compression test question

[Musick17]

What kinds of compression tests are there? I've seen the manual ones where you put a gauge where the spark plug goes and you turn the motor over and it gives you a reading.

The reason I ask is because the people that are working on my truck said they did a compression test where they pressurized the cylinders and left them over night and 3 of the 4 had lost all their pressure by morning. To me, this makes no sense. Wouldnt it be impossible to for a cylinder to hold pressure over night like that and what the heck kind of test would you call that?

 

[alwaysFlOoReD]

RUN away! These people are worse than thieves, at least you can get insurance against thieves.
There is a leakdown test which puts air into the cylinder but it measures leakage right at the moment, not over night.

Richard

 

[JC 97]

+1, I think they are not very good at what they are doing .What kind of problem are you having with your truck?

 

[kunar]

i agree, sounds like a bunch of scammers. a leak down test needs the cylinder at TDC, which means you have to do them one at a time. tell them they are a bunch of liars and crooks, take your truck elsewhere.

were they doing this compression test because you requested it or do they claim its complementary (but its really just a way to try to convince you to get more work done)

 

[Musick17]

So here is how things went down....

Truck started running a little hot on the interstate, nothing critical. Just a needle or two above where the gauge normally sits. Then a couple days later I notice it running even hotter, not even near red on the gauge, so I pulled into a gas station and popped the hood and then something went and it started dumping coolant everywhere. Took it to my regular mechanic and they found a broken idle air control valve.
I kept having overheating issues and my mechanic determined it to be blown head gasket/warped head.

I took it to a family members mechanic back around my home town because the labor rate was much cheaper. They had the head machined and replaced the head gasket and valve seats. While they were working on it I said go ahead and put a new water pump and timing belt on it. After they got that done they informed me it was still overheating and that the fan clutch had gone out, which would have potentially caused the overheating and warping in the first place. After they replaced that it was still having issues so they unhooked the exhaust and said it was running alot better so they wanted to try replacing the cats... and of course that wasnt it. So this is where they did some mystery test on it and determined it was piston rings.

Since I have gone through a family member I understand wanting to be a little extra careful in case something were to happen short after and it would fall back on them. But we are going into week 6 of them having it, and I'm tired of the feet dragging and guessing.

 

[doorgunner]

RESCUE YOUR TRUCK.....! Test each cylinder with a compression tester that has a screw-in connector on the end of the pressure hose....Write down the reading of each cylinder's compression...let us know the results.

 

[RonD]

I agree with all the above, they are either ignorant or dishonest, pick one, there is no 3rd option

The leak down test is done by setting a specific pressure on the incoming line, like 100psi.
Then you put that steady pressure into a cylinder that's at TDC, with a gauge attached.
If the gauge shows 95psi, then you have 5% leakage.
You could use 80psi as the set pressure and do the math, I used 100psi as an example because it's easy to do the math, lol.

Then you put a little oil in the cylinder and turn the cylinder over a few times and retest.
If it now shows 98psi on the gauge, then you have 3% leakage via the rings.

At NO TIME would you test for "how long" a cylinder holds pressure, it's not designed to hold pressure, its designed to contain high pressure for a short time, like when a cylinder fires, a firing cylinder is at 1,000psi to 1,500psi, diesels can be above 2,000psi

A "Compression test" is when you test all the cylinders for similar compression.
All spark plugs are removed first.
Throttle plate is propped open.
Fuel is turn off, pulling the EEC fuse usually works.
Then compression gauge is put into one cylinder at a time, and engine is cranked with the starter motor so you hear 4 or 5 "compression strokes", they are easy to hear.
Write down the psi number and do the next cylinder.
What you want to see is that all cylinders are within 10% of each other.
So if highest cylinder is 160psi and lowest is 145, then it's OK, 10% of 160 is 16psi, and 10% of 145 is 14.5psi, so close enough.....for government work, lol.

Some also do the test "dry" then "wet" by adding a bit of oil, and retesting the cylinder, this is also a way to see if rings are starting to go on an older engine.

Some prefer warm engine compression tests, others cold, I think it's a moot point.
What isn't moot is if one cylinder is way lower than the others, that means further investigation is in order.
If compression is getting down below 120psi on any cylinders then ignition of fuel/air would be iffy.


Check your heater core or by-pass, my '94 4.0l started running warmer and the core was partially clogged, go figure, $20 core and all was well.
Still blew warm air too, but blew really hot air after the change.

 

[martin]

There is such a thing as a cooling system pressure test which involves putting a gauge and valve where the rad cap is and applying pressure to see if the system is leaking anywhere. That's the only test I can think of that would involve a time factor, but not usually more than half an hour. that does not sound like what they were talking about however. Don't forget that a bad rad cap or blocked heater core can play a big part in overheating problems.

 

[Surrey]

Some prefer warm engine compression tests, others cold, I think it's a moot point.

A warm engine means you'll get better seals on your valves and stuff, so you'll get a more accurate reading of conditions when actually testing it I would think... So its not a moot point, but how much effect it actually has I couldn't really say.

 

[RonD]

A Compression test has a few factors that can effect it, battery and starter condition effecting crank speed, is the main one, as is the leak down with time which is why the speed of the crank shaft is important.
i.e. manually turning the crank for a compression test would produce lower numbers than turning crank with the starter motor.

Since the compression test numbers are really only to see if one or more cylinders are much lower than the others, the warm test is, IMO, moot, that is all I meant.

If all the cylinders are coming in between 160-170psi on a cold engine, and then you warm it up and get 170-180psi, have you learned any more about the engine with the warm test?

 

[adsm08]

I am aware of 3 valid kinds of compression tests.

Wet, dry, and relative.

Dry is done before wet. You put the physical gauge in the spark plug hole and spin the engine around. You get a reading.

Wet is the same, but you put some oil down the cylinder first.

Relative is a different beast altogether. A relative compression test is done with a scanner. The computer watches the speed of the crank sensor and catches minor variations in the speed of the crankshaft as it spins. It isn't a true compression test in the same sense as the others in that you don't get a reading in PSI that will tell you if there is a problem or not. The results are read in percent lost, which is a comparison of the cylinders to each other. That means if you have no compression in any of the cylinders a relative test will show as all good. It isn't meant to show a worn out engine, it is meant to show if a single cylinder has a problem that warrants tear down for further investigation.


What he tried to sell you on sounds more like a leak down test, which is meant to test valve sealing by running regulated compressed air into the cylinder and seeing how much of that pressure it can hold. The difference is that this test is meant to be done over minutes, not hours, and the air stays hooked up the whole time. IIRC (been a while) 20% loss is the max acceptable, less than 10 is the goal.