Topic: The most accurate compression gauge...

[fendersrule]

Good question. I'm at 2700 feet.

[Bob Wessner]

This might help.

Compression Test Altitude Compensation Factors Altitude Factor at various elevations.

500 0.987
1500 0.960
2500 0.933
3500 0.907
4500 0.880
5500 0.853
6500 0.826
7500 0.800
 8500 0.773

[fendersrule]

Thanks Bob.

This is information I didn't know. Basically, being at 2700 feet means that in "general", compression numbers would be about 10-12 PSI higher than what I'm seeing on the gauge. I did not realize this before and I will begin making additional calculations after every compression test.

For example with the Actron gauge, my CB750's compression went from 142/123/123/143 to 152/131/131/153 by adjusting for the high altitude.

I've already built a spreadsheet and I plan to compare the Actron and the Kawasaki shop gauge when it arrives. There are three possible findings:

1) The Actron reads higher than the Kawasaki Gauge. This is possible, but I feel unlikely.

2) The Actron and Kawasaki Gauge read the same. I believe this is very possible. This would mean that I would HEAVILY recommend the Actron gauge more-so than I do now.

3) The Kawasaki Gauge reads higher than the Actron gauge. I believe that this is possible as well due to the improved fittings.

Assumption: The higher the reading, (if the gauge is a quality gauge) the more accurate the number. Reason being is that there is only a chance for compression LOSS with gauges/fittings/hoses and not much chance for a gauge to actually add compression. This is a fair assumption.

I'll post the spreadsheet in a few weeks after I do some comparisons.

Some may think this is splitting hairs, but people doing compression testing can easily fall into pitfalls to where they are pulling an engine that maybe isn't as "tired" as they may think.

For example, if you're using an old cheap gauge, forgot to do something important (open the throttle, or warm up the bike), did not adjust valves, or did not adjust for your altitude, you can easily read 40 PSI less than what it's "really" at. That isn't splitting hairs.

It's interesting that Honda never sold a shop gauge. Kudos to Kawasaki on doing that.

[Bob Wessner]

Fenders, I think you have it reversed. The greater the altitude, the lower the air density. I'm thinking you should use the factor in the table and multiply it times the compression you 'would' get at sea level and you will get progressively  lower readings the higher you get in terms of altitude. At least that's my take on it. If I'm the one who has it backwards, hope someone chimes in.

Since I'm at a whopping 900 ft above sea level, I don't even consider any factor. 

[fendersrule]

I "think" I have it right (I did think about this a bunch yesterday) but let's make sure. Greater the altitude, the lower the air pressure. I pulled this from another forum:

"A standard compression reading of about 150 psi at sea level in Los Angeles would measure significantly less in the surrounding mountains. For example at an elevation of 6000 feet, the expected reading would be 150 psi x .8359 = 125 psi. The cylinders would be reading low if compared to sea level measurements, yet perfectly fine at this altitude."

That alone tells me that you need to multiply by 1.xxx if you are located above sea levels and want to see what the measurement is at sea level, which in my mind is the number you are looking for greater accuracy.

If you are at sea level, then you multiply the numbers you gave to see what your compression would be at various altitudes. I think this would be useful if you are bringing a toy to drag race in Colorado and it will give you an idea of the power loss you would expect.

So if you're 900 feet, you need to be multiplying your PSI by roughly 1.03. Around a 5 PSI boost for you.

For me at 2700 feet, I automatically lose 12 PSI. That's significant IMO, so I think it's very much worth factoring in what the compression would be at Sea-level.

[dave500]

ride it down to the beach and check it?

[kghost]

Nah you don't lose 12 psi........

Standard atmosphere at sea level is 29.92 inches of mercury (or 14.696 - lets call it 14.7 psi)

You lose roughly 1"  mercury per thousand feet. The pressure at 18,000 feet is half of that at sea level.

Now that assumes dry air and a standard temp.

Humidity and temperature will change the readings as well.

Its really easy to get overcomplicated trying to factor in too many corrections lol

[fendersrule]

I guess I'm going to put faith into the charts that are posted all over the internet about this and will proceed to perform the adjustment for being nearly 3,000 feet in the air.

That would mean around 12 additional PSI for 150-152 PSI received at 2,700 feet going by the chart. Obviously, it would be great to carry my compression gauge with me and drive a couple thousand miles to sea-level....I'd be pretty hardcore at that point.

I agree that it gets overcomplicated, but altitude is a pretty quick and easy correction to make that seems to be proven. Some people go as far to calculate the hose volume out, but I think that's where splitting hairs occurs. It would take someone with a PhD in physics to fully answer the question if the hose adds cylinder volume with a check valve at the fitting.

I will post both corrected and uncorrected values so I won't be hiding anything.

[kghost]

I guess we can forget about getting the most accurate gauge to attempt to replicate Honda numbers?

As long as you have a quality gauge and know what the readings are........

