Fig. 8.

The noise occupancy measured from eight adjacent channels connected to 18 cm strips (~22 pF). Each point was measured from a million randomly timed triggers, each with a 150 ns window (a parameter that is adjustable on the chip). Note that the most probable signal from a 250 MeV proton is 9.5 fC. From the slope of the displayed linear fit we find a noise sigma of 960 electrons, consistent with the fits shown in Fig. 7.