SpudFiles

saefroch wrote:And that cloud on the jet isn't a "sonic boom". So far as I know, it has little to do with breaking the sound barrier.

Afraid you're wrong. The cloud is condensed water droplets resulting from the shockwave forming around the aircraft. Doesn't happen every time, but under the right weather conditions, it forms as as result of supersonic air flow.

That doesn't necessarily mean that the aircraft itself is supersonic (merely transonic), but it still has an awful lot to do with breaking the sound barrier.

Technician1002 wrote:The chrony the orange is over agrees with you. It's really about 800 FPS.

As I say above, the Prandtl-Glauert singularity occurs as a result of supersonic air flow.

At 800 fps, sphere shapes are transonic - there is supersonic flow over parts of the projectile. So, in theory - although Jack's explanation is more likely - it could be caused as a result of such.

Ragnarok wrote: As I say above, the Prandtl-Glauert singularity occurs as a result of supersonic air flow.

At 800 fps, sphere shapes are transonic - there is supersonic flow over parts of the projectile. So, in theory - although Jack's explanation is more likely - it could be caused as a result of such.

After reviewing the high speed video of this shot, I think you nailed it. Thanks. I didn't think this was the reason for the cloud shape due to the speed clocked by the shooting chony. Watching the orange pass by, it didn't look like the flow behind the orange was keeping up further along the path, but the cloud remained.

I wondered why that cloud had that shape that didn't expand out from behind the orange. The singularity does match what we got on video. I hadn't connected it because it is below supersonic.

Technician1002 wrote:Is there a reason to try for SOS? Is this for a project for something or just a personal quest?

Well, I've heard speed of light is a bit harder to pull off so I figured I'd start here

Seriously though, purely personal quest. I built my first spudgun probably 10 years ago with my brother and our friend. Since then we've probably made 4 or 5, all standard ball valve models, and one with a sprinkler valve. Recently the budget has loosened up, so this weekend I made my first barrel-sealed piston model. We promptly used it to dent a metal garden bed with a balled up grocery bag wrapped in masking tape : D
Now I want to figure out what my next step is. Not really sure where to go from this model.

Too long of a barrel hurts performance. I have a page on trimming the length of a barrel to peak performance for a pressure, chamber size, and projectile mass.

Isn't that only the case when your chamber isn't big enough to sustain pressure for the full length of the barrel? I mean, isn't it not so much "too long of a barrel" and more "incorrect barrel to chamber ratio"?

Callipygous wrote:
Isn't that only the case when your chamber isn't big enough to sustain pressure for the full length of the barrel? I mean, isn't it not so much "too long of a barrel" and more "incorrect barrel to chamber ratio"?

In a nutshell, no. Case in point, take a shop 8 gallon compressor and attach a 1:1 ratio 1/4 inch barrel with a triggered burst disk. It would be a very long barrel. The flow resistance for the length would be very high. The flow out the end would be less than 1/2 the speed of sound even without a projectile. Replace that barrel with one 4 feet long and the flow will increase to nearly SOS and become very noisy.

There is a reason really long hoses is a bad idea on power tools. A larger chamber (compressor tank) won't improve the problems with the really long barrel in the above example.

A guy built an air cannon with a really long barrel and a huge chamber. My much shorter barrel on my much smaller chamber outperforms it by a wide margin. His huge air cannon is here. The large chamber and long barrel did not result in a corresponding performance gain.

A shorter barrel would have performed much better. The oversize chamber could have been trimmed to less than 1/3 the volume with no appreciable loss of performance.
http://www.crazybuilders.com/item.php?i ... ct_section

He published his results. I have compared them to my results. He draws the same conclusions.

For our cannon, the narrowest section of pipe is a major factor that limits the maximum flow rate through the barrel. When the air goes through a narrow section it must speed up to get the same number of air particles through per time as the wider section. When speed of the fluid goes up, the fluid losses go up reducing the apparent pressure on the projectile. As the fluid speed approaches the speed of sound, huge inefficiencies result.

With the limited speed and increases in losses with increases in length, there was too little pressure behind the projectile to provide much acceleration for much of the travel, and high losses, so peak speed was limited.

I don't think I understand what would cause that. The only concept I am familiar with which would cause that is pressure loss due to friction between the air and the inside of the barrel. Are you saying there's a point where you cant really overcome that with a larger chamber, or higher initial pressure? Also, can't you reduce that effect with a wider barrel, and compensate for the larger barrel volume with a larger chamber volume?

