General compression test question... applies to XJs too

OK. So, I recently did a valve job on my "other" bike in order to replace four bent intake valves. Before the work I was reading anywhere from 10 to 90 psi across the four cylinders. After, I'm reading about 105psi on all four.

This sounds pretty good to me, but also had me a bit worried when compared to the typical response of "120 or higher", and also to my brother-in-law's readings of 160psi on his ZX10R down in Florida.

This got me thinking about what's happening when running a compression check. If I understand it correctly, you're (ideally) taking the volume of air in the cylinder, at local atmospheric pressure, and compressing it by the engine's compression ratio, and reading the pressure number as a result of that operation. Boyle's law says, more or less, that if you compress air 2:1 you'll have double the pressure. However, since the compression gauge is actually comparing the difference between ambient pressure and cylinder pressure, it's only going to show one times atmospheric pressure when reading air compressed 2:1, and 0 psi when reading open air.

Given this (assuming I'm on the right track): my ambient pressure up here in Colorado Springs is about 11.78psi... let's call it 12psi for the sake of simplicity. The "other" bike's compression is 9.5:1.

So, 8.5 times 12 (since we deduct 1 times to account for the ambient pressure the gauge is comparing to) = 102. Pretty close to the 105 my, admittedly cheap, gauge shows.

Am I missing something here, or does this make sense? If it makes sense, where do the 200psi numbers I've seen some people posting come from? XJ750's, at least, only run 9.2:1. Sea level and 12.7:1 would give you a theoretical 164psi (which would match the ZX10R in Florida).

Cheers,
Paul

 

Compression figures are wholly dependent on the condition of the Engine during the Test.
A weak Battery, Throttles, leaking Gauge Hose ... all factors.

The test is meant to show you how condition of the Rings and Valves are.
Obviously, the higher numbers indicate better condition.

Low readings, followed by a Wet Test will tell you if the Compression is being affected by Rings of Valves.

 

To effectively perform a compression test the valves should first be set to spec., the engine run to operating temp, the plugs removed, the throttle should be wide opened during cranking, engine cranked 4 - 6 reveolutions (all cylinders should be cranked an equal number), the result should be recorded snd compared to the other cylinders. There should be no more than 20psi between the highest and lowest cylinder and all cylinders should be above 100psi. The process should then be done with a tsp of oil in each cylinder. If the reading increases in any one cylinder by more than 30psi, it indicates a ring issue. If there was no substantial increase in the lower reading cylinders, it indicates a valve issue.

Having said that, if the engine has adequate power, has no misfire, and isn't smoking, why bother. This process is to isolate problems in performance, or oil consumptiom issues.

 

Please read my WHOLE post Rick. I don't think this is all there is to it. For example, regardless of any of your factors listed, an engine tested in a vacuum would read 0 psi. Altitude and designed compression ratio must factor into what the expected numbers should be.

Paul

 

Really?

If I'm right in my analysis, a perfectly sealed XJ750 would read 83 psi if tested in Cripple Creek, Colorado (10000 feet above sea level - a small town not far from me).

Ambient at 10K feet is 10.1 psi. Compression ratio of a stock XJ750 = 9.2:1. 8.2 times 10.1 = 82.8 psi.

Please explain what I'm missing that would get me above 100 psi under those conditions.

Thanks,
Paul

 

Paul, I believe you're correct about altitude affecting compression numbers... I used to have a chart somewhere that went all the way up to 12,000 ft, but I can't find it now. I think the factor for 10000 ft. would be approx. 0.74 (e.g. 120psi measured at sea level would be more like 89psi). IIRC, you have to take altitude and ambient temperature into consideration.

You might consider a leakdown test to confirm whether your compression is where it needs to be.

 

As soon as both valves close ambient pressure is no longer a factor. You're squeezing trapped air, it's not subject to atmospheric pressure, just to the pressure applied by the piston.

I'm not a physicist, I Googled this right after I saw the first post. I thought it might too, but it doesn't. Ambient HUMIDITY does affect compression testing, but only to a minor degree. The biggest variations are caused by greater differences between the air temp and the temp of the motor, which is why we test cold.

 

Altitude would affect the test by virtue of differences in air density.

