The pedestal width when connected to the 16-bit boards was about a factor of 2 worse than the 18-bit board on the L-HRS (1000 ppm compared to 500 ppm). The pedestal width on the right was similarly noisy.
On the R-HRS, there was a much worse problem of the pedestal having a mean value of 30k - 40k ADC channels when the HV cable was connected. This was also not stable, and occasionally jumped to much smaller values for some period before returning to the 40k channels. The pedestals were remarkably similar between the 0.45uA channels on each of the boards.
Tomorrow we will chase this during the target power supply repair access.
Similar problems were seen with the lumis, of course, and we licked it
by breaking the ground line with a capacitor. It may be that we will need
some similar ad-hoc fix this time.