Efficency of an IC engine as an air pump?

[Milanoguy]

What is the thermal efficiency of an IC engine, strictly as an air pump? Turbochargers typically have a quoted efficiency of 90%, modern roots superchargers(Eaton) have a max of about 60%, a twin screw supercharger(Lysholm, Autorotor) is a little better, at about 70%.

Charles Taylor’s “The Internal Combustion Engine”, states that piston compressors can give excellent efficiency at pressure ratios of 1 to 2. From this I gather that as the compression ratio goes up the thermal efficiency goes down.

What is the efficiency of an IC engine at say the 8 to 1 compression rating typically found in forced induction engines?

Do other factors beside compression ratio affect the thermal efficiency of an IC engine as an air pump?

 

[Fabrico]

Only little things like wasting 2 of every 4 strokes.

 

[Fabrico]

Oops, forgot to mention the entire valve train.

 

[NickE]

fabrico -- good one (valve train)

Milanoguy -- as fabrico says the valve train is a huge energy waster, most (well all that I've seen) use reed valves to govern the flow and actate for intate and compression.

 

[j2bprometheus]

Milanoguy -

Where did you find a turbocharger with a quoted efficiency of 90 % ?

For an automotive size turbocharger, the peak efficiency of the compressor is about 75 %.

To illustrate, here's a link to a commercially available compressor where the peak efficiency island is 72 % :

http://www.turbocharged.com/catalog/compmaps/fig1.html

 

[Warpspeed]

Agree with the ~72% turbo PEAK efficiency, and that is only at one exact defined flow and pressure ratio. It will be a good deal less than that everywhere else, as a flow map will quickly show. That is only the adiabatic efficiency of the bare compressor. Frictional power lost in the bearings will lower that efficiency a tad more.

The obvious problem with IC engines used as air pumps is the very low compression ratio. Too much dead volume, and that lowers volumetric efficiency. If you test with a compression gauge, and it can just manage a reading say 160 psi, then obviously if it was pumping against 160 psi back pressure, flow would fall to zero. So flow will fall off very rapidly as pumping head increases.

I recall many years ago, a manufacturer was selling a modified VW flat four engine as a mobile commercial air compressor. Two cylinders on one side operated as a normal gasoline engine. The two cylinders on the other side had special cylinder heads fitted with reed valves, and minimum clearance volume. I believe this made a very effective air compressor.

 

[studedude]

The real problem here is the initial premise: using an air pump {forced induction) to run another air pump. It's like looking in a mirror with another mirror right behind you. Diminishing returns.

 

[globi5]

Diesel piston engines have compression ratios of over 20, which means pressure ratio must be even higher (adiabatic compression). So at least at high pressure ratios piston engines should be very efficient compared to pretty much anything else. The reason is probably due to the excellent sealing performance of a piston engine compared to other compressors.
Also, the Diesel engine with the highest efficiency is still more efficient than the most efficient gas turbine eventhough axial compressors reach effciencies of 90%. If the piston engine had a bad compression efficiency this would hardly be possible.

One drawback of piston compressors is that they are not very compact.

 

[jbthiel]

A little off topic, but I saw an air compressor that was made out of a surplus WWII jeep engine. The engine was reworked with a new head and manifolds so that the inner two cylinders ran on gas and the outer two pumped air. The pumping cylinders were equipped with reed valves and a seperate intake. Certainly not the most efficient but it was compact and cost effective.

The only scary bit was that the lifters of the outer cylinders had to be held in place with hose clamps to prevent all of the lube oil from escaping. I don't know why they didn't include some type of drop in plug - but the farmer swore that was how it was running.

Anyways - Happy Holidays.

 

[MikeHalloran]

That half engine/ half compressor thing has been done commercially with smallblock Fords, too.





Mike Halloran
Pembroke Pines, FL, USA

 

[Azmio]

i have also been wandering on whether an IC engine is an efficient pump during the induction stroke. Not to mention about the piston rings that contribute as much as 40% friction of the entire engine.

Anyone has any idea on whether a dedicated pump air used in as an air compressor operates with much higher mechanical efficiency than an IC engine during the induction stroke?

 

[NickE]

In an electric motor operated compressor, or any constant RPM motor. The induction tract could be designed to increase VE at the operating RPM. (Unless that doesnt work with reed valves.)

Is dead space considered part of the exhaust, or intake?

That would definately have to be changed wether diesel or gasoline were converted to an air pump.

 

[globi5]

I don't know about the contribution of the piston rings towards total mechanical friction of the piston engine.
However, I know that mechanical friction is small compared to the work required to compress air otherwise there wouldn't such a tremendous difference of effort required between cranking a piston engine with and without spark plug.
Also, since there's an oil film between ring and cylinder 40% is a little steep.
Thereofore, I don't think a piston based air compressor is significantly more efficient than an IC engine as far as compressing air goes. However a piston air compressor can be made lighter (no combustion pressure), it doesn't need to withstand high temperatures and there's no need for active valves. Still, dead space of the IC engine will need to be reduced to 0, if used as an air compressor, otherwise it will harm efficiency.