 

Hi Paul

You are probably correct in that analysis. I have been a mechanic for nearly fifty years, but always at sea level (or just above.) The 100psi figure is what we commonly use here. It, like all figures is a guideline. There will be variations depending on conditions and not having to deal with elevation issues, we sometimes disregard what may be pertinent to those who do have to contend with those factors. These conditions are why a carbureted engine set up to run at sea level will require resetting if taken up a mountain. The oxygen content, "thinner" air, manifold pressure
etc etc will affect the air fuel ratio and other tuning items. I guess I forgot that the GTA (Toronto) isn't the center of the universe

As was pointed out the compression of an engine in a vacuum will be zero. You can't compress "nothing."


John

 

pretty sure this is not the case fill a container to 10 PSI at sea level and take it to Denver it still has 10 PSI
my compression gauge says to crank till the needle stops going up so it would take more strokes at higher altitude but the ultimate reading would be the same wherever you are

 

Not so. The valves keep opening and bleeding the pressure out of the cylinder. When you're cranking multiple strokes it's to pump up the gauge which effectively lowers the compression by expanding the volume of the head. The gauge has a check valve so it doesn't bleed down.

I think there is something to this. As SQLGuy pointed out, in a vacuum there will never be any pressure built up.

Bigfitz, we are actually using ambient pressure as a substitute for density. At higher pressures the air has more density.

Now, by my Boyle's law calculations using an engine with a 9.2:1 compression ratio at sea level (14.7psi) yields a compression pressure of 135.24PSI.

Why is the compression spec higher than 135.24PSI you ask? Because the atmosphere is not an "ideal gas". Also, Boyle's law is based on no change of temperature. When you compress that air it gets hot, raising the pressure. With an ideal gas (and no leaks) it'd be 135.24 PSI after it cooled down.

 

If your comp' reads 150 at sea level, it would read 105 at 10000 ft alt'
The speed the motor spins at will also affect the outcome.

 

No, 120.54 PSI. You forgot the initial 14.7 that the gauge reads relative to.

In other words, when you just have your compression gauge sitting on your bench, it reads 0, because the pressure around the gauge body is 14.7PSI and the pressure inside the hose is 14.7PSI, and the difference is 0. When compressing the cylinder's air 9.2:1, you are increasing its pressure by 8.2 times the ambient, not 9.2 times the ambient.

The absolute pressure within the cylinder would be 135.24 PSI, but the gauge would read the 120.54 PSI difference.

Not sure how much effect the heat of compression will account for, but, you're right, there should be some.

 

What if you held the pressure relief button when attaching the gauge? Would that difference then be negated, and you'd get a true reading?
What if you held the relief button for the first stroke or two then released it?
Would the above method yield a more accurate reading?

 

I said PSI not PSI gauge, so I have an out. :wink: You're correct though a compression gauge would read 120.54 PSI.

 

No. The pressure relief button only opens the check valve in the hose, it doesn't have anything to do with the pressure around the gauge body. It's for holding the peak pressure value in the hose.

Think about it the other way around: what would happen if you put the gauge in a sealed jar with the hose sticking out through a grommet or some silicone, then pressurized the jar by feeding some air in from a compressor? Now, assuming the relief button was pressed or tied down, the pressure in the hose would be lower than the pressure around the gauge body, and the needle would read less than 0... same as if you applied vacuum to the hose with the relief button pressed.

 

Holding the relief button just lets the pressure off.

You get an accurate reading. The extra strokes are required to bring the gauge and hose up to compression pressure. Until that happens the instrument is essentially a leak.

 

Add another wrinkle- - everyone is assuming the intake valve closes at BDC, it actually is still open long enough to vent out maybe 2 whole compression points- - just look up the cam specs.

That means you are not reading the total compression capable anyway.
As MiCarl said, the heat and the eventual incompressibility of the molecules comes into play. (you only compress the empty space between the molecules)

And Polock is finally wrong ! That 10 PSI container would read 13 PSI.
It had 25 absolute PSI at sea level.

 

Carl; I understand that. The theory as I understood it was that the pressure "trapped" within the gauge would attenuate the gauge's reading.

My original point was that once the valves were closed, we were dealing with a volume of air no longer subject to atmospheric pressure; if there was a different pressure of residual air trapped in the hose, holding the button would equalize that difference. Once the button is released, the hose, gauge and cylinder (once the valves close) becomes one equalized "chamber" not influenced by ambient atmospheric pressure.

