Idea for homemade HPA compressor

[spudgunr]

Okay, to start, this is the beginning of a thought. I doubt I'll get in to PCPs any time soon, but as I thought about it the big obstacle was the air source; I dont really want to pay for fills, and I like to build things. We'll ignore the economics given how little I would likely shoot it (most likely I'd use a hand pump for a 22cal air gun).

Anyway, the attached picture is the idea. The principle of operation:

• Air compressor makes this more time efficient and retracts the cylinder
• hydraulic pressure acts on the full bore side of the cylinder. This allows the system to devlop higher pressure on the air side than what is input to the hydraulic side (for instance, my tractor is rated at 2500psi)
• While cylinder moves the air in the cylinder gets compressed until it exceeds the pressure in the tank
• Compresor side check valve prevents high pressure from going back to the air compressor
• HPA side check valve keeps PCP tank from filling the hydraulic cylinder, but allows HPA in to the PCP tank when the pressure is high enough
• At the end of the stroke the hydraulic pressure is released. The air compressor pressure pushes the rod back into the retract positio

Here is the unknown: Is 125psi or so from an air compressor sufficient to overcome the friction of the cylinder?

Will normal hydraulic systems, specifically on my tractor if I ever make this, allow the flow in and out from the same connection? This may be the biggest sticking point.

Plusses: hydraulic systems are plentiful and already rated for 3000-4000 psi. Log splitter cylinders are common and many go up to 4500psi. Should be able to source a used cylinder for under $200

Math example; 4" cylinder with 2" rod in a 2500psi hydraulic system:

Full bore is 12.56 square inches surface area
Rod side 9.42 square inches

2500psi on the full bore side would result in 3333 psi on the rod side

A 24" stroke cylinder would have 226 cubic inches volume on the rod side and would go from 125psi to start to 4000 psi* with 3.125% of the starting volume, so end volume would be 7 cubic inches (~115 cc). It'd only take a couple strokes to fill most PCP tanks right? Oh, and there would be 301 cubic inches on the full bore side of the cylinder, that is 1.3 gallons, my tractor hydraulics are rated 5gpm so stroke cycle time would be 16 seconds.

I'd love any thoughts on this and even better if somebody has feedback on the hydraulic side of things.

The larger the rod diameter compared to the bore diameter of the cylinder the higher pressure differential you can achieve. So in the above example if you wanted 4000psi from a 2500psi hydraulic system you might need a 2.25" or 2.5" rod on a 4" cylinder[/size]

*4000 psi chosen because it was easy math; it represents 5 halvings in volume / doublings in psi

[Blacktalon6]

I would think just air in a hydraulic cylinder would melt the seals real quick.

[spudgunr]

This isnt something I had considered, but there woudl still be hydraulic oil on one side. Still, with air on the extend side, is that enough to do the cylinder in? Great question, and enough doubt to make this something worth doing with a well used cylinder (I've seen some compact ones for SUPER cheap, so I'd probably do a proof of concept with one of those based on this concern). That oil though is of course only on the piston there are the end seals aroudn the rod that would never see hydraulic oil, so those would be the biggest risk.

Anyone knowledgeable enough on this to know for sure?

[Back_Roads]

 There are booster pumps that will take shop compressor air to run the booster pump. IIR there are some DIY that use air cylinders from truck air brakes or the like also.

[Back_Roads]

 And years of GTA adventures https://www.gatewaytoairguns.org/GTA/index.php?topic=61263.0

[sb327]

In theory this should work. And if everything seals properly, I think it will.

The heat is probably a non issue due to the thermal mass of the cylinder is large and the compression cycle is slow. And each time, the hydraulic fluid cools the cylinder as the piston is extended. Also, hydraulic cylinder seals tend to be rated at a decently high temperature due to the high temperatures hydraulic fluid tends to get.

Try it and see, it may be possible to use foster fill valves as your check valve’s provided the 125psi will open it with what residual pressure would have when you hit retract.

On that note, hydraulic hoses on tractors act as pressure OR return depending on the position of the control lever. Most on tractors are Closed Center, meaning no fluid can flow in the neutral position. An Open Center would be handy in this case but not necessary.

Good luck on the endeavor!

Dave

[Gippeto]

Something I would think about...just tossing it out there, feel free to ignore. 

How often do you actually stall your tractors hydraulics with the engine throttled up? That's the work load you would be asking them to do in this case...actually develop max pressure....which, ratings aside... they VERY rarely have to do in the real world.

Might be worth a moment or two of consideration when figuring cylinder sizing at any rate. A higher turn down ratio might take longer to fill, but will be that much easier on what could quickly become some fairly expensive repair bills.

