Possible misplaced/rotated SR mask in commonMDI version 1

[Victor-Schwan]

Check duplicate issues.

• Checked for duplicates

Goal

Hi all,

I wanted to raise here something that came up in private chats with @BrieucF and @atolosadelgado so that it’s documented and visible to everyone.

In the current commonMDI v1 geometry, it seems that the SR mask might be misplaced or rotated. To illustrate, I attach two graphics from material scans:

mdi0 – looks fine

mdi1 – possible bug

Before jumping to conclusions, I’d like to confirm the coordinate convention:

My understanding is that the x-axis points outward of the FCC circle, so that the incoming beam pipes are located at negative x.

If this is correct, it looks as if the SR mask (at around 2100mm) in v1 is at least partly inside of the outgoing beampipe rather than the incoming one.

Could someone confirm whether this interpretation is correct, and whether the SR mask placement in v1 needs to be revisited?

Thanks a lot!

fyi @danieljeans

Operating System and Version

Ubuntu 24.04.3 LTS

compiler

GCC 13

The version of the key4hep stack

887f1a4

Package Version

887f1a4

Reproducer

cd $k4gDir
export IF1=${k4gDir}/FCCee/ILD_FCCee/compact/ILD_FCCee_v01/ILD_FCCee_v01.xml
g4GraphicalScan -P -c $IF1 -s ZX -x -300,300 -y 0 -z 0,4000 -n 600 -o mdi1_beampipe.root
git switch mdi0
g4GraphicalScan -P -c $IF1 -s ZX -x -300,300 -y 0 -z 0,4000 -n 600 -o mdi0_beampipe.root

Additional context

No response

[aciarma]

Hello Victor,

please see my PR on this: #527

[Victor-Schwan]

Hi Andrea,
thanks for providing the fix. It seems that only the beampipe wall was rotated, while the actual tungsten mask was not. Since the tungsten accounts for most of the absorption, I do not expect any significant changes in the background levels for any detector model. Do you agree?

[aciarma]

Reasonable, but please consider that:

• this is a preliminary version of the mask, where there is a "beam pipe layer" on top of it as you can see from those plots. An upcoming version instead does not have the notch and the mask is all tungsten -for the same dimensions- so this may have an effect.
• for single beam backgrounds (SR, beam gas, losses, ...) it is under development a dedicated interface with FLUKA for handling the radiation interaction with the MDI elements, including this mask, before the tracking in ddsim.

[BrieucF]

Hi Andrea, thanks for providing the fix. It seems that only the beampipe wall was rotated, while the actual tungsten mask was not. Since the tungsten accounts for most of the absorption, I do not expect any significant changes in the background levels for any detector model. Do you agree?

Hi Victor, significant changes will come from replacing air by vacuum in the pipe (see #430). And as Andrea said, these masks are very preliminary, finding their optimal position and size will be an iterative process.