I'm finishing studying EFI systems, which set me thinking about induction more generally.

We inject fuel from a high-pressure common rail for many good reasons. We're capable of making stratified burn within a cylinder, provided we have full control of the throttle butterfly. We sometimes use exhaust gas recirculation to retard combustion and lower cylinder temperature.

Given those scenarios, why don't we add a common rail of high-pressure atmospheric air, and use injectors to introduce air, and exhaust when required, in a similar way to fuel?

Surely this would give me an engine that can respond quicker because there's no lag in the inlet airflow, have fewer mechanical parts and potentially reduce emissions by allowing me to control the oxygen content at the catalyst more easily?

  • 4
    Patent US 5381760 A had been submitted. Commented Aug 26, 2016 at 15:26
  • 29
    We do. It's called turbo or supercharger.
    – Chloe
    Commented Aug 26, 2016 at 17:23
  • 1
    Using a "common rail" to do all this would be impractical due to the volume of air to be injected at any given time.
    – Moab
    Commented Aug 26, 2016 at 21:05
  • 1
    @Chloe: None of the answers point this out. Would you be able to write one? Commented Aug 28, 2016 at 18:39
  • Another thought : do you consider "injection" just high pressure intake (not suction) or really with some kind of controlled injector like fuel injectors ?
    – Benj
    Commented Aug 29, 2016 at 13:00

7 Answers 7


Simple reason: volume. @ 14.7:1 stoich, your input into the cylinder would need to be 14.7x bigger (or push that much more) through a nozzle than you would the fluid which is fuel.

You state it would have fewer mechanical parts, but is that true? You'd have to furnish a mechanical method to create the high pressure air as well as introduce it into the system. You'd have to have some type of tank which would hold the high pressure air. Then that "high pressure" would need to be in the 3000-5000 psi range at a guess to ensure proper flow. Think of an air compressor which could keep up with the demand you are talking about.

Let's say we throw some math in the mix (and assume I'm not just being completely stupid ... though the jury is out on that one):

A 2L engine has a swept volume of 2L. If this theoretical engine was running, naturally aspirated and attained an 80% volumetric efficiency (VE), it would be taking in .8L of air every revolution of the crankshaft. The math:

  • 2.0L X .8 = 1.6L - Intake volume for all four cylinders @ 80% VE
  • 1.8L x .5 = .8L - Intake volume for every revolution in a 4 cycle engine
  • 600rpm x .8L = 480L - Amount of air at idle needed to maintain idle speed
  • 6000rpm x .8L = 4800L - Amount of air at redline to maintain fastest engine speed

Your system would need to move 4800L of air per minute in order to maintain that engine speed. That's about 170CFM. If you can haul something like this:

enter image description here

around on the back of your car, it might be doable. The 170CFM is a figure for the small, lower horsepower end of the equation. What about performance cars where you have three times as much swept volume (6.3L Chevrolet LT1 engine) with a greater VE (~85% at a guess). Those numbers are hugely greater. You'd be tripling the amount of air needed, which means triple the amount you'd be towing behind the vehicle.

Yes it could be done, but at what cost? The way air is introduced into the engine now is far more efficient and introduces far more air than you could reliably continue to pump air into an engine the way you suggest.

  • 1
    Of course, if you had a different way of supplying air, you build some really cool engines - but they would still be impractical for most applications using a regular combustion engine.
    – brichins
    Commented Aug 26, 2016 at 16:40
  • 1
    170 CFM isn't nearly as much airflow as you're implying with that towable whatever it is. A slightly oversized computer case fan can move more than that. Depending on static pressure requirements a fan like that might not be suitable; but a bit of quick googling suggests a few hundred CFM for the airflow rate a turbocharger; which is a bit bigger than a 120x38mm fan but not by that much. Commented Aug 26, 2016 at 21:17
  • 10
    @DanNeely - You are mixing apples and oranges. No way a fan case fan would make a lick of difference in this case. You have pressure and flow to consider. Especially if you are trying to pump air through a system which is going to inject it directly into a cylinder. With any back pressure whatsoever, the case fan would stop flowing air altogether. Commented Aug 26, 2016 at 21:20
  • 1
    @A.L - The math is general. Insert the engine size, volumetric efficiency of the engine, and max engine speeds to correspond whatever engine you have in mind. This example is for a 4-stroke engine, though. A 2-stroke engine will be computed differently. Commented Aug 29, 2016 at 14:04
  • 1
    @Pᴀᴜʟsᴛᴇʀ2 I replaced "impossible" with "practically really hard" in my last comment. Commented Aug 29, 2016 at 14:34

