Piston Design

[Technician1002]

A simple lap job on the piston and valve seat take care of low pressure performance. My Mouse Musket sealed at only 2 PSI.
One stroke on the hand pump sealed it.

[ramses]

I am personally against venting the pilot area. It removes the need for a bumper.

[long-winded explanation]
If you make the piston travel, say, 2x what you need for full flow, and you start at 11x , there is 10 bar in the chamber, and 10 bar in the pilot.

Once the piston moves back half way, the valve is fully open, the ratio of pilot area to chamber area is 1:1 (since the piston is away from the barrel seat). Chamber-side pressure is perhaps 50 bar. Pilot pressure is now 20 bar, net force on piston is 30*14.7*SA. The piston is still accelerating, after it is fully open!

now it moves to 3/4 open. Since the compression ratio of piston travel/pilot area is now 4, the pilot is at 40 bar. NOW it might stop accelerating.

at 7/8 of the full travel, compression ratio is now 8, pressure is 80 bar...

And so on.

Basically pressure builds in the pilot to ridiculously high levels, slowing the piston down. assuming no dead volume in the pilot (thinking of it as a pump, now), pressure in the pilot would need to be infinity for the piston to hit the back of the pilot.

[/]

As a side-note, the O-rings on my piston hybrid slide in a fairly rough tube, and hold up quite well.

When I make my upgraded valve (1.5" porting) I'm not sure I will keep the O-rings. It would be very convenient to fill from the pilot, but I am concerned the flame-front will propagate through the gap around the piston. Even if that didn't happen, I would probably loose the air-bumper effect I described above, and - if the mix was right- it would diesel a ~80x mix behind my piston.

Any input on that?

[Technician1002]

Add fuel first. Purge the pilot of fuel by adding air through the pilot. Flame propagation into a fuel devoid area should be no problem.

Leakage into the area simply raises the pressure in the pilot sooner so the piston stops acceleration sooner. It then has more air to cushion the movement.. Oh and on it leaking back out, the flow is high loss. It acts as a dash pot and less like a spring for lower energy rebound. A tightly sealed pilot is not an advantage. Hard piston rebound can cause piston face damage.

[ramses]

excellent - less machining to do!

My sealing face is fine, but I usually pump the pilot to far below optimum pressure.

[Moonbogg]

So, even with a leaky pilot area, the idea is that the chamber pressure will rise too fast for the pressure to have a chance to equalize in the pilot? Not saying I don't believe you, but it just sounds scarry to have any leaks in the pilot where any fuel can seep back in or where pressure can find its way in to equalize the pilot with the chamber.

[Technician1002]

If you fill the air through the pilot, even with some diffusion, the pilot should be too lean of a mix to do anything. It should fire well before it sat long enough to diffuse fuel into the pilot. Air under high pressure is dense and diffuses slowly. Pilot pressure will rise exponentially with the piston movement. What is important is to have very little dead space. This means the pilot fill ports must be very small so little flows back out them in the short duration of the shot. A reed valve on the air/fuel inlet in the pilot area is recommended.

[ramses]

Caulk is also good for filling dead volume. I closed my pilot area ball valve, filled the nipple with caulk, and inserted a thin wire to allow air to flow. I know it compresses somewhat, but since my gun isn't broken, I don't think I'm compressing it too much.

I'm not sure a reed valve would take the pressure. A small check valve would do the trick.

[c11man]

when you inject the fuel that wont be enough pressure to seat the piston will it? would a premixed fuel air be better for this aplication? but if you do that then you need to oring the piston to prevent the pilot from igniting.

or you could use a light spring behind the piston to seat it, then inject fuel...
and if you do that you could inject fuel into the chamber section and air through the pilot and have absolutly no fuel in the pilot section

[ramses]

At 10x, there is 6.2 psi of propane in the gun once it is filled. If you fill it from a meter pipe, like most people, you will be pumping perhaps 80psi propane into the pilot, where it will later pass into the chamber. That pressure seems to work fine for pneumatics, so I don't see an issue. You might have to compensate for the minute amount of propane lost out the barrel before it seals, but I doubt it. A weak spring could be used, but since you minimized dead volume in the pilot to create the air bumper, it will likely pancake, unless your piston travel is huge.

If you filled the chamber with propane (assuming your piston sealed somehow) and then injected air, you would push a stoichiometric mixture of air and fuel into your pilot BOOM.

If you used o-rings on the piston without a check valve to equalize, as I do on my hybrid, your fill procedure would be as follows:

• 1. Fill pilot with air
  2. Fill chamber with propane
  3. Fill chamber with air
  4. BOOM
  5. Repeat 2-4 until finished

If you had a check valve in the piston, you would:

• 1. fill pilot with propane
  2. fill pilot with air
  3. BOOM

Now that I think about it, if you didn't vent the pilot with a pop-off, the residual pressure would re-seal the barrel, and bleed to equalize in the chamber.

Does anyone think the vacuum from the chamber cooling would equalize to form a vacuum in the pilot, and pull the valve open from the pilot I know vacuum doesn't suck, air pressure pushes(in this case from the barrel port), but... then the chamber would suck in gas from the barrel, which could be assumed to be mostly combustion byproducts, and screw up the mix for the next shot?

[Technician1002]

Basically pressure builds in the pilot to ridiculously high levels, slowing the piston down. assuming no dead volume in the pilot (thinking of it as a pump, now), pressure in the pilot would need to be infinity for the piston to hit the back of the pilot.

Any input on that?

For evidence of the compressed air, look no further than my empty pop can impact.

The pattern on the bottom of the can shows where it is patterned from the steel plate it hit. This can hit bottom first. The ring of the bottom is top left in the photo.



The center of the bottom of the can should have pancaked flat against the sheet of steel. After finding the center of the bottom, it appears that most of it it never touched the steel. It did get peppered with moss and stuff that was on the steel.



The best is saved for last. The top of the can did not implode inside the can and contact the bottom of the can. The air inside the can stopped the top before it reached the steel sheet. It is relatively undamaged. It is bent a little, but there is no sign it impacted the steel.