View Full Version : mini air compressor

10-15-2010, 12:59 AM
Well, the adiabatic air compression scheme is out the window now, but I still want to build myself a small air compressor for the toolbox. I am thinking about a radial piston design, pancake type. I'm also considering a single cylinder long stroke type, possibly double acting. Just looking for more ideas- any and all thoughts welcome.

The space I have to fit this into is about 22 inches in height, 8 inches wide, and 2 inches deep. That pretty much dictates a pancake design, which suits a radial just fine. The space also suits a long cylinder design, provided I can find a way to run a piston up and down the cylinder in a stroke about 5 times the piston diameter without the connecting rod interfering with the cylinder. One thing I don't want is a tiny piston scorching up and down a short cylinder at the speed of light, making more heat than anything else except noise.

My intended use is for headless pin guns and brad guns. No air drills, or other high CFM usage tools. I fully expect to get only a few shots before the motor has to come on, since the tank is very small also. If the thing has to run constantly to give me one shot every five seconds, that will be fine. The fully charged tank will give me about four shots before needing the pump, and that's within my working envelope.

I'm probably going to use a 3-400 watt range universal motor to keep the weight down, or maybe a PM motor of suitable size. Small treadmill motor, perhaps. No heavy induction motor- I ain't lugging one of those up and down stairs. The whole idea of this workbox is that I have all my required tools in it, and I don't have to separately tote a portable air compressor. The box is built- I can stand on it, my head just short of the ceiling, and I don't have to bend down to the floor to reach the most used tools and fasteners. If I can keep the total loaded weight to about 45 lbs, that will work. That's 30 lbs for hand tools, a jigsaw, two cordless drills, pin gun and headless pin gun, plus assorted fasteners- and the air system, air hose, extension cord- I'm looking at about 12 lbs or so max for the complete air system. The tank weighs about one lb, the motor about 3 lbs- no suggestions that include cast iron please-

10-15-2010, 10:56 AM
Do what I did and use a CO2 tank from a paint pistol and get a regulator for the 3000 psi max pressure. I used this rig to go around for punch list items when I was a finish carpenter. It was so nice not hearing a compressor I got several extra tanks and used this system exclusively for all trim nailing they last a long time even shooting 15ga finish nails.

I just did a search and these rigs are available commericially now. Lowes has one and Mcfeelys has one . Good luck

10-15-2010, 11:45 AM
Darryl, years ago, Sears Craftsman sold a 3/4 hp compressor. It was a carbon piston in a carbon cylinder sleeve. My grandson presently has one, but the sleeve is worn and pressure is down. My point is, this unit is only the diameter of a 3/4 hp cap. start, 3450 rpm motor, which is about 8" and about 16" long. It should easily drive a pin gun with a small tank and regulator.
Alternatively, have you considered "reworking" one of those little 12 volt tire inflators from CTC. With a universal motor, it might serve you very well.

Deja Vu
10-15-2010, 12:15 PM
Hmmm... I'd just go up and down the stairs twice with two or more packages. :cool:
..but I like the spirit of your as yet dream.;)

10-15-2010, 09:03 PM
The 12v tire inflator type of thing is what I'd like to avoid. I'd be building this device myself, and considering it a home shop project, so I don't mind spending the time to come up with something reasonable.

I like the idea of a much higher pressure bottle with a regulator, though it does require attention to filling, and might also represent a safety problem on the jobsite. I think I'd rather stick to the built-in pump-

Thanks for the suggestions.

10-15-2010, 09:08 PM
This comes to mind- I wonder if it's possible to convert an air tool into a pump- say a die grinder or drill- anybody tried driving one with a motor to see if it will work backwards?

10-15-2010, 09:15 PM
A little out of the ordinary, but have you considered a 2 cylinder (opposed) compressor layout with a scotch yoke to turn the rotary into linear - no scraping the cylinder sides and the pressure fluctuations would not be as great with 2 cylinders 180 degrees apart. If you had a pancake motor you might even be able to fit that into a 2" height (say 1" bore pistons).


J. Randall
10-15-2010, 11:18 PM
Have you considered using an old weed whacker, or chainsaw and building from there? It would be a fairly small package.

Edit. just reread you original post, and don't think my idea would fit.

10-15-2010, 11:40 PM
This may be a little bigger than what you were thinking of...

Have you considered gutting a cheap pancake compressor and replacing the compressor motor with one from a refrigerator? I've been toying with this idea for a while and I am convinced that it would make a compressor that was small, light weight, and whisper quiet. The obvious drawback is that the recovery time would be longer than normal and maintaining the proper oil level would not be easy without installing some kind of sight gauge.

Just a thought.