You have almost but not quite described the operation of either a Turbocharger or Supercharger. The idea of the air under pressure being injected from a common fuel rail would likely not work as it would be difficult to guarantee decent atomization.

  • 4
    How does TC and/or SC use injectors to introduce air into the system. It is my understanding the OP wants to do away with the valvetrane and just have two injectors: one for fuel and one for air. A turbo doesn't even come close to doing what is wanted. Commented Aug 26, 2016 at 13:07
  • 1
    Could be. We just took different routes I guess. Should let the OP clarrify. Commented Aug 26, 2016 at 14:11
  • 1
    @Pᴀᴜʟsᴛᴇʀ2, IMO, you (and perhaps OP?) are seeing a difference where there isn't one. If I took the fuel injection system of a car, replaced the components dealing wth fluids to one's meant for gasses (pump -> compressor, tank -> intake, change valve designs, etc), then expanded the pipes to accommodate the required airflow, I think I'd end up with an ordinary turbo system. I don't see a reason in your answer why you'd need 2k+psi, when 5-10psi fills the cylinder just fine in a normal engine.
    – mbrig
    Commented Aug 26, 2016 at 20:04
  • 1
    @mbrig - Never any taken unless it's a personal affront. You are just trying to explain, so good on yah! I always appreciating differing opinions and thoughts on it ... I definitely don't think I have the lock on any of this stuff! Your opinion is just as valuable, if not more so, than mine. Thanks for the input. Commented Aug 26, 2016 at 20:14
  • 2
    Isn't the common rail in this scenario just the inlet manifold?
    – RemarkLima
    Commented Aug 27, 2016 at 19:53

In many ways, you're describing a 5 stroke engine

5 Stroke engines are using a piston to provide a secondary means of compression for the AFR. Although, not injecting air they are compressing the air by mechanical means. What you describe with air injection requires massive volumes of air.

Think of a 5.0 Liter engine requiring 5 liters air every 720 degrees of revolution. At 4,000 RPM you would require 10,000 liters of air to be 'injected' every minute.

Air injection for emissions

The idea of injecting air is not a unique one. Many manufacturers have been injecting air into the exhaust in order to assist with oxidation of unburnt fuel at low RPM in catalytic converters. These were early versions of course, think mid-70's.

  • 1
    I think you are the only one that understands the question.
    – Moab
    Commented Aug 26, 2016 at 21:07
  • 1
    OP asks about injecting air into combustion chamber before (or during) ignition. "5-stroke" takes post-burn, post-expansion gases from combustion chamber to expand them a second time. "Exhaust reusing" is pretty much the exact opposite of "air injection".
    – Agent_L
    Commented Aug 29, 2016 at 11:36

High pressure gas is very difficult to create, much harder than high pressure liquid. It's because liquids are not compressible, so you can squirt them virtually as hard as you want, while gas will just absorb most some of your compressing effort and convert rest of it to heat (adiabatic heating). To compress air to the pressure necessary would require a reciprocating pump slightly bigger than the cylinder itself. So instead of doing that via dedicated pump, we compress the air with a component we already have. In-place compression gives the added benefit of recycling adiabatic heat.

