23 replies to this topic

[mrman_3k]

Does anyone know if one can make a correlation between compression test results of an engine and the compression ratio?

Basically, can we say any engine with an 11.1:1 ratio would have 220psi as a result? Or are there more factors than just static compression ratio, i.e. cams, head design, volumes, etc?

If you could post the answer to this and any extra detailed information about factors especially diagrams, it would be greatly appreciated.

Thank you in advance

[rgagne]

The other info required is the pressure in the chamber at bottom dead center (assuming the valves close at that very moment, otherwise, you need both pressure and crank position).

RGagne


edited: by the way, 220 psi is very high... (or very low if we are speaking diesel engine)

[McGuire]

Excellent question.

The short answer: Among other factors, cranking compression is not so much a matter of static compression ratio but effective compression ratio: cylinder volume at IVC (intake valve closing) vs. cylinder volume at TDC.

There is more to it than that, but there is a useful empirical rule of thumb for diagnostic purposes in the field: cranking pressure should approximate 17 to 20 times the static compression ratio.

Couple things here, though: Naturally, for diagnostic purposes the gross numbers are not nearly as useful as variations among cylinders; compression tests are often improperly performed, and then the results are improperly interpreted; for many purposes a leakdown test may be more useful than a cranking compression test.

Do you want the long answer?

[mrman_3k]

I would very much like the long answer

Thank you guys

[BRIAN GLOVER]

Oh jees, here we go.;)

Quote

Originally posted by mrman_3k
I would very much like the long answer

Thank you guys

[McGuire]

Quote

Originally posted by BRIAN GLOVER
Oh jees, here we go.;)

Never mind then.

[desmo]

Quote

Originally posted by McGuire


Never mind then.

Oh come on

[Catalina Park]

Quote

Originally posted by desmo


Oh come on

Yeah, I am up for it!

[Kjetil]

First of all, I am not into motors. But I vill use common sense.
I find metric measures easier to use.
If you have a 11.1:1 ratio, you initially get 11.1 bar pressure if you have atmosferic pressure to begin with. (One bar is 14.5 psi)
However, as you compress the air, it gets hotter, so you get additional pressure from that. I'm sure there is formulas for this, but I don't heve them.
Some air will probably leak out, so this will reduce the pressure depending the state of the engine.

[McGuire]

This is an interesting subject because of all the four valve events (IVO, IVC, EVO, EVC) intake valve closing is undoubtedly the most critical and sensitive to engine performance. (IVO is actually one of the least sensitive, along with EVC.) Since the intake valve closes well after BDC on the compression stroke, the earlier you close the intake valve the higher the cranking compression will be. However, cranking compression will not tell you much about what the engine really wants. If you put a standard production camshaft in a high-compression, high-performance engine its cranking compression will increase but its performance will decrease dramatically.

[Paul Ranson]

Quote

Does anyone know if one can make a correlation between compression test results of an engine and the compression ratio?

I think the answer is 'no'.

Compression tests are very vague procedures, but even readings across the cylinders is a good indication that the basic engine is fit. If you need to know the compression ratio I think you will have to measure the stroke, bore and the volume of the combustion chamber.

Paul

[McGuire]

Quote

Originally posted by Paul Ranson

I think the answer is 'no'.

Compression tests are very vague procedures, but even readings across the cylinders is a good indication that the basic engine is fit. If you need to know the compression ratio I think you will have to measure the stroke, bore and the volume of the combustion chamber.
Paul

There is a fairly reliable rule of thumb for cranking compression in production and production-based engines: 17 to 20 times the static compression ratio (a figure included in most service manuals, while effective compression ratio is not).

The trouble (or nuisance) in the field is calculating the effective compression ratio, which is a function of static compression ratio, point of IVC, and rod/offset ratio (which determines the exact location of the piston at IVC.) For the typical diagnostic purposes it's not worth the trip, while the empirical rule of thumb (17 to 20x static compression ratio) is close enough for these purposes. Of course, what we are really looking for in a compression test in engine diagnosis is variation among cylinders, with 10% being the allowable maximum (some say 15%).