I believe there is that Scuderi (?) engine concept that uses 2 pistons one for compression and one for expansion, but at least to my mind it has more disadvantages than advantages.

 

[Fabrico]

There have been commercially available conversions for at least S/B Ford, S/B Chevys, and air cooled Volkswagons for years. Some of this type or larger can be found on big semi-trailer vacuum tanks. The more serious conversions seem to use reed valves more often than automotive valves on the air pumping cylinders.

 

[Warpspeed]

The efficiency of a piston compressor can (within reason) be made as high or as low as anyone would desire to make it.

There are practical reasons why some piston compressors have deliberate design features to limit the mechanical drive power requirement so as not to overload the drive motor.

If the compression ratio is kept deliberately low, that is the ratio of internal volume change at the extremes of piston motion, then the output pressure will be self limiting. As the output back pressure rises, flow will gradually reduce. That may sound very undesirable, until you have to spec a motor to drive the thing.

Which is better, a fairly flat drive horsepower curve with rising output pressure, or a very steeply rising power requirement with increasing output pressure? How easily is it able to start up against full applied back pressure?

Another way to limit drive power is with deliberate reduction of intake valve flow area. Very approximately mechanical drive power will be flow x back pressure. So if you can place some deliberately imposed limit on flow, once again the drive power requirement can be somewhat controlled. This feature can be very handy to keep the drive horsepower within an acceptable range if the operating Rpm is highly variable.

Which is better, an acceptably constant flow and drive torque over a wide operating Rpm range, or a steeply rising output flow and drive power requirement with operating Rpm ?

Now suppose you were given the task of designing a piston air conditioning compressor for a vehicle. It had to operate over at least a 10:1 Rpm speed range, and a vastly wide range of operating pressure.

You might be tempted to run a fairly large displacement compressor to get good performance at engine idle, and use the low compression ratio, and small valves trick to deliberately flatten the drive power requirement and even out the flow characteristics. This is in fact done to perfection.

In one respect these compressors are a hopelessly inefficient design. But they are designed to provide efficient cooling over a wide range of operating conditions, not be super efficient within themselves.

Likewise, commercial air compressors are usually designed to offer excellent and safe performance with a given drive motor. If you "hotted one up" by increasing the compression ratio, and porting the cylinder head (don't laugh!) would it be a better compressor if it then sometimes grossly overloaded the drive motor at certain times?

When I was a poverty stricken student, I used a belt driven twin piston refrigeration compressor as an air compressor at home. To start with I just could not get it to sufficiently load the drive motor, no matter how I changed the pulley drive ratio!

So I went to work on the cylinder head, and "hotted it up" a bit at a time. I still use that compressor, and it now loads the 7.5Hp electric motor beautifully.

 

[dcasto]

milanoguy,
Were you talking thermodynamic or volumetric efficiency. PD piston compressors like a car engine could run as high as 82% thermodynamic efficiency at low compression ratios say under 3:1. Volumetric effiency is a function of clearance, RPM, and compression ratio. I've seen combinations of those that result in volumetric efficiecies from 80% to 30%.
Vacuum trucks use V8 engines as compressors. GasJac markets Ford engines where one bank in an engine and the other a natural gas compressor.

 

[Azmio]

another sad thing about using an IC engine to induct air into the cylinder is the fact that there is so much restriction from air filter, throttle, valve and intake port. Not to mention that the maximum air induction happens only when the piston is at its peak speed (somewhere in the middle of the stroke).

I remember a long time ago when i was a kid, we opened up a bicycle pump and if my memory is good, I thought i saw an elastic rubber or leather that will efficiently seal the cylinder during induction and compression. I have a feeling that it has higher efficiency than IC engine acting as a mechanical pump.

I also know that centrifugal pump or regular fan can blow air with much better efficiency, provided that we dont need such a high pressure buildup.

 

[Fabrico]

In my previous post I meant to say disc valves not reed valves.

The better conversions do not retain the original cylinder head(s). The air cooled VW based compressor, which is still around, also has one side powering and the other side pumping.

Warps point reminded me of the restrictor plate stock car engines. Before the limits they were running 17:1 C/R with the thin air.

 

[Milanoguy]

Hi Dicasto

Thank you for being the first poster to answer the question I posted back in June. Yes I was talking about termal efficiency. What is your source for a PD(postive Displacement) piston type compressor having a thermal efficiency of 82% at compression ratio’s below 3 to 1?

What would the thermal efficiency be at a compression ratio of 8 to 1?

Can you link to a website or a white paper which discusses these matters?

 

[Tmoose]

I think of VE as how "full" the cylinder gets.
Plenty of normally aspirated race engines achieve VE over 100%. I don't think its the compression ratio that gets them there, or else every diesel would have VE > 100%.