In situ measurement of swept volume 2

[SincoTC]

I have been asked by a colleague who is involved in club-level motor racing, if there is any instrumentation or method of measuring the swept volume of an engine cylinder "in situ", such that it can easily be carried out by a scrutineer, to see if it complies to the regulations without tearing the engine down?

I could only think of using a burette to introduce a benign light oil through a spark-plug hole at TDC until it spills, and then checking the amount added to spill at BDC, however, he wants to be able to check horizontally opposed engines too, so that rules that method out.

Any ideas? Thanks for any suggestions!

 

[NormPeterson]

P&G gauge, claimed to be accurate to about 1%.


Norm

 

[Lou Scannon]

I see a potential problem at BDC in that typically either the intake or exhaust valve will be open - unless the open valve is higher wrt to gravity than the spark plug hole, the moment of spillage will not correlate that at TDC.

That aside, you could check horizontal engines by tilting them so they are vertical.

 

[SincoTC]

Thanks for a good lead Norm, I'd failed to find that one! Unfortunately, its use requires removing a pair of pushrods, which is a step further than the club's scrutineers want to go for fear of upsetting the owners or causing damage (especially disabling valve motion on OHC units)!

hemi,

I agree, this method would only work where the plug hole is below the lowest point of an open valve (unless it's possible to disable the valve, which is not desired as mentioned above). I'd more or less dismissed this idea, because of it's inability to cope with inclined or horizontal cylinders without the use of some kind of rollover fixture for the whole vehicle.

 

[TDIMeister]

Conceivably, if you inject a known quantity of conditioned air through the spark plug hole and motor the engine in that way while simulaneously plotting the pressure indication, you will end up with a P-V diagram (actually P-°CA), with which you can determine the minimum and maximum volumes, the difference of which is the swept displacement. You need to run several full revolutions; this takes care of valve timings but you will see a distinct and familiar P-V trace developing. For this you will need to design a threaded adapter that fits the spark plug hole that sends air (mass flow rate precisely measured)into the cylinders and also pressure transducers built-in to it (e.g. Kistler). You also need an optical pickup at the flywheel notch or crank sprocket to identify TDC and revolutions. To top it all off you will need a proper data acquisition processor.

This method assumes minimal blow-by; might help to squirt some thick oil on the cylinder walls beforehand to seal the rings in the same trick people use to check compression leak-down.

 

[patprimmer]

It still won't correct for the poor volumetric efficiency of a high speed engine being tested at low speed

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[SincoTC]

Thanks for the additional comments from Pat and TDIMeister.

Pat: I was worried about engine speed and VE when my colleague said that the scrutineers had suggested collecting and measuring the output from the exhaust pipe(s), while the engine was spun over without fuel for a known number of rev's. I told him that it just wasn't that simple but I'd look into it.

The P&G gauge Website says that all plugs are to be removed, so it must only be measured at cranking speed (albeit faster without the plugs), so VE variations will probaly be quite significant, but presumably allowed for in their calibration. The claimed accuracy is one cubic inch per cylinder, but as the range of instruments goes from 500 to 800 cubes, I wonder how well this could be extrapolated for tiddly little European lumps?

TDIMeister: I wasn't sure whether the air injection was to be fitted to a single cylinder so as to administer a known input at the power part of the cycle, or be fitted to all cylinders to "Air Motor" the engine as well. If it was the latter, then I guess that it could be run considerably faster than just cranking on the starter, but at the expense of additional external complexity in the form of a computer controlled valve manifold to deliver correctly timed pulses of air, suitably matched the engine/crank layout.

I really need to find out what volumetric transgressions they're looking for, and that they're not just trying to count "Fairies on a Pin-head" because they think it's easy.

 

[TDIMeister]

Volumetric efficiency is absolutely, positively irrelevant to the method I suggested, which I am sure works, because it's not relying on the breathing of the engine through the valves, but rather on the measurement of explicit thermodynamic characteristics -- mass, pressure and volume -- of a known quantity of air that IS BEING SUPPLIED to engine.

 

[BigVlad]

TDIMeister: Does your suggestion make any sense at all? If air is injected into the plug hole at TDC on the firing stroke, won't the piston just descend until the exhaust valve opens and then stop? I don't think you can "motor" an engine like this. Maybe I don't understand what you mean.
SincoTC: I think the simple answer is that without disabling the valves etc. that it can't be done in any sraightforward way.