Much of the "vagueness" in compression testing is due to the sloppiness and/or ignorance of the tester. Perfomed properly it's a reliable test, with one limitation: identifying the degree of deterioration in an engine that has experienced more or less uniform wear, especially in cases where a moderate degree of wear is still significant in performance (race cars for example). An engine can easily be half worn-out and show relatively uniform cranking compression on all cylinders, and decent pressure across the board as well. There lies the value of the cylinder leakdown test over the traditional compression test.

The major mistakes made in compression testing:

1. Improper/inadequate equipment: the "universal" gauge with hold-in-place rubber boot is useless. Throw it away, or use it as the basis to fabricate a more proper gauge with hoses and adapters.

2. Test must always be performed at 100% throttle opening. (Amazing how often this is overlooked.)

3. Cranking speed must be uniform as each cylinder is tested, and each cylinder must recieve exactly the same number of compression strokes. (For experienced mechanics, how quickly the gauge comes up to maximum pressure is itself a valuable tool in diagnosis, along with using one's eyes and ears to detect where the air is leaking.)

4. The engine must be at approximately the same temperature throughout the test. In fact with some engines (air-cooled especially) performing the test both "warm" and "cold" can be useful...in much the same way as with "wet" and "dry" figures.

Another overlooked point: in engines with relatively small cylinders, like motorcycles etc., the compression gauge (with hose etc) can influence the total cranking compression figure as it adds to the total volume of the chamber at TDC eh. So while here the compression test will give you an accurate picture of variations among cylinders, the cranking pressure figure itself may be misleadingly low.

[Paul Ranson]

Quote

There is a fairly reliable rule of thumb for cranking compression in production and production-based engines: 17 to 20 times the static compression ratio (a figure included in most service manuals, while effective compression ratio is not).

I always reckoned that if you saw at least 160lb/in2 the engine was in the ballpark, which tallies with your rule of thumb. But leakdown will be better for determining, diagnosing and especially monitoring engine condition. I think I've compression tested a serious production based race engine, but I've forgotten the figures....

Anyway the answer to the OP's question is still 'no', for at least some of the reasons you gave up thread.

Paul

[WPT]

From thermodynamics for a polytropic proccess; P*V^k=a constant , where k is equal to 1 for an isothermo proccess and 1.4 for an isentropic (no heat transfer) proccess. As McGuire says, intake valve closing has a big effect on the trapped charge, and the temperature of the engine during the test will affect the heat transfer.

P1*V1^k=P2*V1^k

T2/T1=(V1/V2)^k-1

WPT

[WPT]

Made a mistake, the pv equation should read; P1*V 1^ k=P 2*V 2^ k Sorry, WPT

[McGuire]

Quote

Originally posted by Paul Ranson
Anyway the answer to the OP's question is still 'no', for at least some of the reasons you gave up thread.

Paul

If you say so.

However, if you perform compression tests on a few hundred engines, you will find there is a reliable empirical relation between static compression ratio and cranking compression: 17x to 20x. This correlation is reliable and useful because it covers the range of intake valve timing normally encountered.

The trouble with picking an arbitrary number like 160 PSI to fit all engines: an engine with conventional valve timing and a perfectly ordinary static compression ratio of 7.5:1, in perfect condition, can easily test out at 128 PSI. Oops. However, if an engine with 12.5:1 static CR tests at 160 PSI, either it has very radical intake valve timing indeed or it's hurting inside. So while cranking compression is more directly related to effective CR than to static CR, the latter is still quite useful info when performing a compression test...much more so than simply picking an arbitrary over/under number, for sure.

Electronic engine analyzers employ a very interesting dynamic method for testing compression. At a predetermined speed (say 1500 rpm) each cylinder's ignition is shorted out one at a time and the added load on the remaining cylinders is recorded. If one cylinder is not contributing its fair share of power but still takes the normal amount of pumping work from the engine, its compression is regarded as within normal range (and so it goes in all proportions thereof).

But perhaps even more impressive is the "cylinder misfire" value now used in the onboard diagnostics systems of production cars. The crankshaft position sensor is used to record the crankshaft's velocity (6x per rotation for a six cylinder, for example) and if the ECU sees the crankshaft slowing down slightly at the same spot more than a few times in a row, it registers that cylinder as misfiring and records the event(s). So while it may not tell you what is wrong with a cylinder, it will at least tell you which cylinder is causing the trouble, which with an intermittent or emerging problem is half the battle. These kids in the shop today have things too easy if you ask me.

