Checking Compression...

[crazy steve]

There's been more than few posts over the last months about what a 'proper' compression number should be, with some getting results both low and high. 'Everybody' knows that factory spec is shown as 142 PSI, +/- 14 PSI, for a total range of 128 to 156 being 'acceptable' as long as the variation between cylinders doesn't exceend 10%. This is with the throttle fully open; having it closed will give you lower numbers.

But Yamaha isn't all that clear on just how to check this, other than noting that the throttle needs to be open. After spending a few hours this afternoon checking my three used motors to see which one I want to use in my 'project' bike, I think I can shed a little light on this.

I consistantly got higher numbers than 'spec' with two exceptions. I was checking two '80SG motors, and a '79F unit. The '79 I'm reasonable sure has about 44K miles on it, as the bike looked to be pretty much unmolested other than an aftermarket exhaust. The miles on the other two are unknown, as one wasn't original to the bike and the other had the speedo changed, so who knows with those. The numbers I got were these:

'79F: 1. 195 2. 195 3. 195 4. 190. Good numbers; within 5 PSI/3% of each other, and nice and high. Considerably higher than 'spec'.

'80SG 1: 1. 190 2. 190 3. 160 4. 150. All above spec, but 3 and 4 are a bit too low; 15 and 21% low respectfully. This one will run, but not at it's best....

'80SG 2: 1. 190 2. 170 3. 170 4. 190. Again, all above spec. This one is better, as the difference is right at 10%.

Now, how did I test these? I used a near-new Snap-On gauge, all three motors are out of the bikes, and all three have the carbs removed. Now, I thought that the carbs being missing would make a difference, but it doesn't; I tried installing carbs and it made little or no difference. I saw a few PSI difference on some of the higher-testing cylinders, but that could be within the error of the gauge (you were right Cy!). On all three, I shot some WD40 into the cylinders and let it soak a bit to lube the rings, but I doubt if that's thick enough to really change the numbers if the rings are bad. I'll note that I had previously checked the '79 w/o WD40, and got 190, 190, 120, and 220, so I can't explain that except that maybe one wasn't sealing because of stuck rings, and the other sealed really well for the same reason. I did test each cylinder of each motor at least twice, going through the full sequence each time (1, 2, 3, 4, then back to 1 etc). Some cylinders started much lower and came up as I spun the motor more, but I saw little change in the high ones.

To spin these, I used a 200A 'booster' function on my charger. Now, that probably affects the results, as this really spins these. I'll also note that I had all four plugs out, installing the tester into each cylinder and leaving the others open. So if spinning the motor on just the bike battery, you may see lower numbers. You'll also see lower numbers if you leave the plugs in, and remove them one at a time to put in the tester; I also did this on 1-2 cylinders on each motor, and got 20 PSI lower numbers consistantly because the motor didn't spin as fast.

So, what's it all mean? Well, without stating any conclusions, this would seem to suggest that if you're testing the motor in the bike with the bike battery and if you only pull one plug at a time for testing, then the 142 PSI number should be considered your 'benchmark'. If your results fall within the 'factory range' (and 10% or less to each other), the motor has no issues. If you pull all the plugs at once and then check each one, you should see numbers about 20 PSI higher, remembering that you still need to stay within that 10%. Spin the motor extra-fast with a boost, still higher numbers. If you use a light penetrating oil like WD40 or PB Blaster for lube, you may see about a 5 PSI gain. So, taking my 195 readings of the '79 motor, if you deduct 5 PSI for lube, 20 for no plugs, that drops the reading down to 170. Spin the motor slower, and that could be the other 15 PSI, putting this at 155, or the high end of the factory range. Looks like that one's a keeper...

Any comments or observations?

[Rickrod3]

Hey Steve:

Pretty interesting.
From reading your posts and advice you have given me; I fully acknowledge you are more knowledgeable than I am.

I just wanted to mention that I have read more than once that excessive carbon build up on valves may cause higher compression.
I guess it could make the combustion chamber smaller increasing compression.
How does that fit in.

JAT
Rick

[DGXSER]

Yep anything that coats the surfaces will shrink the size of the chamber, including the piston tops being coated. I would think that has more to do with the high numbers than anything else.

I would not suspect the higher revs to be THAT big of a change in compression. All your really doing is to reduce the time for leakage to occur. The piston still moves the same distances, and the valves still open the same distances, with the same timing. So it should not substantially change the compression numbers.

I have ALWAYS been told that when running compression numbers to pull all the plugs out of any engine, 4 cyl, V8, or V10. Mostly to save the battery form turning against that compression and draining it while your trying to crank the cylinders over.

[crazy steve]

..I would not suspect the higher revs to be THAT big of a change in compression. All your really doing is to reduce the time for leakage to occur. The piston still moves the same distances, and the valves still open the same distances, with the same timing. So it should not substantially change the compression numbers..

That surprised me too; but I checked and rechecked this several times and kept getting the same difference in the numbers. The motor was turning slower, but it didn't seem like it was that much slower.

The other thing I can't say is just what the difference (if any) you would find if you performed these tests with the motor in the bike and through normal starting circuits. I'm guessing that they would be lower numbers, and the difference between plugs in/plugs out might not be so pronounced...

And the slightly higher numbers on the '79 could be due to it's slightly higher compression ratio, although without knowing the history of the motors that's another guess.

[petejw]

hi steve,

Just something else to mention,
in that the engine should be fully warmed up
when checking,
which can be hard to do when there out of the bike.