It is true the losses are higher with smaller bore barrels. Large bore barrels have less loss. Higher pressure does work with longer barrels. There is a practical barrel length limit for a given pressure, diameter, chamber volume, and projectile mass. Pick up a copy of Gas Gun Design Tool and play with it. You will get warnings that the barrel is choking the flow in some barrel lengths.

Real world testing has confirmed this.

I have ggdt and i have been playing around with it the past day or two. What I've found actually, is that I can have a ridiculously long barrel, like 1500 inches, and it doesn't mind so long as I have a really large chamber as well.

GGDT regularly disobeys the laws of physics. It's tendency to give mach + readings in unlikely circumstances is well known.

With plain air, you will generally need to heat it (either before or via compressive heating on firing) to go supersonic. Don't trust GGDT too much.

Some data you might find interesting here.

Basically with a piston valve I was getting less than ideal results for light projectiles, as opposed to using similar parameters with a burst disk valve.

inonickname wrote:GGDT regularly disobeys the laws of physics. It's tendency to give mach + readings in unlikely circumstances is well known.

I wouldn't say that it REGULARLY disobeys the laws of physics... Just that it doesn't handle transonic/supersonic scenarios well.

Yeah, I know what the problem is, I'm just not 100% sure of how to fix it short of a full rewrite (which will likely happen in the not too distant future...basically once I get Vera up and running).

Well, not regularly for ordinary gun modeling. But for higher pressure guns it can often give rather generous predictions. Around transonics, don't take it's word as gospel truth.

jackssmirkingrevenge wrote:Some data you might find interesting here.

Basically with a piston valve I was getting less than ideal results for light projectiles, as opposed to using similar parameters with a burst disk valve.

I loved the thread and the correct conclusion on piston bounce. When the chamber vents quickly and the pilot is venting relatively slowly, this bounce on the air and bumper can hurt results. For this reason I use 1/2 D instead of 1/4 D so even with some bounce, the harm is less. Going to close ratio valves so the pilot pressure is less is another step to fight this. With less pilot pressure when the valve opens, the less air spring there is. Going to a 1:1 ratio valve was part of this progression for a high performance piston. My current build is to work on some low bounce energy absorbing bumpers to further reduce piston bounce.

My current Marshmallow cannon if permitted to bounce with a 100 PSI shot, I sometimes can retain as much as 10 PSI in the chamber after the shot. The piston bounce effect can be seen in my Using O rings video. After the air discharge at the end of the video, the piston is part way closed and hisses some air after the shot. I'm actively working on eliminating the bounce.

Overall, I like the progress so far with the Marshmallow cannon. It flows very well with little hinderance from those ills. This is evidenced by the speed I do get with light projectiles.

In summary, I'm working on elimination of the pilot area trapped air spring and building a bumper based on Newtons Cradle. The first ball (the piston) impacts a mass that takes all the KE and sends it down the line, but instead of using an elastic collision in the transfer, they will be high loss so the energy is dissipated along the chain. The piston hits, stops and stays.

To see this effect, take a Newton's Cradle toy and place about 10 pieces of paper between each ball. Release the first ball and note the rapid damping. Hold the last ball in your hand. This is how you want a piston to come to a stop.

My cannons have enough space behind the piston to make a 5 or more stage Newton's Cradle bumper assy with each stage containing the same mass as the piston. Energy dissipation is a large part of the design. Details later after testing.

There's a not-so-recent gun under "recent developements" on on spudtech.com, he claims it to shoot spuds over the speed of sound. It's a giant piston pneumatic that operates around 150psi, scroll down and you'll know it when you see it. I'm currently building one just like it only a little bigger with a barrel sealer. I'll see if it breaks the sound barrier... I kinda doubt it, but ggdt seems to think it can happen, the values I entered could be incorrect... what do you think?

Getting a spud to exceed 950 FPS is not easy. Getting surfaces smooth to prevent excessive turbulance and losses is required. The chamber shape is important at those speeds along with the number and location of any bends to the flow. The fastest I have gotten an organic projectile has been a little over 800 FPS. Most piston air cannons launch spuds somewhere under 700 FPS.

kiwi of nitro wrote:There's a not-so-recent gun under "recent developements" on on spudtech.com, he claims it to shoot spuds over the speed of sound.

Claims. Joel never actually provided any chronograph evidence.
His reasoning for the claim was that it was much louder and he thought he'd heard a sonic crack - and let's be honest, that's not a reliable method.