Your data is both interesting and suspicious. It's too repeatable for the spikes to be random noise, while the detail view shows a kind of bi-modalness that isn't normal for manufacturing defects. (Pun on the word "normal" not originally intended.) The spikes are large in relative amplitude and seem to alternate between + and - deviations. Overall, it has more the feel of an electronic or numerical anomaly than a physical one.
You referred to some kind of electronic multiplier -- that's where I'm placing my bets. My guess is that the regular spikes every 10 samples are when the algorithm makes adjustments based on its observations over the recent past. That combines with some quantization (I'm guessing) such that those spikes represent one quantum of change somewhere in the system.
During the 9 near-constant measurements between the spikes, there appears to be a smaller variation. Again, just guessing, but I'll venture that the electronic multiplier is generating a constant freq there but measurement quantization toggles between 2 measured values. Meanwhile the 1 in 10 spike is an adjustment or quantization generated by the multiplier.
If you're stuck measuring after the multiplier, can you average across at least 10 or 20 intervals?
-Kevin P
ALERT! LabVIEW's subscription-only policy came to an end (finally!). Unfortunately, pricing favors the captured and committed over new adopters -- so tread carefully.