 

[patprimmer]

OK

I see what you mean, by collecting data of the induction you can determine by the air speed and pressure in the port when the piston is drawing down, when it is stopping and when it starts expelling air back into the intake port as the piston moves up with the valve still open. At constant relatively low speed the error due to elasticity of the air will self correcting as it is opposite at TDC and BDC.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[TDIMeister]

"If air is injected into the plug hole at TDC on the firing stroke, won't the piston just descend until the exhaust valve opens and then stop?"

An engine has inertia. At a sufficient RPM of the motored (driven) cylinder, and all the spark plugs pulled out, the engine will keep rotating past BDC.

Many industrial engines are started exactly by a pneumatic blast into the cylinder rather than electrical starters.

 

[thruthefence]

I've seen an aircraft engine having a poorly executed differential compression check, and let me assure you that an engine will motor just fine with air pressure applied to the spark plug hole.

And regarding the "complex" air valves, needed for a starter, not to worry; a simple manifold with an air tank & engine driven pump works just fine. An example is the Sukhoi Su-26. There is also a hand pump to service the tank, if needed.

For the old WWII era radials, a shotgun shell,( no pellets, of course ) was inserted into a breech, & fired in a similar manner.

 

[Lou Scannon]

The Napier Sabre shared this method, if I'm not mistaken.

 

[BigVlad]

I remain totally unconvinced. If "motor" means turn over continuously as long as air pressure is continuously applied to the spark plug hole - it just doesn't work.
Certainly if air is suddenly applied to the plug hole while the piston is just past TDC on the firing stroke it will spin over very smartly - but, as you would expect, when it comes to the compression stroke it stops and turns in the reverse direction until the piston is at BDC. With the extra pressure in the cylinder (from the airline) the piston will never make it past the compression stroke.
Big industrial and ship diesels etc. have a timed compressed air system to every cylinder for starting.
Some models of the Su-26 have an air starter system - that air was not going into a spark plug hole. It is very possible you could start an engine by kicking it over by air into a plug hole - but how would you get the sparkplug back in after the engine was running?
Cartridge starters were very common to start both piston and jet engines - but the pressure from the cartridge did not go into the engine's cylinder.
The cartridge may have looked like a blank shotgun shell but were purpose-made and usually far bigger than a shotgun shell.

 

[TDIMeister]

Why on earth is there a need to put the spark plug back on after the engine is running? Getting the initial cranking going can be done from the conventional starter motor, to make sure the engine spins in the correct direction and to give it some rotary momentum.

I haven't hashed out the fine details of realising this system, but to my knowledge there isn't something like this that you can go out and buy yet. I should be protecting this idea for a potential patent. If there are doubters even better for me; I trust the laws of thermodynamics are on my side.

 

[BigVlad]

Actually I was referring to thruthefence's interesting comments about starting an Su-26 and "motoring" an engine by putting air through the plug hole. I have no idea how your idea is supposed to work so I won't be trying to steal it.

 

[thruthefence]

You are comparing apples to oranges, but perhaps I wasn't clear enough in my post.

Initially, I referred to a "differential compression test", also known as a "leak down" test. This is an inspection procedure to determine relative health of the recip engine. It is not a starting procedure. It is performed as follows:

And for the purposes of this discussion, I will use an aircraft engine as an example.

1) The spark plugs are removed from the cylinders. ( on aircraft engines, only one plug per cylinder is removed )

2) The piston is brought to TDC, in the subject cylinder.

3) A regulated test Gage set, comprised of two 100 psig calibrated gages in series, separated by a fixed orifice, (.040", .25" in length) is plumbed into the open spark plug hole.

4) And here's important detail; The propeller is restrained from turning.

5) Compressed air is applied to the test rig, and the regulator is adjusted to 80 psig, and the leak rate of the subject cylinder is noted on the second gage ( downstream of the orifice ) thus the name, "differential" compression check

Now, If the piston is not right on TDC, or in the dwell range, the air pressure will drive the piston down it's stroke, turning the propeller with great force. ( I'll leave the calculations to you, assume a 5" bore, and a healthy cylinder, say 75 PSIG ) even through a .040" orifice.

If you read my post I referred to a "poorly executed" compression check, which I will expound on. The technician performing the test, either through poor training, lack of attentiveness, or whatever reason, plumbed the compressed air line directly into the spark plug opening. Thus a full 125 PSI of unregulated air, without the restrictive orifice, was directed to the top of a 5" piston that was just slightly past TDC. The engine motored rapidly until the air was removed, the Magneto impulse couplings firing away. Luckily, the other cylinder's plugs were removed, or it is possible, with a few more "ducks lined up" that the engine would have lit off. Also lucky that someone wasn't killed or seriously injured.

