...
we can see the fitting line (forced across origin) deviates a lot compare with the normal pixel.
Some studies with @philip_adam_hart on the sagging region (high #column ), we take the photon counts ratio between single-pixel-cluster and single & double-pixel-cluster (run29, -50V), it seems like we have less two-photon cluster in sagging area (top: Module1, bottom: Module2)
We got smoother date at -110V (run 148, primary beam energy 11kev )
I guess the sensor really doesn't work well at -50V.