10-16-2010, 12:37 AM
Friend at work said he used to use a refrigerator compressor for his portable air system. One of the smallest ones, like from a 2 cu ft or smaller. I'm kind of leaning away from that because of the smell they make, or used to make anyway. Still that could be a viable way to go.

I do like the scotch yoke idea, that had eluded me. Two pistons opposed on the same shaft would lend well to that configuration. That is a likely candidate for my project, thank you.

I've had another idea as well, using fluid as a piston. There would be two cylinders vertically, side by side, with a fixed volume of oil in them. A fluid pump would work through a reversing valve to alternately transfer the fluid from one cylinder to the other, compressing the air above the fluid column each time the transfer takes place. A sensing device of some kind would activate the reversing valve. Water would collect in the fluid of course, and would have to be drained as required.

Black Forest
10-16-2010, 02:32 AM
I had a sandblasting and painting company. We used huge compressors for the sandblasting. Up to 1600 cubic feet per minute. The older compressors were rotary vane compressors. The new ones were screw compressors. The rotary vane compressors were easily repaired. The screw compressors were much more difficult and expensive to repair. The screw compressors were much quieter.

A die grinder uses a rotary vane compressor as does a impact wrench. I don't know if you could use one in reverse to compress air but I think you could. The volume of air produced would have to be calculated. The oil separator would be the biggest issue I would suppose.

Forrest Addy
10-16-2010, 04:02 AM
OK. Winging it here. Letting loose the crows of creativity. Compressor. Small. Quiet. Compact.

Opposed free piston electro-magnetically driven.

The pistons are linearly coded for axial position.

A big spring between the pistons recirculates kenetic energy.

While the pistons are mass balanced they are two stage and the magnetic drive keeps the pistons on the head end of their stroke (high in any case) making it possible to modulate delivery without sacrificing efficiency.

Reed valves: hard to beat them for weight and durability. Keep the moving mass down and they can operate at dazzlng speed.

I picture a compresser the size of a quart thermos running about 500Hz with 2 1/2" pistons putting out 3 CFM with 1/2" stroke @ 500 Hz.

The pistons are rare earth magnets and the spring tensioned to provide about 1/2 the compresson force relieving some of the work load of the mag driver.

The only detial I can't decide is whether to compress on both side of the pistons or the outboard sides.

Drive electronics? I haven't a clue. Ingenious would be good.

OK I thunk it up; now you guys build it.

10-16-2010, 03:36 PM
Forrest, I like that idea as well. It would be about as vibration-free as could be. I would think that

There was an article written some years ago about solenoid operated valves for an engine. There was a particular type of solenoid that gave a very high 'push' with a fair degree of efficiency, but the stroke would be seriously limited. I'll have to look into that. Maybe a 3 ph drive could be arranged somehow- being an electronical type person I should be able to figure that out. It could be arranged where both pistons are driven towards center, and a single pressure sensor feeds an electronic circuit to control the commutation and drive reversal.

10-16-2010, 03:46 PM
Forrest:- do the crows of creativity live in the same zoo as the dogs of war? (The devil MADE me ask!)

10-16-2010, 05:24 PM
Seems like the crows of creativity have pecked at my last post. Some of the text I had typed is missing in action- or maybe the dogs of war have eaten it

Now I have another question about sealing rings. I have a couple types, one has little side support area, but the other has a substantial 'square' edge to it, with a slight outward lean at the actual sealing lip. Can that square edge be used to offer side support? In other words, do I need to have a substantial length to the piston, or can I get away with basically a disc machined up to fit the seal onto? If I go with a two piston design where both are connected by a shaft, then both are automatically kept square in their bores. There will be some side pressure if I use the yoke design, though I don't know how much. Maybe I should play it safe and use a substantial length for the pistons and just let the rings float.

I have some nicely polished tubing from a shock absorber that I'll use for the cylinders, and I'll insert those into machined aluminum heads. The heads will be joined to a front and back plate, both aluminum, which entire assembly will act to dissipate heat. I'll be using reed valves of some kind, and the plumbing will be pretty straightforward, with a water trap at the bottom, etc.

10-16-2010, 06:40 PM
One of the reasons I suggested the scotch yoke was that I saw it used for an experimental engine some years ago. The designers were able to make the piston skirts shorter than normal because the joining shaft kept the pistons aligned concentrically with the bore. In a normal cranked engine they need to be longer to resist the twisting they may want to do around the gudgeon pin. There's a trade-off that was mentioned (for the engine) and it was that the scotch yoke had increased friction compared to a crankshaft, but they were able to offset that against the decreased piston shirt friction and so came out slightly ahead.
The yoke will (should) eliminate any side forces present, so a thick disc may do the job. If you go the electromagnetic route you will need some length to your pistons to prevent twist.
An interesting thought (as you have the lenght in your envelope) is that one thing you may be able to do is rather than use a motor with a crank to drive the yoke, you may even be able to mount the motor 'in' the crank - have a internal stator, external permanent magnet rotor with an excentric mounted on that and go from there. Depends on the stroke you're after. With high rpm you may be able to get the desired pressure and flow with only a short stroke.