[Paul Ranson]

I've never come across an engine with as low a CR as 7.5...

Paul

[Greg Locock]

turbos, two strokes, lawnmowers. I think you have, you just didn't know!

[McGuire]

Quote

Originally posted by Paul Ranson
I've never come across an engine with as low a CR as 7.5...

Paul

I can think of a few... the VW Beetle (most were less, actually; I think the 1200 40 hp version was 6.8 or something) and loads of Porsches including the 356B, the 90, the 924 and 968 turbos, some single-turbo 962's...the Volvo 240, Cosworth DFX, the Pontiac Trans-Am 301 V8 turbo, the Callaway Sledgehammer, the original Lotus Esprit Turbo with the Dellorto carbs...I can probably think of more but I am getting too damned old. Many of the piston engines in general aviation, the Lycomings and TCM's...

[Paul Ranson]

Quote

turbos, two strokes, lawnmowers. I think you have, you just didn't know!

OK. I've never come across an engine I needed to compression test with such a low CR....

Would a DFX really be that low on methanol?

Paul

[McGuire]

Quote

Originally posted by Paul Ranson
OK. I've never come across an engine I needed to compression test with such a low CR....

Would a DFX really be that low on methanol?

Paul

Yes, for the first few years the Cosworth ran at 80" MAP. For 1979 the 50" popoff valve rule went into effect (which everyone cheated, but that's another story).

[desmo]

Calculating 2 stroke CRs gets complicated.

[McGuire]

Quote

Originally posted by Paul Ranson
I've never come across an engine with as low a CR as 7.5...

Paul

Historically, compression ratio has trended in interesting ways. From 1927 to 1934, the CR of the Cadillac V8 (a rather conventional sidevalve design) rose from 4.25:1 to 6.25:1, with the only major change being the increased availability of what was called at the time "high-octane" fuel...about 66 octane for "Ethyl," vs. 56 for "low test." Depending how you look at it, a piston engine with a CR of only 4:1 is a rather different machine than the one we know today. Here, when the piston is halfway down the bore on the power stroke, the combustion charge has expanded only 2x, eh.

Of course "Ethyl" was the marketing name for TEL, tetraethyl lead, which was discovered as an octane enhancer by Thomas Midgely and Charles Kettering of GM Research in 1921. Since lead was a well-known environmental toxin even then (Midgely himself nearly died from exposure to the lethal compound; in the first plant built to produce TEL in volume, five men died of lead poisoning in the first two months) it took the combined lobbying and marketing efforts of General Motors, Standard Oil and DuPont to put that one across on the public, and even then it took a number of years for TEL to attain wide acceptance and use. This is one of the more interesting tales in the annals of American capitalism, which folks may find illuminating.

Just before WWII the refiners learned, through regenerative catalytic cracking, how to produce high-quality, high-octane gasoline in mass quantities...this advance was largely the work of one man, Eugene Houdry, one of the true unsung heroes of both the auto industry and the allied war effort. (High-octane aviation fuel was key to allied air superiority.) After the war this production capability propelled the trend away from L-heads to the modern high-compression, overhead-valve V8. The 1949 Cadillac V8 came straight out of the box with a 7.5:1 compression ratio, a rather startling development for the time. Through the 1950's CR continued to climb, and by the 1960's musclecar era there were a few US production engines with 12.5:1 CR's, but emissions concerns put a stop to all that business.

At that point, the same company that had invented, developed, and popularized TEL for production cars then led the charge to eliminate it, when GM formalized its plans to phase in the catalytic converter, which made leaded fuels essentially obsolete: the platinum catalyst could not tolerate any amount of TEL in the exhaust and remain effective. But as there were no real substitutes for TEL available at that time, compression ratios would have to come back down.

All through the 1970's and 1980's in the USA, static compression ratio on production cars hovered in the 7.5 to 8.5 range due to emissions requirements, mainly the switch to unleaded fuels. With better chamber designs and electronic control systems (notably, piezo knock sensors) and improved unleaded fuels, they have since climbed back up to 10:1 and above.

[Christiaan]

I did a project on wind tunnels and their instrumentation in 2002. I am looking for an electronic copy of it on my old disks. I circulated a summarized literature review here a couple of years ago but can't seem to locate it. If anybody still has it pliz mail to me