Strange timing (and reading) behavior of multipurpose temp sensor driven by scada report period

[jessicamillar]

It appears that at the top of each reporting period, the temperature sensing is captured at its expected rate, after almost exactly a minute, it drops to less frequently than expected. In addition, in the first minute the average temp is between .5 and 1 degree F lower than the rest of the time. This second issue is particularly important since is affects our numerical calculations of energy use will hamper our ability to do thermostatic control of room temps.

We only see this at high capture rates. We have set the capture period to 1 second on two temp sensors at Beech: statcheck (Room Temp) and outdoor.air.temp (OAT). The graphs below show the same hardware layout on beech, with the SCADA report period at its usual 300 seconds (First two graphs) and then dropped to 150 seconds (third graph). Excel spreadsheet with the relevant data for these two sensors also included.

Beech 1:1:24 Graph.pdf

Also sending the csv file (as an xlsx) that was sent at midnight on 1/1/24, showing the messages from "statcheck" (Room Temp) and "outdoor.air.temp" (OAT). We have color coded them to show the fast report times (red) and the slow report times (yellow).

SCADA Report for beech-20240101-0500-24-hw1.isone.me.versant.keene.beech-scada.report.a.001.xlsx