Hi, I have compared DSM response for good and bad events (i.e. with the wrong token in$ taken with JP & HT EEMC triggers. www-e-->trigger --> 4 entries for runs: #4143027 and #4135005 ===== ppLong-1 , EJP trigger ==== 1) for bad events the HT spectra with input to DSM-0 show ghost pedestal at value $ 2) TPsum spectra with input to DSM-1 show 'mini ghost pedestal' . It is _100_ times larger (comparing to the regular pedestal) then in the good events. This kills our jet-trigger ===== ppTrans-1 , EHT trigger ==== 3) 'ghost pedestal' at value of 4-5 is seen in in all HT spectra with input to DSM$ It is OK because we do not veto JPsums>300 now. 4) also 'ghost pedestal' is seen in JPsums , the same reason. The spectra have the regular pedestal too high, but it is before we shifted it In conclusion, #2 looks like a strong argument that wrong token in EEMC FEE data $ EJP trigger rate are strongly correlated, so if we fix token problem we it should $ EJP trigger , too. With regards Jan