In the process of investigating the pion rejector mystery, I also turned on and off "random" channels trying to get insight from the HV map. In the end, I found this as confusing as the rest. The following plots show the adc spectra for prl1 and prl2 (cut above the pedestal) and the hand drawn arrows show the expected off channels (due to HV being off). So it looks to me that the map for prl1 is ok while the map for prl2 is off by 16 channels.
Figure 1
Figure 2
Figure 3