I don't know the answer, I'm asking questions that test the theories.

 

I'm not sure I understand what you're asking about the relief valve. The relief valve opens the check valve in the hose, so that pressure within the hose returns to ambient.

When testing compression, the check valve maintains the highest pressure within the hose, so the peak value can be read on the gauge. If you held the button during testing, the gauge would drop back down to zero after each stroke.

The gauge is showing you the difference in pressure between whatever is in the hose and ambient.

So, let's say you start by pressing the relief valve. This opens the hose so that the pressure inside the hose is equal to the pressure around the gauge body, and the gauge will now read 0. If you now release the button, the gauge will still read 0 because the pressure inside the hose is still at ambient even though that air is now isolated from the atmosphere. This actually ties well to Polack's example: if you press and release the button on the gauge at sea level, and then bring it to Cripple Creek, the gauge will now read 4 or 5 PSI, due to the ambient pressure around the gauge body being 10 PSI and the pressure trapped in the hose still being the 14.x of sea level.

Anyway, back to your zeroed gauge: you now attach it to the spark plug hole and crank the engine. On each stroke, the intake valve opens and piston moves down, allowing air at local ambient pressure to flow into the cylinder. (Unless you have a SECA turbo, with the engine running, the air filling the cylinder will never be under any higher pressure than ambient.) So, ideally, when the intake valve closes, the cylinder will be filled with (engine size / 4) cc's of air under ambient pressure. Even though the valve is now closed, nothing has yet happened to change the density or volume of this air; therefore it is still under the same pressure as the outside atmosphere.

As the piston now starts moving up, though, the air is compresed. It's volume decreases and its density and pressure increase. This higher pressure air is also forced into the hose of the compression gauge, pushing a spring that was previously at equilibrium between the ambient pressures inside and outside the gauge. After TDC, the piston moves back down, reducing the pressure in the cylinder back towards ambient, but the higher pressure air in the hose remains trapped by the check valve. The air that was in the hose was some additional air that would offset the effective compression within the cylinder. By trapping the pressure in the hose with the check valve, the offset from the true compression reading will be less on the next compression stroke and on each following one, until a fairly true peak reading is achieved.

 

Fitz,

All the discussions we are having are making some assumptions:

1 - We're talking about an ideal gas (which air is not, it's a mixture and variations in humidity change it's composition).

2 - Temperatures are constant.

Please allow me to use the term air as a substitute for "ideal gas" in this description.

The number of air molecules in the cylinder is determined by it's volume, the temperature and the pressure. Since we're not going to change temperature the number of air molecules is only dependent on cylinder volume and pressure.

Now, assuming no leaks the number of air molecules in the cylinder during compression is constant. As the piston moves up the volume decreases. Boyle's law states that as volume decreases pressure must increase (at constant temperature) and that it's linear.

Now your compression ratio is 9.2:1. That means for every 9.2 cc at the bottom of the stroke there is 1cc at the top of the stroke. So, the pressure at the top of the stroke will be 9.2 times the pressure at the bottom.

Now, at sea level, atmospheric pressure is approximately 14.7PSI. That means that the pressure at the top of the stroke is 135.24PSI.

If you take your gauge out of the tool box you'll see it doesn't read 14.7, it reads 0. That's because your gauge is a differential instrument. It measures the difference between ambient pressure and the pressure at it's inlet. So, what the gauge reads at the top of the stroke is 135.24 - 14.7 = 120.54 PSI. This is called Gauge pressure (PSIG). Since most of us don't work in absolute pressures we get lazy and just call it PSI.

Now, let's assume someone is on a high mountain other than at sea level and that atmospheric pressure there is 12.5PSI. Since his pressure is 85% on sea level he'll start with only 85% of the molecules the guy at sea level starts with. So, his final pressure will only be 85% of the TDC pressure at sea level - 114.95PSI. You have to subtract the ambient pressure (12.5psi) to get the gauge reading - 102.45PSIG.

So, an engine that reads 120.54PSI at sea level will only read 102.45PSI on the mountain.

