How to - compression test?

[Trstno1]

Hey guys-

I was searching but could not find a thread on this ins and out of conducting a compression test on the motor in our cars. What needs to be done to power down the fuel pump so I don't flood the motor? Pull the rpm relay? How long should you crank before reading the pressure gauge? And, what is the acceptable PSI compression reading per cylinder? Am I forgetting anything else?

I am just about ready to start going through the fuel system but wouldd like to make sure I don't have another problem with the motor before I spend the money.

[jangell]

I did a compression test on my 3.0L engine (http://www.tmproductions.com/repairs-and-maintenance-blog/2014/11/15/engine-compression-test) outside of the car, so I didn't have to worry about the fuel pump. Pulling the RPM relay will disable the fuel pump, if you're concerned about flooding.

This was the first time I'd compression tested an engine. I record at first puff (i.e.: one crank), and at three cranks (the reading that really matters), since that's what seemed to be recommended when I did some Google searches. I did it with dry; I didn't do the wet test where you put a drop of oil in the top of the cylinder.

I'm fuzzier on the exact PSI you want, and failed to find any information for the 3.0L engine. Googling just suggested that anything less than 100 PSI after three cranks is Very Bad. Mostly it seemed important that they all read about the same PSI relative to each other. Depending on the source you use, that should be less than 10% to 20%. My worst-case was a 12% difference, which I considered acceptable for my application. I did show it to a local mechanic to be sure, and he wasn't worried about what my results showed.

-- Joe

[David T]

I did a compression test on my 3.0L engine (http://www.tmproductions.com/repairs-and-maintenance-blog/2014/11/15/engine-compression-test) outside of the car, so I didn't have to worry about the fuel pump. Pulling the RPM relay will disable the fuel pump, if you're concerned about flooding.

This was the first time I'd compression tested an engine. I record at first puff (i.e.: one crank), and at three cranks (the reading that really matters), since that's what seemed to be recommended when I did some Google searches. I did it with dry; I didn't do the wet test where you put a drop of oil in the top of the cylinder.

I'm fuzzier on the exact PSI you want, and failed to find any information for the 3.0L engine. Googling just suggested that anything less than 100 PSI after three cranks is Very Bad. Mostly it seemed important that they all read about the same PSI relative to each other. Depending on the source you use, that should be less than 10% to 20%. My worst-case was a 12% difference, which I considered acceptable for my application. I did show it to a local mechanic to be sure, and he wasn't worried about what my results showed.

-- Joe

You usually do about 3-4 "bumps" or *puffs*. Not terribly important but it is important you do each cylinder the same as all of the others. Shut the fuel pump off by lifting the plunger on the inertia switch or unplugging the RPM relay or pulling the #7 fuse. Block the throttle wide open. The test is supposed to be done on a warm to hot motor. Remove all of the plugs. Do each cylinder and record the results. The absolute # is not as important as how all of the cylinders compare to each other. On a new motor you should see around 150 psi but I would rather see them all within 5% of each other at 100 psi than some at 150 and others more than 10% different. The "wet" test is to see if there is a big difference and that would be due to worn/leaky rings. When you interpret the results you need to consider the motor's history. If it ever had a valve job or adjustment, if it uses oil, how many miles on the car, if the oil was changed regularly, etc. Good to do to record a baseline measurement so in the future you have something to compare to. On my motor at 3,000 miles I had a high of 145 and a low of 130 cold. Higher equals more power and less oil consumption. Lower equals more wear, lower gas mileage, higher oil consumption, harder to pass emissions testing, and lower quality idle. Large differences (over 25%) can mean burnt or out-of-adjustment valves, cracked rings, bad head gasket. Usually a motor headed for major repair.

[DMCMW Dave]

If you see a cylinder with low compression, the next thing to do is a leakdown test. This is where you pressurize each cylinder with compressed air and see how long it takes to bleed down, AND where the air goes (i.e. out a bad valve into the intake or exhaust, or out past rings into the crankcase). The compression test by itself does not tell you any of these.

[Trstno1]

Ok, pull rpm relay or fuse 7 to shut the fuel off. Do I need to disable the ignition system? If so, what do I need to do for that? And you said remove all spark plugs from the motor. Do I just remove all spark plug wires to get around the ignition system, then remove 1 spark plug at a time as I test each cylinder? Also, could I just settle on a test crank time, say 4 seconds? Then what is a good & bad cylinder pressure for our engines?

[jangell]

Dave and David know more than I do, but I think you want the exact number of cranks. It's very obvious when you're looking at the pressure gauge, as it jumps up each time the cylinder hits the compression stroke. I admit that you'll need a long hose on your pressure gauge, or a way to crank the engine from the engine compartment to be able to monitor the gauge.

I also think the reason you do 3 or 4 cranks is because after that it doesn't really change a lot, so there's no need to crank further.

I don't think you have to disable the ignition system -- there won't be any spark because you don't have any spark plugs installed, or at least have the wires removed from the plugs. My notes don't say if I removed all the plugs or only the one I was testing, and it was long enough ago that I can't remember anymore. I'm reasonably sure that I removed them all, though.

I did not do a leak down test on my relatively low cylinder, mostly because I didn't think to do it. It's one of those things I'm aware of, but have no idea how to go about it.

-- Joe

[DMCMW Dave]

I don't think you have to disable the ignition system -- there won't be any spark because you don't have any spark plugs installed, or at least have the wires removed from the plugs.
-- Joe

I don't likes sparks all over the place, so I'd still pull small wires off the coil just to make sure it's dead. Or just pull fuse #1.

[Bitsyncmaster]

You pull all the plugs before testing so you get the same cranking RPM when your testing each cylinder. Besides its much easier on your starter and cranks faster.

[Trstno1]

Ok, so what's normal pressure for the cylinders in our cars? After testing mine, cylinder #1 looks lower than the others. I got the following results:


#1 - 120
#2 - 135
#3 - 130
#4 - 145
#5 - 140
#6 - 130

[DMCMW Dave]

Ok, so what's normal pressure for the cylinders in our cars? After testing mine, cylinder #1 looks lower than the others. I got the following results:


#1 - 120
#2 - 135
#3 - 130
#4 - 145
#5 - 140
#6 - 130

Not perfect, but not bad enough to warrant a major repair. A cracked piston/ring will typically read more like 40, stuck valve anywhere from zero to 50. A really good engine will read 140-150 all the way across.

Just for fun, squirt some motor oil into each cylinder and repeat the test. This tends to seal things up a bit and you may get better readings.

[Bitsyncmaster]

Not perfect, but not bad enough to warrant a major repair. A cracked piston/ring will typically read more like 40, stuck valve anywhere from zero to 50. A really good engine will read 140-150 all the way across.

Just for fun, squirt some motor oil into each cylinder and repeat the test. This tends to seal things up a bit and you may get better readings.

I think a lot depends on how long the engine has gone without running as to how much oil is left on the cylinder walls. So defiantly try the oil squirt test. If you can, squirt the oil to the top side of the motor (piston) so it can coat the whole ring.