Al

[Doug Wall]

I'm not an expert on hydraulics, but two things to consider:
1) Hydraulic cylinders need hydraulic fluid in both sides to push back and forth. I suppose that you could use some big springs to retract the cylinder.
2) Hydraulic fluid is pretty much incompressible. Therefore, the cylinders aren't designed to minimize dead volume (like piston end and piping). This will absolute destroy your ability to compress air, which is compressible. In your example of a 24" cylinder with 226 cu in, if there was one cu in. of dead space in the cylinder, you would never be able to put out any air above 226 bar and at any pressures below that, your output woud be way reduced.

[Snatchcan]

I've often thought about doing the same thing.  My first concern has always been turning the cylinder into a very simple, and very large combustion chamber,  if any oil were to seep through, because eventually it will.

I'm a technician on compact equipment, tractors, and diesels. Every brand I've worked that use open center valves have an acceptable rate of drift for their cylinders/hydraulic systems. Most of them are over an inch every 10 minutes measured on the rod. Now this doesn't mean only the cylinders specifically are bleeding by, the spools in the valve along with many other valves and other components are just as likely to have a small amount of seepage.

My solutions are as follows.
Use two cylinders. One run off the machine, the other being only for compessing air, being driven by the first cylinder. This will allow you to extend and retract the air compression cylinder without a weight or a spring.

The second cylinder would have pure silicone oil on one side for seal lubrication, probably the rod end, connected to a reservoir with a sight-glass, so you will know when you're losing oil. The other side of the cylinder would be for compressed air. I would also machine at least one extra groove for an extra piston seal, to help eliminate any seepage.

If you use the base end of the cylinder for the compression chamber, you're going to need to machine a piston that screws directly onto the rod and leaves very little to no dead space between the base end and the piston. Most pistons slide over the rod and are attached with nut, which would lower your fill pressure significantly.

If you use the rod end for your compression chamber you're going to lose volume and pressure, however most pistons extend until they reach the head and already leave very little dead space. So this would mean less work but you may need a larger cylinder, to get to your desired fill pressure. This will also increase the cost. 

Another idea I've floated around, is just a single cylinder set up the same way as previously stated, one end air the other silicone oil and a reservoir. But you attatch this one mechanically to the machine, between the frame and the loader and use the loader to run the cylinder up and down. You could even get an identical cylinder for your loader and attach it with extra long pins right next to your hydraulic cylinder.
 

[oldpro]

Fast forward to 1:40 and you will see probably the best idea for a pcp compressor yet to be used.

[AlanMcD]

That is an interesting idea - of course in the AC compressor all six of the pistons driven by the plate are equivalent, resulting in a single stage compressor with six pistons (single stage is plenty for an AC system).  If that arrangement could be redesigned so that each piston fed the next one in the series, it could achieve very higher pressures, with reduced work in each cylinder.  It would still require a good internal heat management system to manage the excess heat from compression (which would be much higher than in the AC configuration), as well as a way to trap and periodically release condensation between the stages to prevent damage (the AC compressor does not face this issue), but it could be made to work quite well.

The challenge is managing all that condensed water - imagine having 5 or 6 vent points - which would be required since our air compressors don't work in a closed loop. The highest volume of water will condense out in the first few stages.  Honestly, that is one of the things that I felt was brilliant about the Shoebox compressor - it used a common shop compressor as the first of three stages, and thus took 90% of the condensation work off the table and made the remaining amount a non-issue, at least for the compressor (it still matters to us, so most use a drier before the Shoebox to eliminate the issue).

I do like the idea - although it is probably pretty expensive at low volumes.

[AlanMcD]

OK - now that I have had a cup of Joe my brain has awakened, and it told me why this might not work:  Washplate driven pistons work well because the load is balanced by the fact that all pistons are producing the same work, and thus the same load - which clearly is not the case if all pistons have the same swept volume.  One could vary the volume and compression level in each to lower the peak loads on the washplate, resulting in a multistage compressor that would be more balanced in load (but nowhere near as balanced as in the AC compressor).

In the end, the system would function much like a radial dive compressor, except the pistons would all be longitudinally aligned and driven by the washplate rather than the central crankshaft.  I think the only advantage this would have would be the potential for reduced package size, but that generally is not an issue for a stationary compressor, unless one is targeting a smaller portable unit and is willing to pay the added costs for any given output level.

Still interesting though - it might be able to make a nice small lower speed compressor . . .

[Back_Roads]

 No batteries run it off your cars serpentine belt 

[MJP]

Fast forward to 1:40 and you will see probably the best idea for a pcp compressor yet to be used.

Do a search on Rix sweet air, they have used the swashplate construction on hpa for decades.

Marko