Can Anyone Tell Me if I Did My Compression Test Correctly?

Hey, guys.

Just did my first compression test and I hope I got it right, because if so, it may mean that I didn’t fork over 1500 bucks for a 1992 Ranger Custom with a junk 4.0 engine. I performed this test because I’m having a terrible problem with vibration out on the road and when I pulled off one of the wires on the coil pack (corresponding to plug #5, located on the driver’s side), I noticed that it made no discernible change in the idle, while the others all did. I took the spark plug out of cylinder #5 and switched it with one of the others, but there was no change.

Following the instructions that I found elsewhere on the site, I disconnected the coil pack, removed all the wires, pulled the fuel pump relay fuse and started taking out the plugs. Unfortunately, the ones on the passenger’s side are a b@tch to get to and the one in the back on that side is almost impossible (due to the AC box). So I ended up leaving the plugs in on the passenger’s side and just removed the ones on the driver’s side (since that’s where the aforementioned plug is located).

With all 3 of the plugs out on the driver’s side, I screwed in the test gauge and cranked the engine over a couple of times on cylinder #5 and got about 165 lbs of compression. I then re-read the instructions and cranked it over for about 10 seconds (which seemed excessive, but it’s what the instructions called for). This time it read around 180 lbs and so did the other two. I was thrilled.

So I’m just wondering if having the passenger’s side plugs still in when I did the test could have somehow thrown the test off? Also, the instructions called for the engine to be run for a while, then shut off for 10 minutes before the test. I drove the truck around on the road a while before conducting the test, but it was cooling down for more like a half hour or so before I actually got to cranking her over and reading the gauge. Just wondering if either of these things could have led me to have a higher compression reading than I otherwise would have. Also wondering if cranking her over for close to 10 seconds could have led to a false high compression reading. I don’t think so, but would love to hear it from someone else.

Thanks so much and sorry for the long thread.

The reason you pull all the spark plugs is to get good crank speed for the test, no compression strokes to slow it down.
You can pull fuel pump fuse or fuel pump relay if using a Remote starter button, if using the key to turn engine over just press down gas pedal all the way before cranking engine, that shuts off fuel injectors, also opens throttle plate for better air flow into the engine.

You don't need to warm up an engine to do compression test, you get slightly higher numbers if you do, but the point of compression test is the comparison of all cylinders to each other, so really doesn't matter if engine is cold or warm.
I prefer cold, much much easier to work on cold engine.

Since you didn't have all spark plugs out you couldn't hear compression stoke of cylinder being tested, usually you want 5 to 6 "hits" of compression stoke to get a good reading.
Starter motors turn engine's crankshaft at approx. 200RPM, 200 revolutions per Minute
A 4-stroke engine needs 2 revolutions of crank for all cylinders to complete 1 cycle.
So there are 100 complete cycles per Minute, assuming 200rpm crank speed.
That's 1.6 complete cycles per Second(this applies to 1cyl, 2cyl, 4cyl, V6, V8 or V10, lol), its crank RPM so nothing to do with number of cylinders.

So 5 seconds should be 8 complete cycles, but again this is based on good battery, good starter motor and 200rpm crank speed, so 10 seconds is long but will not hurt anything.

4.0l OHV engine has compression ratio of 9.0:1
Cold engine multiplier is 18, 18 X 9.0 = 162, so expected PSI would be 162

4.0l SOHC engine uses 9.7:1 ratio, 18 X 9.7 = 175

For warm engine use 19 as multiplier

Compression gauges are not calibrated, but as long as you use the same gauge for all cylinders then numbers are valid

Your 162 would be fine as long as all 3 were 152 to 172
180 is not "better" its just a number, and as long as all 3 were 170 to 190 then test shows no #5 issue
If one cylinder was 162 and the other 2 were 180 then that would indicate a possible problem.

Hey thanks, RonD. Sounds like a lot of good information in there. There are a few things that I’m confused/curious about though. Could you maybe help me understand the following?

1. Does leaving 3 of the spark plugs in (as I had to do) invalidate the results of the test on the other 3 that were done? In other words, could I have gotten false high compression readings because it somehow raised pressure in those cylinders that I did take the plugs out of and test?

2. Why does nailing the pedal to the floor shut off the injectors? Does the fact that the truck isn’t running somehow tell the pump to behave differently than if you were bombing down the road and floored it?

3. Does cranking the engine longer (say, for 10 seconds instead of 5) somehow lead to higher compression numbers on the gauge? I would think that an individual cylinder would “top out” on how much pressure it could generate after a few seconds, but then again, it did seem like the reading in that first cylinder rose from 165 the first time to 180 when I cranked it longer on the second try. Don’t really understand why this would be, but if so, then I’ll need to be pretty precise about just how long I crank it....

Thanks so much and sorry for all the follow-up questions. Much appreciated!

1) It is best to have all plugs removed, but consistency is the real key. As Ron said you are trying to compare the cylinders to each other, rather than a set value. So realistically it doesn't matter much if you remove one plug, test that cylinder, replace it's plug, and move on, take all of them out and leave them out, or take out just the half of the engine you want to test, as long as you aren't doing something like test cylinder one with all plugs in, test cylinder two with plug 1 missing, test cylinder 3 with plugs 1 and 2 missing, etc. As long as you were consistent.

2)There is a strategy in the computer called "clear flood". It is engaged when the throttle position sensor reads fully open during cranking (or when the engine RPM is low enough). It shuts off the injectors to allow a flooded engine (one with excess fuel in the cylinders) to be cranked to clear it out without adding more fuel. During a compression test you also want the throttle wide open to get the most air flowing.

