After moving Cerenkov ADC to ROC4 and puting into 13-bit mode by Alex, the overfload issue of its ADCs is resolved. From the plot you can see that there is not shape peak at highest channel and the ADC don't saturate at low channels.
For PRL2#14 overfload issue, I reduced its HV from -908V to -850 and now as you can see from figure.2, its ADC distribution looks normal by comparing with other channels, and overfload peak has gone.
Figure 1
Figure 2