Since the 1877 TDCs are multihit modules, I checked if the problem with uncorrelated ADC and TDC data is due to multiple TDC hits. As far as I can tell, this is not the case. Roughly 1% of events has more than one hit in one of the 24 scintillator TDCs (s1 and s2) in run 1189. This percentage is way too small to explain the observation.
I am a little confused as to what a value of 0 means in the TDC. Usually, the TDC reports overflow (e.g. 4096) if there is no input. Not sure how this one works, 0 may be fine, or it may indicate a bad TDC. Bob: any comment?
I do notice that all three detectors with this problem, L.s1, s2, and cer, share a single TDC module, crate/slot = 4/23. I would look for a hardware problem with this guy (bad common stop signal, bad TDC module altogether). Even if the input is fine, the module may not be.