3) You pretty much hit it on the head. You want to hit several compression strokes to allow the engine to get going as fast as it can with the starter, which still isn't even idle speed, but it will top out at a maximum. I have heard several theories about the best way to do it, crank until the needle stops going up, crank for 5-6 seconds, crank for 4 or 5 compression hits. Ultimately I have never seen drastically different results with any of these various methods. I usually see the needle stop going up in the first 5 to 6 seconds anyway.

Your difference may have been because enough oil got circulated to make the second test "wet".

Part of the oil's job is to help the rings seal to the block. So when you first do a compression test it is considered "dry", as in no or little oil on the rings. If you come up with low numbers on one or more cylinders the next test is to put some oil in through the spark plug hole and retest it, this is called a "wet" test. If the numbers come up a lot then the rings are the problem.

Hey adsm08, thank you so very much for all of the good info!! You really helped me understand things a lot better. The “wet” test was included in the second part of the instructions I had, so I guess I’d better refer to that as well. I think it sounds as though I can tentatively trust that my number #5 cylinder isn’t dead, but I’ll try to do a more accurate test during the day when the weather’s a little better. The next thing I’ll try is testing the coil pack with a volt/ohm meter. It does arc on that tower, but it also looks old as the hills (possibly the original, since it’s Motorcraft and was badly corroded where it mates to the motor).

On a side note, given what we know about the oil helping to seal the rings, wouldn’t it be more accurate to do a compression test on a cold engine? And wouldn’t the first cylinder you test also be more accurate than the second, third, etc since more oil has begun circulating at that point? I know the answer must be “no” (otherwise compression tests wouldn’t be worth much), but maybe you could explain why to me?

Thanks again, man. I really appreciate your time!!

The rings don't get a ton of oil when the engine isn't running, or even when it is. They get all their oil by wiping down what gets splashed up onto the cylinder walls by the crank weights hitting the oil in the pan.

The amount you are supposed to add to a cylinder for a wet test is a table spoon or two, significantly more than is ever there during normal operation. And you are adding it to the side that is usually dry.

I think the idea that I suggested about doing enough cranking to move that much oil up to the rings is a long shot, but not technically impossible. It would account for the jump, but your initial numbers seem high enough to indicate that there is nothing to really by worried about from compression.

If the miss is bad enough that you can find it pulling plug wires the other thing you can do (aside from my timing light test from the other thread) is swap wires 1 and 5 at the coil, and then see if the misfire has moved to cylinder 1. If it has then your issue is 100% the coil.

Thanks again for all the good info, adsm08. The miss can’t really be heard at idle (at least not by me), but I found that by pulling each wire at the coil pack, I noticed that each time the idle changed, with the exception of wire 5. The arc between the wire and the tower looked good (which makes me wonder), but I couldn’t hear any idle change at all. I swapped two of the plugs, with the same result. I think I’ll try swapping two of the wires to see if maybe the wire at 5 is bad. Not quite sure I understand the part about changing the wire position on the coil pack (seems like this would cause it to run terribly). Sorry for being so thick about this...

It will cause it to run exactly the same as it did before. Your coil is not an old school distributor where having every wire in the exact right position is critical.

Your setup is what is called a waste spark system. There are actually three individual coils serving the six cylinders, enclosed in that one coil pack. Each pair of towers is hooked to a single coil, so when the coil (and consequently the plug) for #1 fires so does #5. So functionally swapping 1 and 5 is no different than having them in the "correct" positions.

Same with 3 and 6 and 3 and 4.

adsm08 said:

It will cause it to run exactly the same as it did before. Your coil is not an old school distributor where having every wire in the exact right position is critical.

Your setup is what is called a waste spark system. There are actually three individual coils serving the six cylinders, enclosed in that one coil pack. Each pair of towers is hooked to a single coil, so when the coil (and consequently the plug) for #1 fires so does #5. So functionally swapping 1 and 5 is no different than having them in the "correct" positions.

Same with 3 and 6 and 3 and 4.

Click to expand...

Got it. I’m learning.

So the coil pack and wires checked out okay (I’ll post my methodology and reading later, but for now, I have to run to work). So I’m thinking that the problem with cylinder #5 must be fuel-related. Anyone know if there’s a way to test the injectors? Also, if that one’s bad, can I replace just one or do they have to be done as a set?

Thanks, all!

EDIT - I may have been pre-mature in saying that the coil pack is definitely okay. According to the tutorial I watched, testing the primaries (the four poles where the wiring harness plugs to the coil pack) should result in a reading of .3 to .9 Ohms, when using the 200 Ohms setting on the Ohm meter. Instead, I get a consistent reading of 1.2 on each pole. I didn’t give it too much thought, since they’re all the same, but does anyone know if this could cause a problem in cylinder 5 somehow? If it matters, this test was performed on a cold engine.

.5-1.5 ohms is acceptable for coil "packs", single ignition coils have the .3-.9 spec, with standard volt/ohm meters

First test for injectors should be to unplug coil pack and crank engine over a few times then pull spark plugs #4 and #5, or #5 and #6, and compare "wetness" on the tips, see if #5 is "drier", not scientific, but still useful for sequential fuel injection.

You can unplug 3 injectors and then measure OHMs of each, 11-16ohms is spec for Ford injectors, but if all 3 are about 18ohms that's fine.
You do 3 to get an average to compensate for OHMs Meters bias, same as you do compression test using multiple cylinders to compensate for Pressure Gauge bias.

Run a can of Seafoam in that gas tank, or similar injector cleaner, it will clean the tips of the injectors.
I do this once a year on my EFI vehicles and it smooths out the idle after about 1/2 tank.

Dirty tips drip fuel, instead of spraying it, so mix at idle can be poor, causing misfires, at higher RPM, higher air flow, mix gets better and misfires go away