1. Here is the LED vs gain variations for all of the 63 channels www.star.bnl.gov/protected/spin/yuxip/SlopeVSday/.

   The overall gain variation for each cell (day-by-day) is estimated from slopes of ADC spectra. Note that some of the cells do not have a valid slope due to the lack of a threshold shape in their ADC spectra.

   There is a common trend in the small cells that LED peaks dropped by ~ 10% from day80 to day89, while it is hard to see if there is any correlation between the overall gain shift and LED peaks. Because the slopes were not esimated very well.

2. Here is the averaged LED among cells (small/large) within the same channel vs slopes:

   www.star.bnl.gov/protected/spin/yuxip/AllrSmlLedSloVSday.pdf   --> averaged over small cells within the same channel

   www.star.bnl.gov/protected/spin/yuxip/AllrLrgLedSloVSday.pdf   --> large cells

3. Then I turned to estimate the overall gain shift by pion mass for each day (80-89)

   www.star.bnl.gov/protected/spin/yuxip/dPion_Sml.pdf  -->pions with daughter photons impact small cells, over 8 days

   www.star.bnl.gov/protected/spin/yuxip/dPion_Lrg.pdf  -->daughter photons from large cells

  relative LED position averaged over small cells vs pion mass: www.star.bnl.gov/protected/spin/yuxip/AllLedVSday_Sml.pdf  It shows that only part of the LED variation (~10% ) can be explained by the gain shift (~2%). The rest is due to the change of pulser intensity.

  averaged LED for large cells vs pion mass: www.star.bnl.gov/protected/spin/yuxip/AllLedVSday_Lrg.pdf

  Because of limited statistics I didn't compare LED vs pion mass for each cell.