After much chronograph testing and many spurious readings (up to 897.5 fps) I revised my set-up and got:
Total mass of the launcher is 790 grams wet (with snow) which I think might be very light, 1.2 m long overall. It looks pretty much like my first sprinkler valve launcher but has a cell core reservoir about 4" longer and the blowgun pilot is unmodified from how I found it discarded at a construction site. I think that might be where all the parts came from actually! Oh, not the fill valve... I bought that in a pack of 4 for about a dollar.
It is sleek (less than 13 cm tall) and actually is comfortable to shoot from the shoulder. Barrel threads out for storage.
The Hunter valve is converted to barrel sealing and has been modified for better flow like I did to this Rainbird:
4.5 gram corks come out around 615 fps. 60 gram at 206 fps.
Why all this? I used to turn my nose up at sprinkler valves. My first launcher was a hammer valve from scratch with no lathe back in 2001. I figured to give sprinkler valves a try because they came free. Well turns out they are fine, and I speculate they are faster than most people think. My piston valves and commercial QEV are not significantly better (they are smaller ports and barrels) at the same pressure. If you are launching below 500 fps, and the valve has the cross section area you want, sprinkler valves are great!
If you have a faster sprinkler valve launcher let me know... I might need a push to try and build a better, lighter projectile
Video coming when I learn this new-to-me video editing software.
A word about the spurious chrony readings: styrofoam pusher wads gave a lot of them. On frame by frame (240fps) I did occasionally see bits of pink ahead of the projectile. That means probably the air is moving around 900fps or more to send debris past the projectile, and so perhaps a faster shot is just a matter of a lighter, better fitting projectile that dos not fly apart.
Other weird readings may be from extreme tip-off, where the projectile yaws hard as it exits the muzzle. One cork (marked 4.1) did that repeatedly and gave low velocity readings. It sounded like a ricochet from a .22 rifle. It was one of very few corks I could find, because it didn't travel far.
Also I backed away from the chrony a bit, stopped using masking tape around the corks (it was stripping off sometimes) and moved the backstop away (I think debris was sometimes bouncing back over the rear sensor on the Chrony.