Now lets look at how the compression gauge effects this:

Before you start the air inside the gauge and hose is at ambient pressure. As the piston moves up the pressure in the cylinder increases. Since the pressure in the hose and gauge is lower some of the molecules will leave the cylinder to pressurize the instrument. At the end of the first stroke you're just like the guy on the mountain - less molecules in your cylinder so your pressure is low. Now, on the next stroke there instrument is already above ambient pressure (it has a check valve) so less molecules need to move in there so the reading after the 2nd stroke is higher. After a few strokes the instrument is pressurized and no more molecules will leak from the cylinder. The gauge quits jumping up and you have your true reading.

Now the reason that the readings will be higher is because the calculations so far assumed no change in pressure. In the real world compressing the air raises it's temperature a lot (the output pipe on your air compressor gets hot enough to burn you!). That higher temperature also increases the pressure.

 

So higher ambient atmospheric pressure automatically (and directly) affects air density?

 

Yes

http://www.engineeringtoolbox.com/air-t ... d_771.html

 

For the engineers in the group. Not my work, but cute:

A thermodynamics professor had written a take home exam for his
graduate students. It had one question: “Is Hell exothermic (gives off
heat), or endothermic (absorbs heat)? Support your answer with a
proof.” Most of the students wrote proofs of their beliefs using
Boyle’s Law (gas cools off when it expands and heats up when it is
compressed) or some variant. One student, however, wrote the following:

First, we need to know how the mass of Hell is changing in time. So, we
need to know the rate that souls are moving into Hell and the rate they
are leaving. I think that we can safely assume that once a soul gets to
Hell, it will not leave. Therefore, no souls are leaving. As for how
many souls are entering Hell, let’s look at the different religions
that exist in the world today. Some of these religions state that if
you are not a member of their religion, you will go to Hell. Since
there are more than one of these religions and since most people do not
belong to more than one religion, we can project that all people and
all souls go to Hell. With birth and death rates as they are, we can
expect the number of souls in Hell to increases exponentially. Second,
we look at the rate of change of the volume in Hell because Boyle’s Law
states that in order for the temperature and pressure in Hell to stay
the same, the volume in Hell has to expand as souls are added. This
gives two possibilities:

1. If Hell is expanding at a slower rate than the rate at which souls
enter Hell, then the temperature and pressure in Hell will increase
until all Hell breaks loose.

2.Of course, if Hell is expanding at a rate faster than the increase of
souls in Hell, then the temperature and pressure will drop until Hell
freezes over.

So which is it? If we accept the postulate given to me by Miss Therese
Banyan during my freshman year that, “It will be a cold night in Hell
before I sleep with you,” and take into account the fact that I
still have not succeeded with her, then #2 cannot be true, and so Hell
is exothermic.

The student got the only “A”.

 

Yep. I recon you'd be all day tryin' to get your Compression Checked in a Vacuum.
Them vacuum's sneak-up on you, too.

You'll be standing there minding your own business and suddenly a vacuum happens and sucks a burst of air right out of your butt!

 

AND rightly so,original thought = original comedy i give it an A just for the definitive expression of a hopeless answer in a helpless world !!

 

Just roll the POS down to the beach & check it there. 8)

 

Hey! POS!?

Not neccessary, anyway. Unlike some interminable restoration projects, this bike is currently ride-able. Still a long way to the ocean from here, though. Death Valley, which is a bit below sea level, is about 1000 miles / 1600 km.

 

Wow, you guys are all a bunch of nerds. LOL!

I really like these kinds of threads. You can learn a lot in a short period of time!

Don't get me wrong, you are all still a bunch of nerds, but the discussion was really cool!

I like hang out with folks who can exchange ideas like this. I think it makes us all better people!

Just so we are clear, I'm a geek, not a nerd.

greg

 

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
SEE IQ POSTING FROM YESTERDAY BOYZ,THINK THIS GUY IS ONE O THOSE ONES. ANY GUESSES? I BET 79

 

The only thing I learned was some people can make something very simple pretty damn complicated

105 across all cylinders, so long as it runs well and isn't using oil the rest doesn't matter, but the "wet test" Rickomatic suggested would be worth doing

I've always been told to test at operating temperature

 

Performing the compression test should be done at operating temp. That's when all the parts fits as they are supposed to. Until the engine reaches operating temperature valve clearance, ring end gap, and piston fit are all loose. The cold testing is just less accurate.