And regarding your assertion that the flywheel effect of a 60 lb propeller would not carry the piston back up the stroke, I leave those calculations to you. But remember, as the piston approaches BDC, the exhaust valve opens, relieving the compressed air somewhat & allowing the crank to coast through.

You are correct in your description of the "shotgun shell" starter system. I stand corrected.

 

[SincoTC]

thruthefence,

It sounds like that technician (or his assistants) had a very lucky escape as the prop would carry quite a lot of momentum and is not a very "person friendly" device!

However, I'm a little confused when you say that "the engine motored rapidly until the air was removed" and other references to simply blowing reciprocating engines over with a continuous supply of air. I'm with BigVlad in having doubts that it could get past the next compression stroke, as surely it would start out slightly "blown" and then have the incoming air as well as closed valves to overcome. However, as you've seen it and I haven't and a prop is a good flywheel especially without the other plugs in, so I don't doubt that it actually happened!

All the air starters and Cartridge starters that I have seen (Coffman/Breeze), are basically separate devices, similar in concept to an hydraulic swash-plate motor (or latterly, a vane type rotor), where a continuous high pressure medium is converted without valves, into rotary motion to drive the main engine's crankshaft, often through a conventional ring gear. This was considerably ligher and more powerful than an electric starter and their heavy batteries, and with the bonus that the air reservoir could be recharged very quickly, or even by hand pump if necessary. The Cartridge type were fed by a short steel tube from the breech chamber.

Large marine diesels often have compressed air starters where the gas is injected directly into the cylinders, but I think you'll find that every cylinder has it's own accumulator and cam operated valve to give a timed burst to each piston just after TDC on its firing stroke.

I've seen the thread about "Air Motoring" a week or two back, but believe that motor was was of a design that WOULD simply "blow over".

I'm not into gas dynamic theory, but I think TDIMeister should pursue his idea, if it will spin on a single cylinder with a timed pulse (probably with the other plugs removed), after giving it a nudge in the right direction with the starter, and if the P-V diagram will "average out" the affect of the opening valves and give a reasonably accurate indication of swept volume, then it's worth a shot.

I'm of a generation where I would rather try and get the cylinder vertical, disable the valves and burette some incompressible liquid in to measure it. Oh Sh*t; I forgot it's got Dykes pressure-backed rings and I've just filled the sump!! I knew I should have pulled the head and got the calipers out in the first place!

 

[Lou Scannon]

I see this as an opportunity for a materials engineer develop and apply a suitable viscoelastic materal and delivery system to force a metered amount of the material into the cylinder at TDC and BDC through the spark plug boss. Ideally the procedure would be staged: position the piston at TDC and meter the material into the combustion volume, then position the piston at BDC and meter the additional amount of material required to fill the swept volume. A required characteristic of the material is that it has sufficient tensile strength to enable the process of withdrawing the all of material back through the spark plug boss. A desired characteristic would be that the material dissolves in motor oil and does not present any appreciable hazard to the engine, when dissolved in the crankcase oil in limited concentrations. A nice feature to include in the delivery system would be measurement of the recovered material volume, as a guard against subsequent "hydraulicking" the cylinder due to accidental jettisoning of some of the material during the withdrawal process.

 

[BigVlad]

On the subject of air-motoring through the plug hole etc. I actually tried this on a bench-test engine I have. Not as dangerous as a propellor but still a good way to lose a few fingers if you are not careful. I am pretty sure with this engine (which has typical valve timing) it would never get past the next compression stroke. Starting just past TDC on the firing stroke the air pushes the piston down until about 50 degrees BBDC, when the exhaust opens and the pushing ceases. The flywheel momentum then coasts the piston through the upward exhaust stroke (the air coming in through the plug hole just escaping out the exhaust without any effect on the piston), through the following downward intake stroke and then into the start of the upward compression stroke. The intake valve closes at about 50 degrees ABDC so then the piston has to face the same force/pressure for the same amount of sroke length that originally got it moving - plus the compression forces. So the particular engine I've got will never get anywhere near TDC again. And in practice, when I try it, the engine behaves as described. If the engine had extremely late intake valve closing it may continue to "motor" - but not with normal valve timing.
I also wish TDIMeister well with his idea. Just because I don't understand it doesn't mean it doesn't work - it means I don't understand it.