Surface leakage currents in these detectors are not very stable. Measurements are repeatable in a coarse sense but values can fluctuate by factors of two or three within a period of 10 minutes. Despite this, a few consistent features seem to be present.
This plot shows leakage current between pixel and strip on detector Y-2 (9.63 mm). All strips are at 0V. Voltage is applied to all pixels. This is the total leakage into strip 8 from all pixels. Strip 8 response is reasonably typical of all strips. The leakage increases rapidly for voltages above +30V. When the polarity is reversed the leakage is much less and nearly linear with voltage.
This plot shows leakage current between pixel and strip on detector EV-3 (5mm). The EV detectors also show nonlinearity but the difference with polarity is less.
Detector Y-2 does appear to have a bad cell. Leakage current between pixel row one and strip one is an order of magnitude higher.
Strips are numbered left to right 1 to 8 viewing the detector from the cathode through the crystal toward the anode. Pixels are numbered bottom to top 1 to 8 (see pinout).
This plot shows total cathode to anode leakage vs. cathode voltage. All anode contacts are at 0V and the cathode is at negative HV. (The measurement is quite light sensitive. These readings are taken with the detector in darkness.)
The guard ring collects most of the current. At 500V the guard ring collects 85%, 44nA out of a total of 52nA. That leaves only 7nA of bulk leakage which would be spread between all strips and pixels.
Detector EV-11 shows lower leakage. The guard ring collects 74% of the cathode current.
Cathode leakage varies widely between detectors.
