Thermal telemetry occasionally gets channels wrong

[chazshapiro]

This happens from time to time, where the channels are getting mixed up. Here, the dewar T is temporarily being swapped with the heater power on I channel. Also the CCD T channel is either being dropped or swapped with something that's 0.

I think this is a race condition where there are multiple requests to the same Lakeshore in quick succession.

Not a huge issue, but it will interfere with alerts once those are set up.

[astronomerdave]

With what regularity does this occur?

[chazshapiro]

Scanning through the plots...

10/18 10/15-10/16 Then looks OK going back to 9/17, and then there's a long non-logging period before that.

Anecdotally, I have noticed it in the past, so it's not new.