I recently chrono'd my HW100 over a whole fill of air, just to see what variation there was.

It showed me that the shot to shot output varies by around 3fps with the very occasional 5-7fps.

There is a gradual change across the whole cylinder of air, it starts slightly high, drops gradually, then climbs back up before dropping off again - more or less like a sine wave. I assume this is a characteristic of regulator. I don't think this gradual change is really very important, as long as it's not too big. On my rifle it's only 4fps change.

Worst case the spread between the fastest and slowest velocity shots out of 113 shots was 10fps. The worst spread between any two consecutive shots was 7fps.

What does stand out, is that the shot to shot readings nearly always go high, low, high, low. For example +2.8fps, -3fps, +1.6fps, -3fps and so on...

I was curious if this is a real effect or just random chance and if it's real, what causes it.

I have yet to compare FPS readings with pellet impact point to see if there is a strong correlation. In theory, if this was predictable and did have an effect on impact point, you could adjust for it with a slight change in point of aim, alternating high, low. Trouble is you'd have to synchronise with the changes in fps otherwise you could make the effect worse!

Lastly, does anyone know what's involved in bringing the FPS spread down on the HW100? Is it even possible? Is a shot to shot variation of 3fps as good as it gets? Is what I'm measuring more to do with variation in pellets than what the mechanics of the gun are doing?