We are trying to understand some periodic noise that we have observed on 10K Thermistor measurements taken with a 34980a. It seems to occur on the 1st or lowest numbered channel setup to measure 10k-Thermistors. The reported output from six such thermistors can be seen in the chart below.
This is data of the output of 6 10K Thermistors that we would expect to operate very similarly and was surprised to see this pattern in our data.
The period between the “pulses” is approximate 10 min, 10 Sec. Other examples have shown the same period of 10 min, 10 sec.
If we feed this same setup into a 34972 we see nothing of these spikes. It almost acts like a “feature” of an internal calibration. It is extremely repeatable.
How can we best avoid or eliminate such variation in reporting?