In the later months we have engaged in the market survey for the sensors of the CMS MTD. A long time has passed since my previous post focused on the beta setup and I have learned a lot. However, I was still not able to reach the final expected time resolution for the detectors. After Federico from Torino spent a few days in our lab, for which I am very thankful, we finally were able to set everything up and reach results compatible with theirs (Torino’s) in our setup. The plot below shows the time resolution obtained for a pair of identical detectors (FBK, UFSD4, T10, GR3_0, deep gain layer):
Two additional points were measured at higher voltages but one of the detectors went into breakdown. As seen, at 180 V a time resolution of (31.2±1.5) ps was achieved, which is compatible with Torino.
The two devices were mounted in two Chubut boards, respectively, placed one on top of the other such that the betas can go through the two detectors. The first stage amplifier is in the boards themselves, while as second stage amplifier our EBay amplifiers were used. After this, straight into the oscilloscope. The devices were placed inside the climate chamber and the temperature was set to -30 °C.
There were several reasons why we could not reach the expected time resolution, but among the most important, and this was a surprise to me, was the vertical scale in our LeCroy WaveRunner 9254M oscilloscope. Surprisingly, if the vertical scale is greater than 100 mV/DIV the noise increases considerably. This is because of some circuitry internal to the oscilloscope, as one can hear a relay switching when this vertical scale value is crossed. Other contributions were the ambient light, the temperature, and we using a reference trigger that was not good enough.
For reference, here are the results from Torino: