[h0t5tuff]

The underground argon closed 6 mm below the re-entrance tube’s top, leaving a 55.5 kg copper lid in reentrance_tube_copper. this is an artifact of how the argon profile is made: the tube is open at the top. The argon now closes 1 µm (PROTECTION_GAP_POLYCONE) below the top, so the lid is about 20 g. Only one GDML point changes; I also tested with Geant4’s overlap check and it passed

[MoritzNeuberger]

I think at some point we really have to refactor this module since it is getting really difficult to understand what is going on (at least to me :D ). Your PR seems good to me though.

[gipert]

@h0t5tuff have you validated this? can you provide a plot that demonstrates the fix?

[h0t5tuff]

@gipert Yep, plot attached: top 10 mm of the tube, before vs after (this PR), with 1000 geantino vtxs sampled uniformly in reentrance_tube_copper.
Before: a 55.5 kg copper cone (24 % of the EFCu), and 240/1000 vtxs landed in it.
After: just a flat 1 µm slab (28 g) and no vtxs land there. The tube’s EFCu goes from 231.5 to 176.1 kg.