A good strategy for when a powervalve needs to open is outlined by Urich and Fisher Holley Carburators and Manifolds, at least the older editions. It was written for tuning for oval, road and autocross racing, but the concepts apply. The underlying concept is that the power valve needs to open when the load is gets high enough that the engine needs enrichment. They test for the engine's needs by crowding manifold vacuum down to power valve opening point.

As their book was written in the late 70s, they of course did it with a stop watch on a closed circuit. But with todays eddy current dynos and consumer priced dataloggers, its possible to rework the strategy for these new tools.

Basically the idea is to start at a moderate load and speed, say 35 mph, and add enough throttle to drop the manifold vacuum to 14 in hg, and hold that vacuum while the car accelerates. Then do this again 2" lower, and again 2" lower down to power valve opening. Shrinker wrote that with a g-gas, the proper PV opening point is revailed by the change in combustion seen on a 5-gas. Since most of us don't have access to a 5-gas, we just have to look for where performance starts to drop off and while trying different jetting and PV opening points.

Let's say we're testing on an open road rather than a dyno. The car accelerates lightly when crowded to 16", 14" and acclerates a little more when crowded to 12" but kind of goes flat at 10" and 8". Below 8" the power valve opens and the car takes off like you expect. Next, we install slightly larger main jets and it accelerates about the same, but the rpms and g meter show that crowding 16" it's a little lazy from 50 mph up. Based on that, the next test would be to go back to the original jetting trying a 9.5 and 10.5 power valve.

Opening too early is a bit harder to identify. It should be a little slower on acceleration, but its tricky maintaining and repeating part throttle acceleration once the PV opens.