What you are proposing would fit nicely to a 2-stroke engine. It already has a common rail of moderately high pressure air, the air admission into the cylinder can be controlled by inlet valve (if there is one) just as common rail injectors open to inject fuel. But the power required to inject air would be huge, just to put your needs into perspective: the 2-shaft Junkers Jumo 205 should theoretically require very strong gears to transfer half of it's power from lower shaft to the upper one where power was taken, but compressor was run off the lower shaft and took so much power that very little actually remained. Almost half of gross output was taken by a compressor and that engine reached intake manifold pressure nowhere near what you need.


Here's a variation that I've thought about at great lengths. Even doing some of the preliminary math.

IC engines don't need air. They need oxygen. So... eliminate the valvetrain entirely, and have two sets of injectors: One for liquid hydrocarbons, and one for liquid oxygen.

Granted, I'm not considering the expense or safety issues in this brainstorm (I rarely do.) I also have not really found a piezo or solenoid type injector, or even a HPOP diesel type, that would operate at the frequency and pulse widths needed at the LOx temperature around -300 degrees F, with crank RPM in the 7000 range.

However, there's more to it than valve train elimination. Imagine the adiabatic cooling from LOx returning to a gas in the combustion chamber. I am confident with the right crank, rod, and piston materials, you could safely run 15:1 or 20:1 compression, and have a wonderful emissions profile as well. The head would be reduced to nothing more than a thick durable injection plate... no moving parts. Exhaust could be handled by a two-stroke or wankel style "reveal" port, with a modified Atkinson cycle with a longer exhaust stroke.

This is a very long way from reality (much like myself), but I do think it illustrates a practical variation on the OP's concept. Compressing the air in order to inject it through a very small orifice would likely cost more power than the gains realized. But a tank of liquid oxygen already has the "work" put into it, is reasonably mobile/portable, and has that huge added cooling effect -- perhaps so dramatic as to scale back or virtually eliminate a water/glycol cooling system.

I'll be taking volunteers in a decade or so for Official Test Pilots. The glory will be yours. cuz no way am I gonna ride in it...

  • 1
    ULA is working on this as part of their Integrated Vehicle Fluids (IVF) concept, for their ACES rocket stage: basically they run an internal combustion engine off the propellant and oxygen supplies of a rocket stage. (ulalaunch.com/uploads/docs/Published_Papers/Extended_Duration/…)
    – Hobbes
    Commented Jun 19, 2017 at 11:30
  • Nice! probably not in my lifetime
    – SteveRacer
    Commented Jun 20, 2017 at 5:36
  • The big drawback of using LOX here on Earth is that the tank needs to be kept cold. You need thick insulation and an active cooling system to keep boiloff to a minimum. Also, rocket engines that run on kerosene/LOX, run on a mixture of ~2.2 LOX:1 kerosene, so for every 50 litres of fuel you need 110 l of LOX. An LPG tank that size would take up half the trunk space, an insulated tank is even worse.
    – Hobbes
    Commented Jun 20, 2017 at 6:20

I think the concept would be like taking a piston compressor to pump air into a piston engine, so energy to pump the air compressor pistons would counter the energy developed by the engine pistons. Adding loss in the engine do to heat would seem likely to be a negative gain.

But is it possible a gain could be realized in this concept in a compact and self contained form would be taking 1/2 of the pistons in a V8 and turn them into compressors to pump the air into the driven pistons.. Maybe turn the whole thing into a two cycle with adjacent pistons using the scavenge port for intake tied to the output of the pump piston.

  • Not really answering the OP's question, yet nice idea :) It somewhat needs a calculation of the power loss and overheating of air, maybe needing an intercooler as in turbocharged systems. I also wonder how to deal with the cams & camshafts to avoid unnecessary compression of this air in "compression" phase which would be avoided with your technique.
    – Benj
    Commented Aug 29, 2016 at 12:59

Direct to cylinder Fuel injectors used to add just a little bit more air into cylinders straight after inlet valves shuts and before the air gets compressed (which has to be fast on/off), no air tank needed if it runs only when engine runs (via a belt). And if it stops, it'll have no effect on engines normal performance because it's one way valving and not interference. That should give just a little more power depending on size of injectors used.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .