So I've talked to a few more people internally about this and formalized my reasoning in the attached report.
I'd be curious to get some thoughts on it.
Obviously, it would be nice if anyone could give me a good rationale for Filtered Limits, but people seem to be onboard with my hypothesis.
My only sticking point is the prevalence in the late 70's - preceding the introduction of peak velocity in API - of pseudo peak (1.44xRMS or 1.57xRectifiedAverage) in the most common analyzers. Maybe my assumptions of peak referring to true peak in 1981 is not a good one...
(328.05 KB, 21 views)