After getting data logging earlier in the day, Chris installed the detector and we took a look at data.

The amp board gain is set to lowest setting.  Still, when looking at pedestal subtracted data, the preamps were picking up clear noise.

This may have impacted the pedestals, which were still noisy with preamps off, but long cables/detector connected.

Therefore cables disconnected, and noise demonstrated to be expected (Attachment 1).  In this and subsequent plots, the x-axis sample number (approx 0.36ns/sample) and y-axis is ADC (roughly 1mV/count).

As a training exercise, Chris configured and logged these data sets.

Detector was reconnected and powered, with reference plots for ch0, ch1 in subsequent 2 attachments.

The noise is clearly correlated between channel pairs, as they are interleaved/looking at same signal.

Looking at a different channel (ch6), the noise isn't so clearly correlated.

Because each trace is 4k samples, it is easy to focus on outliers, so a less biased approach is to look at the ensemble distributions.

An example of the two conditions for ch. 1 is posted as the last 2 plots, where the RMS spread is seen to increase from about 2 to 7 ADC counts RMS.

Interesting to compare values in the tunnel.