Help with CO2 engine

[skyjive]

So recently I have been branching out a bit from just building pneumatic guns and applying the skills and techniques I learned from spudding to other hobby projects. My latest project is an attempt at constructing a CO2 powered vehicle. The basic idea is that compressed CO2 pushes a piston down a cylinder, and four of these cylinders turn a crankshaft that turns the drive wheels. The tricky part in building the engine was coming up with a system to fire each cylinder (which is essentially just a mini pneumatic gun with a piston instead of a bullet) with exact timing as each piston is returned to the end of the cylinder by the rotation of the crankshaft. Here is a very crude diagram, not at all to scale.



I have finished building the engine, and the valve system is as follows. There is one pressure chamber which feeds all four cylinders, and is itself fed from a CO2 tank. At the base of each cylinder, gas flow is blocked by a fairly standard piston valve inside a Tee. The piston has a return spring to help it close against the cylinder end. There is no pilot valve, but instead the mechanical force of the returning piston is used to push on the piston face and force it open:



As the piston returns down the cylinder, the red rod, which is attached to the piston head but not the valve piston, pushes the valve piston back against the spring and greater pressure on the rear of it, allowing air to flow into the cylinder and drive the piston head forward again.
The problem I have is that I’m not sure how far back to have the piston head go. This is because the valve system presents the difficulty of cutting off the flow of gas into the cylinder as soon as the piston moves forward a little bit:



Then ->



So basically only the air that makes it into the cylinder before the piston moves enough to close the valve will be available to propel the piston the rest of the way down the cylinder. My dilemma is that if I have the piston return only say halfway down the cylinder, leaving plenty of available space to fill with air before it moves, the piston stroke is shortened, whereas if I have it return all the way, there is no room for air to enter and there will not be enough air to drive the piston properly once the valve closes.



So generally I’m just asking for thoughts/advice on how to design the system. The engine, valves, and cylinders is already built, the crankshaft and pistons is not. The dimensions of the engine are (all Sch40 PVC)
Valve Tee: ¾”
Piston Seat: ½”
Cylinders: 1.25” (cylinder diameter must be greater than seat diameter to ensure that the stroke of one cylinder provides enough force to open the piston valve in the next cylinder, with the force being transmitted between pistons through the crankshaft)
Cylinder length: 11”
Chamber volume: 26.6 in^3 or 59” of ¾” pipe (somewhat irrelevant since nowhere near all the CO2 is vented into the cylinder on each firing, but here it is anyway)
Thanks in advance!

[Dave_424]

Personally I would have a blow through hammer valve for each cylinder.

when the cylinder reaches TDC (top dead centre) it pushes a tube upwards. This tube dis-lodges the seal which lets CO2 move through holes in the tube, through the tube and onto the face of the cylinder thus forcing it downwards

If you don't know about a blow through hammer valve, check the valve that is used in the airforce condor

http://i2.guns.ru/forums/icons/forum_pi ... 577492.gif

Dave

[jackssmirkingrevenge]

Seen this?

[jrrdw]

Good find Jack, but the check valve ball is going to get dimpled and cause leakage into the cylinder and make the engine slow or stop.

Skyjive, you need a exhaust and I would leave the connecting rod the original length because once the casing is charged the first time it will not have to exhaust the entire amount of CO2, only what is displaced in the stroke.

You also need a flywheel, weighted and balanced to start the engine and to help the pistons through the entire stroke.

[psycix]

Dont get the piston to top out halfway. This would be a waste of air.
The volume you want to have as dead space would be something like:
V1 = (V2*P2)/P1
The starting volume = The expanded volume times the pressure you want to vent off divided by the starting pressure

Note that this calculation does not keep adiabatic expansion in mind: the cooling of the gas.
Oh, and the pressures are absolute pressures, in bar.


Nice project. With a good flywheel you wont need 4 cylinders, saving lots of work.

[jackssmirkingrevenge]

Good find Jack, but the check valve ball is going to get dimpled and cause leakage into the cylinder and make the engine slow or stop.

This is how most CO₂ motors actually work, if the ball material is harder than the pin pushing it it should not be a problem - and if it is, you could easily replace it with say a schrader valve.

[tomthebomb137]

Im surprised im the first to bring it up, but although its deffinatly not a dead end, as any paintball player, or spudder who has used a Co2 system knows, CO2 cools as it expands, like every other gas, except CO2 does it to a much grgeater extreme. For this reason, HPA is used instead of CO2 on speedball (fast) paintball guns. With the amount of expanding CO2 such a device would require, you will encounter a large amount of freezing problems in all likley hood, as oil will solidify at -15F (for 10W30) and -30 (for 15W50) as will almost all seals that are constantly at work.

[jackssmirkingrevenge]

With the amount of expanding CO2 such a device would require, you will encounter a large amount of freezing problems in all likley hood, as oil will solidify at -15F (for 10W30) and -30 (for 15W50) as will almost all seals that are constantly at work.

Apparently there are ways around this problem

[youtube][/youtube]

[tomthebomb137]

hmmm i stand corrected. now we just need to figure out how


now go to bed jack, your making me look bad

[tomthebomb137]

but hey, i didnt say it was impossible!

[jackssmirkingrevenge]

now go to bed jack, your making me look bad

You mean "go back to work", Central European Time here

[inonickname]

Compressed engines are nearly all a 'pin and ball' configuration.

It's usually the pin that suffers the most. It will end up mushrooming slightly.

Rotary valves are also and option, and work much better (are simpler to implement) on larger (multi cylinder) engines, plus if well made and lubricated there's no parts to replace.

[jimmy101]

Probably to late but what about just using an air tool? A grinder if you want rotation or a nibbler if you want recipricating action.

Cheap ones are, well, cheap. Proabbly get a air driven grinder for less than $20.

[POLAND_SPUD]

yeah an air tool would be the best...

well I know it's off topic but I met a guy who wants to use air tools to generate electricity from wind power...
since batteries cost a lot and one would need an inverter to run AC appliances and good low RPM generators cost a lot he wants to power a small compressor with a wind turbine and use a big tank to store air... air tools have a quite high RPM so they'd be ideal to power an AC generator (becasue high RPM generators are really cheap)

but to tell you the truth I am not sure if it would be efficient...
first of all wind power is converted into rotation of the turbine... which is then converted into air pressure... then pressurised air is stored and finally its energy is converted into rotation that is in turn converted into electricity

so what's your opinion?

[jackssmirkingrevenge]

Since you'll be losing energy at every stage (air compression, storage and turning the air tools) it sounds like the setup will be extremely inefficient compared to just using a wind turbine to generate electricity directly.