Thanks for the tip on lighting, but the difference is not due to manual valve, since I measured the speed on the cannon with an electronic valve, with pressures within about 2-3psi. And no, I am not using pressures above 50 for any of my air cannons (3, a mini one I finished today, the big powerful one where I tried measuring speed, and a tiny manual one that I built to test enclosures). Also, I never thought to wrap the chamber in tape to contain an explosion, but that seems like a good idea, so I'll add it to mine. Thanks for all your help.
What I have observed is that there is enough residual pressure after the fisrst launch, when using an ASCO brand valve, similar to the sprinkler valve, to do a second launch, while a similar opening and closing of the ball valve leaves almost no residual pressure. The volume between the valve and the projectile is about 2 inches of 1inch steel pipe. The volume of the tank would hold 25 pounds of liquid carbon dioxide. Probably a smaller tank would be adequate, but smaller tanks were not available for free, and low cost was a big consideration at the time.
I would tend to trust the measurement of speed based on a video camera. I think bright lighting helps the video camera work with a fast shutter speed for minimum smear when photographic fast-moving objects. If you have any doubts you can do a video of an auto at speed. If you are using a manual valving arrangement then I would expect that could explain wide differences in velocity.
I also hope that you are not using high pressures (like 100 psi or more) in PVC pipe. Even though the pipe may have a rating like 480 psi at 73 deg F there are many documented cases of these pipes exploding. Be aware that PVC is not approved for compressed air applications.
Your modification to open the ball valve quickly does sound like it should be efffective. At about $12 I think the electric valve from Home Depot is a simple and effective valve.
I measure 60% of the work ends up in the projectile kinetic energy based on ideal adiabatic expansion of the gas. I might be able to improve that value with a better valving system.
If you have computed the efficiency of your energy transfer I'd be interested in knowing it. If not i'd be glad to compute it for you. I'd need the muzzle velocity, projectile weight, volume and pressure of the compressed air, dead volume between the valve and projectile when the launch begins, effective barrel diameter, and barrel length.
It is not that hard to open a ball valve quickly. We use a similar method to open the air-drive valves on our commercial crash sled systems, which work quite well.
The secret is a spring and a lever to rotate the valve, arranged so that the mechanical advantage for the spring increases as the valve rotates toward the full open position. Of course, the mechanism must be designed so that the ball valve is not driven past full open. Of course some sort of latch is required to keep the valve closed until launch time. Another option is to use an air cylinder in place of the spring, which is what our systems use. The valve opens in much less than 100Msec once it starts moving, which is delayed until the pressure rises enough to start things moving. Very simple, but quite effective.
One model of my air launcher, with a ten-foot tube, will send a paper "rocket" as high as most of the Estes-brand rockets fly. Of course, that is with 100PSI in the tank and a cup of water to improve the seal.
Unfortunately, I have already tried that, but mine was just too inconsistent , I got everywhere from 80fps to 270fps. I don't know what it is, but it looks like I'll need a different method. As for the ball valve, I found that I works just fine. My first air cannon used a 1-1/2" ball valve, and since it didn't open smoothly, but stuck closed, then opened quickly when turned, the gun worked very well. At least, until I got sand in the valve, but that's beside the point. I'd guess a 2" ball valv would allow even more air, and therefore probably work for this as well.
To measure the launch speed I use a 30 frame/sec video camera which is fairly standard. By filming the launch perpendicular to the direction of travel, you can measure the distance the projectile travels in 1/30 of a second. In my case I can use the rocket body as a length reference. Divide the distance by the time and you have the speed.
A large ball valve should have better flow but it sounds like you are opening it manually. I don't see how you can open it fast enough. From the time my launch initiates until the rocket has separated from the launch tube is approximately 24 msec. You need a pretty quick wrist to fully open the valve in less than say 10 msec.
Festo's BionicKangaroo combines pneumatic and electrical drive technology, plus very precise controls and condition monitoring. Like a real kangaroo, the BionicKangaroo robot harvests the kinetic energy of each takeoff and immediately uses it to power the next jump.
Design News and Digi-Key presents: Creating & Testing Your First RTOS Application Using MQX, a crash course that will look at defining a project, selecting a target processor, blocking code, defining tasks, completing code, and debugging.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.