10-16-2010, 10:07 PM
I have also considered a direct drive motor. It would have to be a pancake design and be very torquey at fairly low rpms. It would eliminate having a pair of toothed cogs and a belt. There is nothing available in my price range that suits this, so it would also have to be a homemade part. That's not a problem either, except everything I do from scratch will take time, and I'd like to have this thing on the road within a month or so. The model airplane guys build their own motors from scratch, using scrounged motor laminations and neodymium magnets, and it's not rocket science. I'd be looking for sufficient torque at maybe 500 rpm or so, and that's not the easiest, smallest, lightest thing to do- could be a non-starter. I'm no stranger to electronic commutation, so that part at least wouldn't be a stumbling block. It would in fact result in a much more efficient motor, so I probably should look into this idea a little more.

10-17-2010, 01:06 AM
I agree with EVO803 on the CO2 tank. I have taken it a step further and scrounged an empty co2 tank for a soda fountain. Traded it for a full tank [B]with a dip tube[B] and refill my own bottles as needed. Cost for the large tank refill is about $15.00 at the welding shop and it will fill the bottles that Lowes sells many times. I keep one full in the truck at all times to fill a tire. Also great to pop a blow gun on to help with clean up or to blow the lint out of a computer.
Once you have the set up you won't be without it. By the way, the paintball tanks you can get a walmart are the same thing but less $$$$.

10-17-2010, 08:45 AM
[quote=darryl]I have also considered a direct drive motor.[quote]

Since you mentioned a treadmill motor, how 'bout that and a rotary AC compressor from a car? But it might be heavier than the 12 lbs. I'd say or the size/weight you're thinking of, I'd say a CO2 setup would probably be best. I know a couple of folk in my Jeep clubs that have a PowerTank, and can run air tools for a looooong time on it before it needs a refill.


10-17-2010, 05:24 PM
Any moment now I'm off to check out the CO2 tanks, refilling, etc. In the meantime, I have the details worked out for the two yoked piston design. I have a 1 inch ID cylinder tube to use, figuring that it wouldn't be hard to find a lip seal for that size- wrong! Not in my town apparently. There are lots of shaft seals with the inner lip, but no outer lip seals. I've been up against this a few times now. Last time I simply turned a disc from high density poly sheet to act as the piston, then dug a sharp cutting tool into the face very near the rim. This not only takes out a wedge shaped groove, it also tends to swell the outer rim outwards. On the HDPE disc, if it's turned to a nice fit as a piston, the lip becomes an interference fit in the bore. In other words, it seals.

My biggest concern with it is the high rate of expansion of the poly with heat. I could easily find my piston binding, and maybe even welding itself to the cylinder. In my vacuum pump there's little heat generated, and I've never had a problem with the poly pistons. In an air pump, the heads are going to get hot, and I'm doubting the ability of the poly pistons to stand up.

I'm having some problems with the math also. I can figure out torque from the motor, and the torque requirement to push the piston to the maximum pressure wanted, but there's a grey area. First, the piston meets with increasing resistance as the air is compressed. Secondly, the eccentric reaches a point at TDC where it is holding against the highest pressure, but isn't requiring any torque at that point. Over the 180 degrees the eccentric goes through, the torque reflected to the disc carrying the eccentric goes from near zero, reaches a peak, then decreases to zero again at TDC. I need to know what the maximum torque is so I can calculate the drive ratio. I'll see if I can draw a picture to illustrate this.

10-17-2010, 06:05 PM
Here's a drawing of the setup. It's shown with the eccentric at the halfway point of compression. Here, the eccentric should be placing a load on the driven disc of about half the maximum pressure, and that load becomes less at the rim of the disc because of leverage. As the driven disc continues to turn, the pressure above the piston continues to rise, but the disc reaches a point where it no longer needs to supply torque. The eccentric still bears the entire pressure of the air above the piston. Somewhere between those points, the torque required to turn the disc reaches a maximum. This is where I'm math challenged. Offhand, I'm going to suggest about 45 lbs of pull on the belt is going to be needed, where the eccentric is halfway between the rim and the central axis. Does this compute reasonably well?


Your Old Dog
10-17-2010, 06:09 PM
I run my brad driver from a 7 gallon portable tank and I can just about drive two sheaths of 1 1/4 brads for molding jobs. I carry it back out to the big shop compressor and fill it up to about 95 pounds. You might be able to do the same.