Using a multicylinder engine as a compressor

[Justme]

The average engine compression is 130psi

At cranking speed. I would expect it to be more at running speed.

 

[hotponyshoes]

In terms of compressing air it will be the same at any speed. The amount will be dependent on the compression ratio.
Well, it will vary slightly as at low speeds some air will seep past the piston rings etc and at higher speeds it might not have enough time to draw in or push out the full capacity.

Any increase over that would require forced induction or something like fuel burning with the air to increase pressure.
Google showed something like 300psi on tick over to 1k psi loaded.

 

[outofthefire]

In terms of compressing air it will be the same at any speed. The amount will be dependent on the compression ratio.

The amount will be dependent on the displacement of the cylinder, nowt to do with the compression ratio. The air's not getting compressed in the cylinder, whats left in the gap between the top of the piston and the cylinder head will still be there when the piston starts to go down.

 

[hotponyshoes]

The amount will be dependent on the displacement of the cylinder, nowt to do with the compression ratio. The air's not getting compressed in the cylinder, whats left in the gap between the top of the piston and the cylinder head will still be there when the piston starts to go down.

I ment the amount of pressure available not the volume

 

[Wallace]

There will have to be compression in the cylinder to overcome the exhaust reed valve as pressure builds up behind it when closed, at least that’s how I see it.

 

[Seadog]

There will have to be compression in the cylinder to overcome the exhaust reed valve as pressure builds up behind it when closed

Of course you will. If you've an elephant sitting against one side of a door you're not going to be pushing it open with your finger. you need to exert enough force to overcome the elephant

 

[hotponyshoes]

Yea,
If the cylinder has a capacity of 1 litre when the piston is at the bottom of its stroke and a 10:1 compression ratio it will draw in 1 litre of air at atmospheric pressure (1bar) and compress it to 10bar.
That pressure will push out of the exhaust valve and pump into the tank until the pressure in the tank reaches 10bar.
At that point it will not go anywhere and when the piston goes back down the 10bar will decompress to one bar again and it wont draw any more air in to continue until the tank pressure starts dropping (when you start using air from it)

In practice you would need some way of dealing with this to stop it overheating I expect

 

[Screwdriver]

I was hoping to provide a few rough calculations to derive pressure and cfm from swept volumes, compression ratios etc. but it looks like a nightmare.

Here are some handy links I found: http://www.aircompeq.com/formulas.htm, https://weldtalk.hobartwelders.com/...ng-processes/15744-calculating-compressor-cfm, http://www.epi-eng.com/piston_engine_technology/volumetric_efficiency.htm

None of them specifically help with converting a petrol engine to air compressor except for one common element. It appears the cfm largely depends on horsepower (unsurprisingly) but usefully it would seem you can expect 3 to 4 cfm per horsepower. I flicked through a fw compressor specs and that looks about right. So if you did chop a VW engine in half and assuming a regular VW engine is about 50 bhp and half of that is 25 bhp, you might hope for 20 bhp from two cylinders, maybe 10 bhp if it's not running flat out at high rpm.

10 bhp should give you circa 30-40 cfm which is not to be sniffed at. Pressure will depend on the compression ratio as @hotponyshoes is describing above but that would be a minefield to attempt to calculate, lets assume it should be a respectable 120-140 psi.

So the answer is that the advantage in using a converted internal combustion engine is that it gives you access to high levels of horsepower and therefore cfm without the need for a massive electric motor that would probably trip most consumer mains. Efficiency will be horrible but you would put up with that for the availability of massive cfm. You would be better off driving a regular compressor with a separate petrol engine but then you'd need a big expensive compressor

I think the upshot is, if you're planning on building a sandblaster for example, then a converted petrol engine is actually a pretty good idea.

 

[hotponyshoes]

The calculations look complicated.
As a starting point you would need to know what tool you wanted to drive from it and how much cfm at how many psi it needed.
Work backwards from that to see what you would need. I don't need one so I'm not trying to work that out!

 

[Wedg1e]

id use a rover v8. have the centre 2 on each back firing and the outside 4 as compressors, take the pushrods out to take them out of the normal valve sequence, leave the exhaust valves in to block the port and remove the inlet, fit poppets to the inlet valves and the spark plug holes. . should work quite well and i cant think of a simpler solution. will obviously need to fit a modified intake manifold too.

Just to stir up an old thread...

With the RV8's firing order being 1-8-4-3-6-5-7-2 I fear your plan would be doomed
You'd have puff-puff-drive-drive-drive-drive-puff-puff when you'd really need drive-drive-puff-puff-drive-drive-puff-puff or even better drive-puff-drive-puff-drive-puff-drive-puff...
I can say from experience that an RV8 will chuff along even at idle with 2 cylinders on each bank turned off.

 

[minimutly]

Not to mention you tanks per gallon would be dreadful - running a Rover v8 on choke for warmup would make your eyes water - for a couple of reasons!

 

[hotponyshoes]

It could well be possible with a 4 cylinder motor.
I had a rover 820 once. That spent most of its life running on 3 cylinders and the remaining time on 2.
Not a lot of use as a compressor as the valves sticking open was causing the problem but that was a pretty big car and would still do 70mph on 3..

 

[minimutly]

Common problem on them, they modified the valves on later versions. I have a tuned one in a stripped out Metro....

 

[minimutly]

A lot of guff has been written on this by people who should know better - stuff about compression ratios, 10 bar pressure, not a lot or nothing in fact about controlling the engine.
It's simple, at least I think it is. You modify two cylinders to be fully scavenging - ie no space above the piston, so they empty fully. You have the normal pressure switch and unloader for control - effectively stop start, or rather start unloaded, load up, once up to pressure unload, time delay, shut down, repeat.
Lastly I would bin the carb and run it on lpg, with some means of cold start enrichment and a blowback valve...
I still think an old crossflow would be ideal due to the flat head....

