Skip to content

MCH: Carry the slat geometry in double - #16152

Open
sawenzel wants to merge 1 commit into
AliceO2Group:devfrom
sawenzel:swenzel/mch-slat-double
Open

sawenzel wants to merge 1 commit into
AliceO2Group:devfrom
sawenzel:swenzel/mch-slat-double

Conversation

@sawenzel

Copy link
Copy Markdown
Collaborator

This removes sub-nanometre gaps and overlaps between the layers of the MCH slats (stations 3–5). The layers were stacked from float sums, so neighbouring faces overlapped or were up to 2e-8 cm apart.

With TGeo navigation these faces produced empty MCH hits (zero length, zero energy loss), 8–10 % of all MCH hits in pp. With VecGeom navigation the gaps gave zero-length steps in the chamber air. Constants and placements of Station345Geometry are now double, and a unit test checks that the slat layers touch exactly (342 near-touching face pairs before).

With TGeo, the empty hits drop from about 1000 to 88 per 6000 pp events, and kVecGeom/kTGeo MCH hits go from 0.905 ± 0.026 to 0.992 ± 0.026.

Assisted by Claude Code.

This removes the sub-nanometre gaps and overlaps between the layers of the MCH slats and adds a unit test.

- The slat layers were placed from float sums, so neighbouring faces overlapped or were up to 2e-8 cm apart.
- TGeo navigation wrote empty MCH hits at these faces, and VecGeom navigation took zero-length steps in the gaps.
- The constants and placement coordinates of Station345Geometry are double now.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@sawenzel
sawenzel requested review from a team and shahor02 as code owners November 18, 2026 14:50

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Development

Successfully merging this pull request may close these issues.

1 participant