[crazy steve]

Just something else to mention, in that the engine should be fully warmed up when checking, which can be hard to do when they're out of the bike.

Yep, that's another variable that's not factored in....

So, should compression go up, or go down when the motor's warm?...

[petejw]

the compresion should go up
when the engine is warm as the rings
should be sealing,
i was surprised to see those numbers on a cold engine.

[psycoreefer]

motor spinning faster or slower should make a difference, and the difference may be different for early and late motors because of the difference in the cams.
There are two types of compression I've heard referred to as "static" and "dynamic"

When the motor is spinning slower the "static" compression is usually lower. I've heard this is because the amount of time that both valves in a cylinder are open is longer (not more degrees of rotation but more time in fractions of a second) and therefore allows more of the air to be pushed back out.
The other reason is that even though we don't think about it air has mass - and at lower speeds the air doesn't pick up as much speed in the intake track and therefor doesn't cram into the motor like it would at higher speeds.

When the motor spins faster both these things (and likely a million more) are reversed and you get a higher "dynamic" compression.

So I'd say your results line up with what I'd expect - and BTW the important part is the <10% difference. As long as all the numbers are "close" to each other the motor will likely run, although it will probably burn oil and be down on power if the numbers are low. I once had a truck that tested around 60lbs on all 6 cylinders, I drove it for 2 years without a problem, I just had to put a quart of oil in with every tank of gas.

[WMarshy]

Interesting conversation... I just wanted to add, wd40 and PB Blaster are not lubricants. They have solvents in them that will break down the lubricant on the cylinder walls. Your better using transmission fluid or 10w 40... I've seen bearing get distroyed in short order from someone "lubing" it with wd40...

[81xsproject]

I would imagine that spare engines laying around probably haven't had the valves adjusted which can make a big difference if you have valves with zero clearance. When I did my valves after getting the bike, I had 3 different ones with 0 clearance. If they aren't closing all the way, it has to hurt compression.

[natemoen]

I would imagine that spare engines laying around probably haven't had the valves adjusted which can make a big difference if you have valves with zero clearance. When I did my valves after getting the bike, I had 3 different ones with 0 clearance. If they aren't closing all the way, it has to hurt compression.

But the issue with that would cause the valves to be held open and would cause the compression to drop off, and that obviously is not the issue here so its a moot point.

[psycoreefer]

what do you mean would cause it to "drop off"?

Every compression tester I've ever used has a check valve in it, it recordes the "highest" pressure it sees and then you have to press a release valve to reset the gauge to 0 for the next cylinder. It would show a cylinder with a stuck valve as a cylinder with "low" pressure.

Now if you were to use a leak down tester then you could see the pressure drop and then by some diagnostics determine if it was worn rings, stuck valve etc...

[natemoen]

I meant it would be low not that it would leak out. I know about the check valve in compression testers.

[81xsproject]

But the issue with that would cause the valves to be held open and would cause the compression to drop off, and that obviously is not the issue here so its a moot point.

I wasn't referring to his specific numbers so much as compression checking in general. However, "'80SG 1: 1. 190 2. 190 3. 160 4. 150. All above spec, but 3 and 4 are a bit too low; 15 and 21% low respectfully." I would say that there is at least a chance that 3 and 4 on these could need valve adjustment. And, zero clearance doesn't necessarily mean that the valve isn't seating or very close to it as I would imagine it would be difficult for the valve to wear its seat much once it is always riding the cam.

[crazy steve]

Besides telling me which one of these motors will be the best candidate for my project bike, the main reason for doing all this was to give some sort of context as to what these numbers may mean. I've seen more than a few posts where somebody has checked their motor and wants help to figure out the results. There's been some who were concerned because they got higher numbers than was 'speced' (worried about detonation), but it now seems to me that those worries may be unfounded, depending on how the motor was checked.

To me, that 190 PSI number is a good number when testing like I did. After all, 8 of the 12 cylinders I checked showed that number and it's unlikely to be mere coincidence. If you were looking to buy a used motor that's out of the bike, this would likely be the method you would use, and these are the numbers I would want to see. If you got the 'factory spec' number under these conditions/procedure, you would probably find you had too-low compression once the motor is in the bike. I'll also agree that valve adjustments can be an issue; I haven't checked those on any of my motors, and that could be the reason for the low cylinders I got. Sticky valves could be another issue if the motor has been sitting and all lube has drained off. But mainly what I was after here was what should be a good number; the '10% rule' will tell you if you have low cylinders, whatever the numbers you get.

Carbon build-up has been mentioned as a possible reason for these high numbers. Well, I looked at the plugs as I pulled them and the only ones that didn't have a tan/brown color to them were the ones from the low cylinders, which were more black. If you had enough carbon to raise the compression by the amounts I saw, I would think you'd find evidence on the plugs. Of course, the only way to get a definitive answer to this is to do a before/after test if removing the carbon.

The other thing that was raised is the hot vs cold measurements. I'm not so sure you would actually get an increase on a warmed-up motor under all or even any conditions. Yes, I'll agree that ring seal will probably be better, but you also have the effect of the 'growth' of the aluminum cylinders/head. I know for a fact that the aluminum cylinders on Harleys can expand up to .100" in length from cold to fully warmed; that will very definitely affect compression numbers. While I doubt if the shorter XS cylinders grow that much, that is another likely factor. Again, a before/after (cold/hot) test will be the only way to get a definitive answer. Anybody for adding to the knowledge base?