Registered users: Bing [Bot], Google [Bot], jimmy101, Yahoo [Bot]
Aside from a simple ball valve cannon, I haven't made a pneumatic (nor combustible, for that matter) cannon that requires extensive engineering. Before I start, I have been ravenously scavenging these forums, the spud wiki, and various how to videos on the subject, so please forgive me if I rase any concerns that have been already solved.
My situation is a little unique, as I have acess to a fully equipped machine shop, and are starting from scratch in my valve construction. Many of the tutorials and videos I've seen are geared towards people with limited manufacturing capabilities, and a limited selection of materials, so while they do a good job to explain the concept, I'm still left in the dark about the specifications that make a good working barrel sealing piston. I have worked out a .9375" Piston, sealing against a 0.625" OD Barrel. My Pilot volume (piloted by a blowgunned sprinkler valve), consists two 1/2" NPT elbows and one 1/2" NPT Tee, which I am over-guestimating to be 3 cubic inches just be safe.
The part I am unclear on is the required length of the piston and it's housing, as most tutorial pistons are constructed within a tee, the piston conforming to it's dimensions. Is there not some math I have to do between the piston size/housing volume and the pilot volume for it to open properly, or is it just between the sealing face diameter and piston diameter?
Secondly, because I'm able to achieve machined finishes, an O-ringless piston will work, yes? I have in mind a simple solid piece of aluminum, or UHMW, with the rubber sealing face of course. Would any material bear significantly better results? Also, in my forum searching the general consensus of piston tolerance is "should slide easily"? If that's the case, shall I machine the piston to a press fit, then make small cuts by the thou' till it slides nicely, or is there a numerical tolerance I should strive for?
The last concern I have is regarding the bumper of the piston, and if it makes any difference if the rubber or whatever shock abosribing material I can find is attatched to the piston itself, or at the back of its housing?
If anybody is confused to what I am machining, or why I'm not using a tee, it's simply for the aesthetic, and I have the time. The main machined component I'm referring to is a steel receiver which houses a sliding barrel breach, the valve behind this, and a hole for the chamber pipe to thread into.
I thought this would also be a good time to make sure my GGDT data is alright. My main concern is the vent size. I'm using the sprinkler as a pilot, but it is a 3/8 NPT nipple coming out of my piston housing which connects to the pilot circuit, so do I go by that smaller dimension? Secondly, I am unsure what the dead volume pertains to, as I can't translate "downstream of the valve seal but upstream of the barrel breech", hence why I didn't adjust it. Is it the empty space in the piston chamber between the sealing face and the barrel when the piston slides back?
Where to begin... I'll start at random..
O ring less. Yes in smaller sizes, it works well. High machining tolerances are needed to limit the leakage. My Mouse Musket was ringless with a piston lapped into the cylinder. A small amount of grease sealed it well, so a small EQ hole was required to be drilled through it.
Forces on the piston to hold it closed and then no force holding it closed, to force to open it are all based on nothing than the ratio of the piston face seal diameter and the piston diameter where it seals into the pilot. A piston with the two diameters close to each other will remain closed longer requiring the pilot to vent to a lower pressure before it balances then opens. The lower the pressure in the pilot when the tripping point is reached, the faster it opens when it does trip. They snap open. Pistons with a wide ratio (Large piston OD vs small barrel seal) open earlier with higher pressure in the pilot when it opens. This compression into the pilot can limit the speed the valve opens and even force it back closed as the chamber pressure starts to drop.
Pilot volume and pilot valve size - On a wide ratio piston to compensate for the pilot pressure on valve opening, a large fast way to vent the pilot is required to get decent performance along with a small volume pilot.
On a narrow ratio piston, a smaller pilot valve can be used as the piston remains closed until it snaps open. Narrow ratio pistons are more temperamental as they tend to fail to open due to low pilot pressure required. Piston leakage may prevent low enough pressure and friction may cause stuck pistons. A moderate pilot volume is desirable on these.
Last edited by Technician1002 on Thu May 03, 2012 11:21 am, edited 1 time in total.
Sounds like it will work.
The piston only needs to move back 1/4 of the barrel inner diameter full flow. However in practice, as it bounces back when it hits the rear face of the piston housing, aiming for 1/2 of the barrel ID is a safer compromise.
A tight fit will work only if you get more flow from your pilot valve that can flow past the piston, since you're using a diaphragm valve this should not be a concern. That being said, tighter piston means more efficient use of power, as long as it's not mitigated by friction.
As to materials, a lighter piston accelerates faster and puts less stress on your system.
I'm for the former method though since you have machining facilities, surely a piston with a floating o-ring is not that hard to achieve.
If it's attached to the piston, it's adding weight which contributes to the amound of impact force the bumper has to absorb, not to mention reducing performance.
Providing a diagram will allow a more informed critique
It's the volume between the base of the projectile and in your case, the sealing face of the piston when closed - basically where air gets to expand in between valve opening and hitting the projectile.
Some thoughts on your design:
- since you have a relatively small bore, have you considered a smaller chamber and higher pressure? You would need to pilot with a QEV instead of a sprinkler valve, but these are very cheap on ebay so no biggie. Actually come to think of it, a 0.75" would be perfect as a main valve, though you clearly want to make it yourself
- piston can be made smaller (0.75" for example) to improve efficiency, especially if it's a good fit
- that's a pretty heavy projectile for the calibre, what is it?
My thoughts on piston movement.. to full open.
In a coaxial design, when the piston has moved back 1/4 the orifice diameter, The area of the opening between the piston and valve seat is the same area as the opening in the orifice. Many consider this to be full open as both areas are the same.
I do not consider this fully open. This is why. The orifice is a restriction to flow. A second restriction to flow equal to the orifice restriction (gap between piston and seat) is still a restriction to flow. This should open farther so the only effective restriction is the valve orifice. This is for a valve of a coaxial design where the flow has 360 degree entry into the valve. Opening the valve a minimum 1/3rd larger then the minimum 1/4 diameter will reduce the 2nd restriction to a low value.
On a valve built into a T where the flow all enters from one side instead of 360 degrees of the sides, full flow is at about 1/2 the seat orifice diameter. Opening these 2/3 the orifice diameter or more is recommended.
In my QDV valves, I designed the porting into the valve to have an area double the valve seat area for good flow.
For flow bigger is better.
Now on to piston kinetic energy. The further the piston flies into the pilot under force the higher the kinetic energy of the piston. This can result in some serious damaging impact damage of the piston and pilot. To limit this energy consideration must be given to both the acceleration zone and deceleration zone.
Less distance is better. lower mass is better. Longer deceleration zones are better.
With this in mind, excessive piston travel is bad.
Since a coaxial valve is considered fully open at about 1/4 orifice diameter, this can be considered a good place to begin the piston deceleration. A big squishy bumper can be effective here as the piston will continue to open into bumper for a full valve opening while still providing safe deceleration forces. See my ABS cannon for an example of a long soft bumper. Piston hits the bumper at about 1/3 the total travel.
Who is online
Registered users: Bing [Bot], Google [Bot], jimmy101, Yahoo [Bot]