Piston Valve Design

[FordGtMan]

I know how a piston works, but what are you using to equalize pressures between the pilot and the chamber? A small enough hole would allow equalization and the piston to move forward wouldn't it?

[Hotwired]

A greater pressure but a smaller volume is an alternative but for practical reasons it's much easier to have everything at the same pressure.

What your example should actually read is:

38.48 sqin x 100 psi = 3848 lb

3848 lb / 3.14 sqin = 1222.93 psi


So you need twelve times the pressure to keep that valve shut against a chamber pressure of 100psi.


Personally I've not bothered going beyond a plain (plain-ish anyway) cylinder shape for a chamber piston. You can't get away from having to have a pilot volume equal to the displacement of the piston moving back without some whacky pressure differences.

[DonTheLegend]

Wow stupid math on my part. Thanks for the help. Back to the drawing board, be back tomorrow.

Edit: Noticed more stupid math: Area should be 9.6 sqin and .785 for the piston face and pilot volume, respectively. End result doesn't change much.

[clide]

Overlooked that point. In theory this would still work if the pressure on the back was much greater to overcome the difference in areas correct? This would follow the formula Force = pressure x area so assuming a 1" pilot ID and a 3.5" sealing face on the piston

F(sealing face) = 38.48 sqin x 100 psi = 384.8 lb

384.8 lb / 3.14 sqin = 122.293 psi required to over come the force on the sealing face and hold the piston closed.


In theory would this allow the valve to open faster due to the reduced pilot volume? (the previous cannon used a split pressure system so the parts are already in place to continue to use it on the new version).

You have the right idea. And kudos for actually doing the math, but your doing your area calculations wrong. You are using pi*d^2 it should be pi*r^2 or pi/4*d^2

However since you are doing both of them wrong and dividing them by each other the lack of the /4 term would cancel itself out, but you made another mistake:

38.48 sqin x 100 psi = 3848 lb not 384.8

So you would need over 1200 psi in the back of that thing.

Another problem with the design is that it would have a strong tendency to oscillate when trying to open. Once the valve opens a little bit and begins to pressurize the barrel, the chamber pressure will only effectively be acting on the same surface area as your 1" pilot. The problem is that unless you have a monster pilot exhaust valve, you'll still have more pressure in the pilot than in the chamber so it will want to close again.

If you haven't already I would suggest looking over the "Piston valves explained visually" sticky.

Edit: Oops, didn't see page 2

[DonTheLegend]

I had looked at that sticky earlier, but didn't take the time to fully understand it. Those animations are excellent by the way, I went back and looked at them again and it makes much more sense now. We came up with a slightly different design, but I don't have enough time to draw it up yet, hopefully get to it sometime tonight or tomorrow. Ill post it whenever it gets done.

I don't know where my head was at when I did those initial calculations every single number was wrong

With that tendency to oscillate, has anyone thought of using that to make a full auto valve of some sort? It would have a wicked ROF, the problem would be regulating it....just a thought.

[Hotwired]

ROF isn't necessarily a good thing

If you have a snuffle through some of the auto threads you can see a few experimental valve systems which cycle extremely fast, way too fast for feeding ammo.

[dongfang]

Hi,

I´m not sure if someone already pointed it out: What you need is a longer piston, so that the pilot chamber (on the right in your diagram) is separated from the exit part (up in your diagram) when the valve is closed.

There will be pressure in the pilot chamber, but not in the barrel. You need a good piston seal (O-ring).

With a barrel sealer, you basically switch the barrel with the chamber. You will not need a really good seal with that.

Regards
Soren

[DonTheLegend]

The pilot chamber is on the inside of the piston...It's a hollow tube, as is the tube coming out of the backplate and they seal together with an O-ring. This isolates the pilot chamber from leaking anywhere. As noted in other posts, however, the current design won't work due to the difference in surface areas between the pilot side and the piston face.

[DonTheLegend]

Ok modified valve design:

Basically a partially hollow piston to allow the necessary surface area to hold it closed. Its kind of hard to tell but the area that is a light blue would be filled with air.

[LynyrdSkynyrd]

he means within the light blue there will be air.

[Carlman]

he means within the light blue there will be air.

why did you need to add that when he already said it in his last post?

[JDP12]

yea thanks lynyrd, but kind of already figured that out by what the original author said..

[Gippeto]

Your new "hollow" piston does not change the effective area.

You have complicated your design without solving the original problem.

[potatoflinger]

It looks to me like this thing won't work at all, what you have on it now is just a piston that seals against the barrel, with the chamber pressure holding it there.

With the design you have now, when you release your pilot area, nothing will happen because there has to be pressure on the barrel side of the piston so that when you release the air from the pilot, there is force on the other side to push the piston back.

[LynyrdSkynyrd]

its not a barrel seal. The piston head seals against the chamber, so the pressure difference from the chamber will cause the valve to open.