# Is it possible to build an air compressor with no moving parts?

Is it physically possible to build an air compressor with no moving parts? I'm envisioning a thermodynamic cycle capable of compressing air with no moving parts and operated in a stationary way. There is no limitation on the compression factor, provided it is significantly greater than one (1.1, 2, 100...) but the design must be realizable.

Zero moving parts is a big constraint. You can interpret it as the absence of pistons, crankshafts and other complex mechanisms that will wear out over time. If some moving parts are needed, what are the minimum required moving parts, with the lowest maintenance requirements?

• How much pressure are you anticipating being developed? Are there any limitations on the power source? – hazzey May 14 '15 at 19:28
• So this compressor is part of the thermodynamic cycle, where the isentropic/adiabatic/isothermal compression would normally happen? – Trevor Archibald May 14 '15 at 19:38
• A valveless pulse-jet engine would accomplish this, although it wouldn't really compress air so much as it would compress combustion products. – Carlton May 15 '15 at 1:53
• Trapped air in a radiator – Ewan May 20 '15 at 12:59

It is possible to make a air pump/compressor with absolutely no moving parts.

Make a small, electrically non-conductive chamber and stick 2 electrodes into it. Pulse an arc through the electrodes so that the chamber pressure rises and falls rapidly. Use tesla air flow check valves (because they have zero moving parts) for intake and exhaust.

When the arc forms across the chamber, the enclosed air will become superheated and expand mostly out of the exhaust port due to the tesla valves, then the chamber cools and pulls in fresh air through the other tesla valve.

This can also be done with any pulse-able heat source.

If some type of sealing check valve is used to replace the tesla valve, you can maintain high levels of compression too.

You won't attain very high flow nor very high pressure (plus you must control the gas outflow), and it's a bit of cheating on the "no moving parts" but you can make a compressor with no moving solid parts. Plus it's very wasteful on energy - you'd be far better off converting the energy to electricity and using a classic compressor, but if for some reason you want to get pressurized air with no mechanical parts, that would be it.

You need a strong flow of water and a large altitude difference. Take a rapid, strong flow of water. Using either the Venturi effect or some other way aerate the water - mix it with air/gas bubbles. As the water travels down the pipe rapidly, the bubbles travel upwards slower than the water flow - they are carried downwards. Despite rapid flow, the outlet of the pipe is somewhat constricted and as result, the water pressure grows with the height of the water column, and with the water column pressing down, the pressure in the bubbles grows as well.

Then the pipe turns sideways. The bubbles are no longer dragged downwards, and so they move towards the upper side of the pipe, eventually escaping the water flow and amassing in a reservoir above the upper edge of the pipe; they are at the same pressure as the water - which must yet to pass a constriction, so its pressure is fairly high.

Of course it won't work with narrow pipes and low flow as water viscosity would reduce the downward speed and distribution of pressure. And the amount of air taken out must be controlled because - unlike in normal compressors - the pressure is maintained at all times, but the volume of the gas drops, and if you exhaust it, you will start drawing water. And obviously there are better uses for a high-volume, high-pressure water flow than compressing some air. It's wasteful, because most of the water's energy is lost. Still - the concept is sound; scaling the height of the downwards pipe you can attain quite reasonable pressures, at about 1 bar per 10m. And the only physical moving part is the outlet valve which does not need to move during operation.

Trompes are very similar and were once used to provide compressed air to furnaces.

The high-voltage construction in an ionocraft provides air flow without any moving parts. The pressure difference is very small, but combining multiple stages will increase it.

The Knudsen pump has zero moving parts and is based on thermal diffusion (gas flows from low to high temperature end of a tube). The back pressure the flow can resist is termed the thermomolecular pressure difference and is a function of the ratio between the gas mean free path and the dimensions of the tube walls - modern advancements in the concept have used various materials such as zeolites consisting of nano-scale pores to improve this ratio.

Yes. An amazing device called a tromp or trompe. The flow of water over and down a funnel with a straw or tube elevated above the water level. The flowing water drags the surrounding air along with it, pulling the air through the straw and oxygenating the water with small air bubbles. The water travels through the tube under the river or stream it's placed in and as it moves horizontally through the tube the air bubbles escape into one or two air tanks connected to the pipe..this compresses the air. As long as water is flowing, even quite slowly, the air will compress in the tanks.

About 100 years ago a large scale mining operation in Canada used a tromp to power all the pneumatic drills etc. Why it's not being used today is puzzling?

Interpreting "moving parts" as meaning that every solid part of the device has to be rigid, and the requirement as the capability to deliver a constant stream of air at a non-trivially greater-than-ambient constant pressure, I suspect the answer is a qualified no.

I'm also assuming that fuels and working fluids don't count as "parts."

It seems reasonable to assume that any remotely practical air compressor will operate under conditions where the ideal gas law applies, so $p \propto \frac{NT}{V}$.

One approach would be to imitate a typical air compressor and try eliminating as many moving parts as possible. For example, something like a hydraulic ram could eliminate pistons, impellers, screws, etc. and allow us to extract energy from a moving body of water to compress air, but it still requires valves. A valveless pump as seen in this video requires a special rotating piston. A basic siphon has no moving parts at all and it can create pressure if you enclose the lower reservoir, but is totally impractical as part of an air compressor—and even if it weren't, you'd still need some sort of valve to deliver the pressurized air.

