PR 431: https://github.com/SBNSoftware/sbndcode/pull/431 intended to correct for incorrectly-mirrored channel mapping on the FEMBs mounted on the top of the APA frames, but I assumed the number of them incorrectly. I had flipped half of the U and V channels on each side, but in fact only 26 FEMBs worth of channels are connected on the top, and the others are connected to side FEMBs. This only affects U and V, I had undone the Y flipping correctly. This was detected with data with tracks taken this morning (March 19). Plots below: unfixed event display, fixed event display, and the RMS vs channel for run 11840.
You can see the effect of the unmodeled gap between the two half-APAs in the Y view for the track that crosses the gap. The MC deadens six wires here, but we don't have enough channels to put all the data in them and still have six dead ones. Marco del Tutto knows about this and this will require a new GDML. For another day!
PR 431: https://github.com/SBNSoftware/sbndcode/pull/431 intended to correct for incorrectly-mirrored channel mapping on the FEMBs mounted on the top of the APA frames, but I assumed the number of them incorrectly. I had flipped half of the U and V channels on each side, but in fact only 26 FEMBs worth of channels are connected on the top, and the others are connected to side FEMBs. This only affects U and V, I had undone the Y flipping correctly. This was detected with data with tracks taken this morning (March 19). Plots below: unfixed event display, fixed event display, and the RMS vs channel for run 11840.
You can see the effect of the unmodeled gap between the two half-APAs in the Y view for the track that crosses the gap. The MC deadens six wires here, but we don't have enough channels to put all the data in them and still have six dead ones. Marco del Tutto knows about this and this will require a new GDML. For another day!