• Welcome to the Contour Enthusiasts Group, the best resource for the Ford Contour and Mercury Mystique.

    You can register to join the community.

compression testing

BrApple

No Life But CEG
Moderator
Joined
Dec 23, 2001
Messages
21,731
Location
Gales Ferry, CT
so I picked up a compresison tester and a leak down tester ... both instructions say to run the tests when the engine is at operating temp.

Now I thought it was bad to be removing and installing spark plugs in aluminum heads when they are hot. I have always waited untill the engine is completely cold to do anything with the plugs on these cars.

so should I be doing this testing at operating temp or cold?
 
I had read all of that about doing it on a hot motor as well, you 'should' be okay as long as you are careful removing your plugs and installing the compression tester.

I have tested mine both at operating temp and cold engine.

It was much much easier on the cold motor, no burnt hands and what not.

I wouldn't think that any possible diagnosis would change. The idea is to find (or hopefully not find) differences amongst cylinders. I know the motor temp could casue a different compression result but that difference would be the same for all cylinders. You would still be able to see which cylinder(s) are off at either temp.

One idea if you really want to do the test on a hot motor. Remove plugs first....liberally apply anti seize to the treads and re-install. Then run the motor to operating temp and proceed with the test.

Even if you already have anti seize on the threads doing it again right beforehand should make removal really easy.

Note: This is a new idea i have not yet tried it.

Be sure to post your results. I would think as long as they are all within ~10 psi on compression your good. Never performed a leak down test...isn't that more to test the condition of the valves? Off to the interweb for more research....

FWIW - These were my compression tests results on my 04 3L. As viewed from the front of the car..

213 215 215
215 213 214


Internet Search ...


A leak down or "cylinder leakage" test is similar to a compression test in that it tells you how well your engine’s cylinders are sealing. But instead of measuring pressure, it measures pressure loss.

A leak down test requires the removal of all the spark plugs. The crankshaft is then turned so that each piston is at top dead center (both valves closed) when each cylinder is tested. Most people start with cylinder number one and follow the engine’s firing order.

A threaded coupling attached to a leakage gauge is screwed into a spark plug hole. Compressed air (80 to 90 psi) is then fed into the cylinder.

An engine in great condition should generally show only 5 to 10% leakage. An engine that’s still in pretty good condition may show up to 20% leakage. But more than 30% leakage indicates trouble.

The neat thing about a leakage test (as opposed to a compression test) is that it’s faster and easier to figure out where the pressure is going. If you hear air coming out of the tailpipe, it indicates a leaky exhaust valve. Air coming out of the throttle body or carburetor would point to a leaky intake valve. Air coming out of the breather vent or PCV valve fitting would tell you the rings and/or cylinders are worn.

A leakage test can also be used in conjunction with a compression test to diagnose other kinds of problems.

A cylinder that has poor compression, but minimal leakage, usually has a valvetrain problem such as a worn cam lobe, broken valve spring, collapsed lifter, bent push rod, etc.

If all the cylinders have low compression, but show minimal leakage, the most likely cause is incorrect valve timing. The timing belt or chain may be off a notch or two.

If compression is good and leakage is minimal, but a cylinder is misfiring or shows up weak in a power balance test, it indicates a fuel delivery (bad injector) or ignition problem (fouled spark plug or bad plug wire).
 
Last edited:
You should do both tests on a warm engine for the best "real world", and most accurate results.

Removing and reinstalling plugs in warm heads is not a concern if they are torqued properly. Like SHOgofast said, anti-seize is helpful here, though I personally am deliberately NOT liberal with it's application, and I only apply it on the initial install of new plugs.

An engine that’s still in pretty good condition may show up to 20% leakage. But more than 30% leakage indicates trouble.
This is a bit archaic. Anything more than 10% is trouble nowadays. More than 10% leakage will cause driveability issues in an OBDII car. And if it gets to 20% leakage - you'll have at least an obvious misfire.

Air coming out of the breather vent or PCV valve fitting would tell you the rings and/or cylinders are worn.
I've never understood why articles like this never seem mention that a washed cyilnder will allow compression to leak passed the rings also. If you've got a dead cylinder that's not burning the fuel injected into it due to say, a lack of spark from a bad plug, a leak-down on that cylinder can potentially show compression loss passed the rings. This can easily lead to an incorrect diag, especially if the plug isn't obviously bad.
 
FWIW - These were my compression tests results on my 04 3L. As viewed from the front of the car..

213 215 215
215 213 214
04 3L also

as read from front of car

170 160 170
185 155 165

done on a cold engine that has been sitting a few days.

looks like I should go back and do it on a warm engine.
 
i cant tell you how many compression and cylinder leakdown tests ive done on cars. the differences are so minimal that i dont even bother running the engine before hand. if you have a problem, it will be obvious on the gauge readout
 
ok I re-ran the test. this time I cranked the engine untill the needle on the gauge stopped moving

looking from the front

190 195 200
200 190 200
 
Thats WAY BETTER Brian. I was a bit worried about those first results.

Did the instructions say to only go so many revolutions?
 
you should only let it crank 4-5 times
Most of what you read on the topic will agree with this. However, these recommendations aren't typically written by the people who DO the work and diagnosis. Sometimes, you have to use a little common sense... like Brian did by cranking to a max reading for his second test. Varialbes like battery condition and piston position at the start of the test for a given cylinder, mean the decades-old "5 cycles" compression test recommendation will often yield inaccurate results, (as witnessed by the readings from his first test). His original numbers were still within spec, (all cylinders within %25 of each other), though they might lead you to looking for a problem when there really isn't one.

Personally, I like to add a dynamic test to the 5-cycle & max static tests, to get even more of the picture of what's going on in an engine.
 
Back
Top