 

[cheb]

I'm about to scrap a diesel Modus too....

 

[minimutly]

Hmmm, that would have a flat head? You'd need to engineer a flat topped piston, maybe from a different model engine?
Some diesels came in direct/indirect forms I believe, this might be an answer? The Perkins Prima was a flat head dished piston direct injection, is there a piston/rod that would fit? The petrol engine rod and piston would be the right length, but the be journal is bigger on the diesel.

 

[Screwdriver]

A lot of guff has been written on this by people who should know better - stuff about compression ratios, 10 bar pressure, not a lot or nothing in fact about controlling the engine.
It's simple, at least I think it is. You modify two cylinders to be fully scavenging - ie no space above the piston, so they empty fully. You have the normal pressure switch and unloader for control - effectively stop start, or rather start unloaded, load up, once up to pressure unload, time delay, shut down, repeat.
Lastly I would bin the carb and run it on lpg, with some means of cold start enrichment and a blowback valve...
I still think an old crossflow would be ideal due to the flat head....

I think you mean the opposite of what you're actually saying. Scavenging works by introducing an overlap where both the intake and exhaust are open at the same time. It's a gas flow thing and frankly, best left to those who study fluid dynamics because the calculations are horrendous. Certainly the stock valve timing is going to be a big issue with compressor engine hacks but my simple evaluation suggests the last thing you should expect is efficiency.

Having "near zero" space above the piston is like saying maximum compression and implies some serious modifications to both the valve timing and the valves themselves. That could also lead to unforeseen side effects and I rather suspect you'd be generating a LOT of hot air. What benefit do you expect to get from "no space" above the piston?

 

[hotponyshoes]

Some of the older indirect injection diesels could be reduced to very little cylinder space with a thinner head gasket. Should be easy enough to blank out the injection chamber or even fit it with a reed valve.

Save messing about with the valve timing just remove the camshaft, weld the valves shut. Use the ready-threaded injector and glow plug holes for nrv's and job done.

I still think there are easier/better ways of making an engine driven compressor if yoynwant one fitted in a running vehicle.
An aircon pump looks ideal to me. Maybe fit 2 of them (might find a 2nd bracket from a lhd car?) Set one on a high ratio pulley for quick tank fill and the other on a low ratio to build up higher pressure. Switch the existing 12v clutches with a tank pressure switch.

Turbos also blow out a lot of air. Fit a exhaust manifold and turbo to a non-turbo engine and you could use the output to fill the tank quickly. Not sure how much pressure a turbo can generate but I think F1 ran 5bar at one stage. The boost is limited to protect the engine which would not be an issue so I expect you could go a lot higher. Possibly need sequential or compound turbo charging.

I can see it being an interesting project but in real life I'm sure just getting a compressor head and belt driving it from a air cooled engine would be a lot simpler, cheaper and probably more efficient.

Edit: Tractor pulling contests in the states seem to be claiming up to 300psi of boost from a turbo..

 

[minimutly]

I think you mean the opposite of what you're actually saying. Scavenging works by introducing an overlap where both the intake and exhaust are open at the same time. It's a gas flow thing and frankly, best left to those who study fluid dynamics because the calculations are horrendous. Certainly the stock valve timing is going to be a big issue with compressor engine hacks but my simple evaluation suggests the last thing you should expect is efficiency.

Having "near zero" space above the piston is like saying maximum compression and implies some serious modifications to both the valve timing and the valves themselves. That could also lead to unforeseen side effects and I rather suspect you'd be generating a LOT of hot air. What benefit do you expect to get from "no space" above the piston?

You reallyndon't understand this do you. A compressor is not an engine, you don't need or want a space above the piston - that said I accept your statement about scavenging - to a degree. Scavenging happens due to differences in pressure - whether that is from exhaust suction or by the piston pushing it out is immaterial, but I accept with a compressor it will all be by the latter, whereas an engine uses both. No calcs needed, the valves would be removed and replaced with reeds.

 

[mtt.tr]

Some of the older indirect injection diesels could be reduced to very little cylinder space with a thinner head gasket. Should be easy enough to blank out the injection chamber or even fit it with a reed valve.

Save messing about with the valve timing just remove the camshaft, weld the valves shut. Use the ready-threaded injector and glow plug holes for nrv's and job done.

I still think there are easier/better ways of making an engine driven compressor if yoynwant one fitted in a running vehicle.
An aircon pump looks ideal to me. Maybe fit 2 of them (might find a 2nd bracket from a lhd car?) Set one on a high ratio pulley for quick tank fill and the other on a low ratio to build up higher pressure. Switch the existing 12v clutches with a tank pressure switch.

Turbos also blow out a lot of air. Fit a exhaust manifold and turbo to a non-turbo engine and you could use the output to fill the tank quickly. Not sure how much pressure a turbo can generate but I think F1 ran 5bar at one stage. The boost is limited to protect the engine which would not be an issue so I expect you could go a lot higher. Possibly need sequential or compound turbo charging.

I can see it being an interesting project but in real life I'm sure just getting a compressor head and belt driving it from a air cooled engine would be a lot simpler, cheaper and probably more efficient.

Edit: Tractor pulling contests in the states seem to be claiming up to 300psi of boost from a turbo..

On vehicle i think the old swash plate ac pumps are hard to beat, just need an oiler on the input and oil trap on the output.

This was something intended to do on my landy years ago. Even had a nice ali trcuk receiver