 

I was dealing with a half-way stripped engine on which I'd just finished a valve job. As I mentioned, before the valve job, I was reading anywhere from 10 to 90 PSI. Clearly, getting up to a consistent 105 was an improvement, but it would certainly be nicer to know whether I needed to do rings as well without having to put the rest of the bike back together and run it to see how well it ran or whether it used oil.

Cold I could still pull the head and jugs back off without having to go out and buy another gasket set. To me, that bit of math to figure out how much lower good compression should be up here is less complicated than having to open the engine back up later for more work.

 

Doing a straight ratio your 105 comes out to about 120 sea level (I used your 11.78 ambient pressure vs. 14.7 sea level). A bit low, even for a cold engine.

I'm assuming that you can do a wet test? Be interesting to see your wet numbers. If it didn't go over 115 wet I'd slap it in the bike and ride it like I stole it. Remember, just enough oil to seal the rings not enough to change the compression.

 

Sorry SQLGuy, I didn't know you had a sense of humor lobotomy.

 

Didn't see any or some such around Kickaha's comments. To me they read as his feeling this was a pointless discussion.

 

Too late, the bike's back together at this point. Also, I got yelled at by a local guy that builds race engines about the "teaspoon of oil" test being something that's way outdated and should never be used these days.

Are you at sea level? If so, can you do a cold test on an engine you feel is in good shape? I'd be interested to see those numbers as well. You say 120 is a bit low. What are you used to seeing on a 9.x:1 engine? By the way, 14.7/11.78 * 105 = 131.

In the meantime, I'll probably do a hot test on mine over the next few days.

Cheers,
Paul

 

sorry, i should have explained thisis the gauge i use
and remember to index your plugs

 

What the race guy says & does isn't important, he would rate engine wear in laps, we deal in years.

 

here ya go, at 1250 feet, Pressure: 29.75 in (Rising) Humidity: 100%
1 138
2 140
3 125
4 140

 

Tell him that your engine isn't from these days so the teaspoon of oil will work.

I'm at about 700 feet here and XJs come in at 140-150PSI. The books are at the shop but I think the spec is 135-155.

 

Thanks Polock. Was that cold or hot?

I just checked mine hot - oil temp at 200 degrees F. All cylinders were between 117 and 120 on my gauge. So, adjusting for altitude: 146 to 150.

MiCarl, are your 140 - 150 numbers hot or cold?

 

Uh-oh, Rosie is in trouble then...

With a "cold" engine

at 91°f and 80% humidity about 120 ft above sea level...

1 - 121 psi
2 - 120 psi
3 - 120 psi
4 - 123 psi

Plugs are the color of creamed coffee and no smoke anytime.

But my compression gauge is a POS.

 

that was stone cold, hadn't been run in 4 days

 

At 200 feet, 70 * and unknown humidity, my 750 pulled 147's with the battery boosted, cold. And I think that's pretty durn good.

So at your altitude, I'd pull 118 ??- - sounds like you're good !!

 

One aspect that hasn't been mentioned that will have an effect.. at least on the absolute math..Valve lag/overlap.. Let round 2 begin

 

My new project bike, non-YICS 650 (9.2:1,) bone cold (hadn't run in 15 years) gave me 153-149-150-150, one of the reasons I bought it. We're at about 740' above sea level.

Skippy344 Rosie's fine. Absolute numbers aren't as important as differences; try a different gauge I'll bet you get different numbers.

 

Man... I still don't get it. 150 cold, somewhat above sea level. That means you're getting at least 30 PSI "extra" from one or more of non-ideal compressibiliy of air, heating of air during compression, or deposits reducing the combustion chamber volume.

I soda blasted my combustion chambers and piston crowns while I had it apart, so no "extra" from deposits, but also probably a bit of loss for the time being because I disturbed the carbon ridge at the top of the cylinder.

Still, all in all, it sounds like my numbers are a bit low. Especially since this engine is 9.5:1.

On the othe hands, the humidity here is pretty low, and my mixture is showing lean... I wonder how much water vapor and fuel vapor affect the compressibility of air....

Cheers,
Paul

 

for the sake of science, you should make yourself a leak down tester
then the tale will be told
just make sure you hold that crank tight

 

I'd start by using a different gauge. That's why I own two. I'm assuming they're both reasonably accurate as they always read within a pound or two of each other on the same cylinder.

And offhand I'd say it's the non-ideal compressibility of air, as my spec is 135-156 (from my factory 650 book.)