What is the correct way to measure engine compression?

Pat Martone

Pat Martone

Senior Member
I have measured my engine compression 2 different ways and would like to know if either way is "correct":

- With key off, and a spark plug removed to insert compression meter into successive cylinders, while the other 3 still have spark plugs inserted; cranking the starter, I measured 70-72 PSI in each of the four cylinders.

- With intake and exhaust manifolds removed and all spark plugs removed; cranking the starter, I measured a range of 95-100 PSI across all four cylinders.

I do have a high compression older B-F head.

Is one of these two compression measuring techniques "correct" and if so, which one gave me the correct compression values.

Thank you.
 
I was taught years ago to remove all plugs, and crank the motor with the starter, and the throttle OPEN! This lets the air in the cylinder being tested, and with the other plugs out, it cranks easier. I do this with dry cylinders, then a second time after oiling the cylinders. If the compression rises considerably on the second try, the rings are worn.
 
Your second approach is closer to correct, but you do not need to remove the manifolds.

The approach 2manycars outlines is more normal. You want all the plugs out so that the engine will spin with the starter as quickly as possible.

The most important thing is that the pressures measured are close to each other. I record 3 numbers in a row for each cylinder, and compare them.

I also do the test on an engine that was recently running and more or less at operating temperature (not cold).
 
Was told to remove all plugs, set throttle wide open, take three tests each cylinder, and add a squirt of oil to cylinder to seem if compression comes up. Would imagine a cold engine might turn over easier. Also was told that the first compression reading after one turn should come up to half the max reading. So a 55 lb. reading should get about 25-28 lb. on first rotation.
 
I was taught to let the cylinder pump at least 10 strokes or until the guage no longer rises & with throttle wide open.
If a cylinder is low on first test, try again several times as sometimes carbon loosened when plugs are removed may get stuck under valve face. Bill W.
 
I am trying to understand what you are testing. When we test compression on aircraft engines, we measure the difference between the air pressure applied and that which the cylinder maintains. Just to look for leaks in the valves or rings and overall health of the cyclinders. Is there a difference?
 
rglaab said:
I am trying to understand what you are testing. When we test compression on aircraft engines, we measure the difference between the air pressure applied and that which the cylinder maintains. Just to look for leaks in the valves or rings and overall health of the cyclinders. Is there a difference?
Click to expand...

That's a leakdown test.

To be honest i can't immediately think what the difference is between that and a compression test, other that the procedures are different. They essentially seem to test for the same things.

A leakdown test is a very quiet affair as there's no cranking and you can listen for where the leaks are. Arguably therefore easier to pinpoint the problems?

Chris
 
BILL WILLIAMSON said:
I was taught to let the cylinder pump at least 10 strokes or until the guage no longer rises & with throttle wide open.
If a cylinder is low on first test, try again several times as sometimes carbon loosened when plugs are removed may get stuck under valve face. Bill W.
Click to expand...

I was told to prevent this you should break all plugs loose and start engine and then shut it off and remove the plugs. They claim it blows the excess carbon away and off the plug so it doesn't get jammed up in the threads.

I've personally always been too lazy to do this.....
 
A compression check on an aircraft engine and a typical auto engine are 2 different animals..
A compression test and a leak test are 2 different animals..
That said, as mentioned, a compression test is faster and easier for most folks to perform than a leak check.. A compression check should be done with a somewhat warm engine with all the sparklers removed.. Its usually best to blow air around the sparkler base or loosen and start the engine to blow away any 'crap' especially on an engine whether the sparklers are stuck in some dark deep hole..
I was taught to always take note of the first needle deflection and spin the engine until the needle stops climbing which should be within 7 or 8 strokes.. The readings should be within about 10% of one another and if there is any question then a little can be stuck in the hole and tried again which could give indication of ring condition.. You can perform your own leak test by making an adapter to stick a certain limited amount of air into the cylinder while locking the cylinder at TDC and then listening where the leak is..
 
I have always considered a cylinder leak down test as "the next step" after finding a cylinder(s) to have low compression. I regard it as the next diagnostic step to finding out why there may be a low cylinder. If compression is within 10% across the board, you're good. If there is low compression across the board, but it is even across all cylinders, chalk it up to worn rings. Otherwise if there are one or two out of specs, a leak down test will be conclusive as to what the cause is - burned valve(s), valve(s) needs adjustment, ring blowby, hole in a piston, etc.
Vacuum tests can be pretty conclusive, as well, as you notice the gauge fluctuating, low, etc while idling, snapping the throttle open and closed, steady throttle, and so on.
My Plymouth 218 flat 6 shop manual reommended setting the dizzy timing based upon max. vacuum at idle...interesting and simple concept.
 
You must log in or register to reply here.
Back
Top