Another approach is to manipulate temperature, which sounds like what you've got in mind. It's easy enough to generate heat without moving parts; a burner or an electric coil will do it. But how do you get around the valve issue? In order for the pressure to build, you need an enclosed space, and once you have pressure, the air needs to exit that enclosed space. If you wanted to get creative, you could try something like a diaphragm with an aperture that only opens when the diaphragm has expanded; the pressure will then make its own exit. But an expanding and contracting solid diaphragm or bladder also seems like a moving part, to me. It might be more durable than other types of moving part, I suppose, but then again it might not.

To produce a constant stream of pressure then you need a holding tank and the magnitude of your delivery pressure will be reduced significantly based on the upper limit of pressure you can develop in the holding tank and how quickly you can develop it. The Tesla valves suggested in netduke's answer are very clever but they're really differential flow-limiting devices; I don't see them being able to develop and hold pressure in a tank that you could release on demand for pneumatic power.

So the reason it's a "qualified" no is this. In theory, if you accept that your air compressor may be totally impractical for most purposes, you don't count elastic deformation as movement and you "cheat" a few times with valves and regulators, then yeah. You can create a device that compresses air in a tank, and then do with it what you will. In practice, it sounds like a poor idea that doesn't scale well, but it's an interesting exercise to play with.

Another qualification is that you might get a completely different answer in a microfluidics context.

• Well, no moving parts constraint is a problem. Minimal number of moving parts with a preference for low maintenance ones. – user3368561 May 14 '15 at 22:34
• Minimal moving parts is not much of a constraint at all. What does "minimal" mean? How is one moving part judged against another? The constraint is the only thing that makes this question interesting, in my opinion. – Air May 14 '15 at 22:38
• Maintenance, durability, manufacturability, etc. A matching valve for one particular application can be found virtually anywhere. In contrast, a piston and crank must be custom manufactured. A real world example: valved pulsejet vs gas turbine. – user3368561 May 14 '15 at 22:45

It's absolutely possible and has been done for some time now in thermoacoustic compressors. Originally they were developed for cryo-coolers to condense gases into liquids and that remains their principle application though there are companies working to bring this technology to a consumer level. These compressors have no, or at most one, moving parts (the sound source). They also have the benefit that they don't use any greenhouse gases.

• You can't make the claim of "doesn't use any greenhouse gases" for anything that requires energy to operate. – whatsisname Jan 17 '16 at 3:23
• @whatsisname To satisfy your pedantry, what I meant is that there are no refrigerants containing greenhouse gases used in these compressors. – DLS3141 Jan 17 '16 at 5:04

Sure you can! But don't know how practical it will be. Let me describe a simple way to do it. My assumption is, when you say compressed air, you mean increase the pressure of the air to do some work. If that is the case see the following.

Take a 12 inch inner diameter steel tube or pipe say 24 inch long and weld a steel plate on each end. The weld must be around the entire circumference. In the center of each side plate bore a one inch hole and tap it for a screwed connection.

On one side of the pipe where the tap has been made, screw in a small ball check valve with a spring to keep ball closed. The valve should be positioned that the ball will open inward to the pipe.

On the other end of the pipe, screw in a 1 inch 90 Degree hand operated ball valve. The ball valve should be positioned such when opened it will vent the air inside the tube out.

On the side plate that has the ball check valve, drill a hole and tap it for a suitable pressure gauge and install the gauge.

Close the hand operated ball valve to trap the air inside the pipe. Use a heating element or flame to heat the pipe. As a result of the heat the trapped air will expand and the pressure will rise. When the pressure rises to the desired pressure, release the the compressed hot air by opening the hand operated ball valve.

This is easily done using a supersonic flow. Either heat addition or shock waves.

• Could you add some more detail? – hazzey May 19 '15 at 2:26
• If you have enough speed you can use geometry to compress air, but it only will work if you have a stagnation pressure greater than desired static pressure. My question talks about a case where supply flow has lower stagnation pressure than desired static pressure, so you have to increase it with a suitable method. – user3368561 May 19 '15 at 9:41
• While true that this works (see ramjets) I am downvoting because the question specifically states that it should be stationary. – regdoug May 20 '15 at 14:10

Assuming this question is a theoretical one, the answer can be heating the air. It is similar to afterburners in military jets. See: Wikipedia\afterburner You do not have to add the combusting fluid into your flow, if you use channels, as in a domestic water heater.

The principle of the Afterburner, as related to increasing the pressure of an airflow is quoted hereby: "The afterburner then injects fuel downstream of the turbine and reheats the gas. In conjunction with the added heat, the pressure rises in the tailpipe and the gas is ejected through the nozzle at a higher velocity. The mass flow is also slightly increased by the addition of the fuel."

• I am aware of the principles behind the afterburner, I mean how can you utilize that increase in pressure without moving parts? Not to mention that reaching the flow speed in afterburners require A LOT of moving parts. – Algo May 22 '15 at 19:18
• oh, I see now.. my explanation suggests heating the air while it is moving inside a channel, as in a 1D flow, and does not need to include the turbomachinery part of the engine.. will clarify. – Gürkan Çetin May 22 '15 at 19:36