made some interesting plots that show that something is still up with our TPC z-positions in the current tags.
Status as I see it:

• Earlier in the year we saw that the TPC was offset in the GDML by ~5cm. Marco identified this and made a PR: Update to GDML v02_05 - Fix TPC Z Dimension #760
• Linyan mentioned at the workshop that despite the fix they were seeing an offset the other way by ~1cm. I took a look at diff sbnd_gdml_v02_04_base.gdml sbnd_gdml_v02_05_base.gdml and you get the following.

138c138
<               <constant name="dimDetector_Z" value="5094.000" />
---
>               <constant name="dimDetector_Z" value="5001.000" />

• Based on the description of the PR I think this is a typo and the intended corrected value was 5010.000 - this explains what Linyan saw.
• However, beyond that Rachel is seeing something else. She's looking at the difference between the reconstructed and true positions (vertex) and finding a ~5cm bias in z (no bias in x or y).
• This suggests to me that there is at least one more parameter in the gdml that needs adjusting. It's possible (especially since we often use the gdml with 'no wires' that the truth and reco are picking up on different pieces of information.
• We'll try and dig into what & where but I thought you should be aware this is a bug that should be on the planning for the next validation release.