Its relative LOL

[dave500]

?why do a calculation to compensate?your riding at that level/height anyway and that aint going to change?just see how balanced they are,while were on the topic,checked your tyre pressures lately?

[Markcb750]

I would Calibrate the gauge by applying a known pressure, remember how far off it is and evaluate subsiquent readings per Honda's balance requirement of +/- 5%.  Save the money you will spend on an ultimate compression tester, spend it on the best torque wrench available.  Accuracy is essential when torquing the engine bolts the surgeons are encouraging you to loosen.

My 12 dollar digital volt meter indicated 14.1 V on a battery which should have read about 12.5. I went for years with this in my head and knew when systems were charging or batteries had a bad cell.  I did eventually pop the back off & adjust the calibration pot to indicate 1.5V on a new dry-cell, but that was only to keep me out of the house a while longer.


Remember

It's not the wrench it's the wrench monkey.

[Vinhead1957]

Just a gauge!  An indication of condition!  Precision is unimportant.  150 across the board great, 140-120 across good.   Now is getting into the grey area, below 110 or a cylinder not equal to its neighbor within reason +or- 15. A gauge that is 15 psi off will be 15 psi off all the cylinders so accuracy is unimportant!  The most important part of the numbers is what they read in relation to the group! One cylinder very low says you need immediate work.  Slightly low compared to others, work that can wait till off season : )

[John Eberly]

Assumption: The higher the reading, (if the gauge is a quality gauge) the more accurate the number. Reason being is that there is only a chance for compression LOSS with gauges/fittings/hoses and not much chance for a gauge to actually add compression. This is a fair assumption.

You are assuming that the "quality" of the gage is established before you purchase it, but gages CAN read high. Are you checking it against a standard?

Also, the volume of air in the gage hose should not matter - you are measuring pressure in pounds per square inch which will be independent of the volume of your system.

[Markcb750]

I want to say something witty, but I am left wondering:

Does the poster realy mean what is written?

Is it just a typo?

Was his science teacher actually an english teacher filling in? 

Who knows?


just saying.

[strynboen]

i think a  cb 450 are the most.....

[fendersrule]

If you guys think that accuracy doesn't matter, then please keep using the $15 gauge at Harbor Freight, and exit this thread.

Thanks.

[Markcb750]

i think a  cb 450 are the most.....

Indeed, The most...

My gauge is from NAPA, cost about100 bucks, has the check valve in the threaded end...still does not provide the numbers Honda indicates for the 750 or the 900, or others I have checked.  It is repeatable, and interpetable. 

No one buys Harbor Freight Gauges...

[fendersrule]

Just a gauge!  An indication of condition!  Precision is unimportant.  150 across the board great, 140-120 across good.   Now is getting into the grey area, below 110 or a cylinder not equal to its neighbor within reason +or- 15. A gauge that is 15 psi off will be 15 psi off all the cylinders so accuracy is unimportant!  The most important part of the numbers is what they read in relation to the group! One cylinder very low says you need immediate work.  Slightly low compared to others, work that can wait till off season : )

150 across the board is good when Honda calls for 185-195 (in the case of a CB450)? Leaves you wondering. That's the point of this thread. Let's stop wondering. No one is arguing that the primary purpose of a compression gauge is to determine cylinder variance, and then proceed to perform a leak down test. But me and many others would also argue that the compression reading should also be highly accurate as well. You may not feel so, but I do.

This thread is about accuracy, whether you care about it or not. This thread is for the people who do care about it.

My gauge is from NAPA, cost about100 bucks, has the check valve in the threaded end...still does not provide the numbers Honda indicates for the 750 or the 900, or others I have checked.  It is repeatable, and interpetable. 

And that's why you should be interested in this thread because some of us are tired of seeing 30+ PSI differences from what Honda prints. I'll be testing using a Kawasaki service shop gauge soon, which is the closest damn gauge you are going to get that is used in the shop manuals. It probably is the sam gauge, and for sure I will have the same adaptors seen in the manual. As kghost would agree, It's time to get myself a nice gauge for christmas.

[KJ790]

A compression tester was never meant to be a perfectly accurate test. It gives you a general idea of which cylinder is shot if you have a problem (i.e. if one cylinder tests at half the compression of the other three). This is like searching for the most accurate ruler because you want to measure your cylinder bores. There is no real hard line for compression readings that means it needs to be rebuilt. Nobody would ever say "oh, 140psi is fine, but 139psi means you need a rebuild"...

A leak down test tends to be more accurate, and tells you where you are losing your compression to (which a basic compression test does not). A compression test only tells you good or bad, then if it is bad you need to do a leak down test to find out where the problem is. Because of this, I don't see any reason ever to do a compression test, I just go straight to the leak down test and be done with it.

[Vinhead1957]

+1

But you can't get through with reason

If it means anything my gauge is over 30 before HF!

[TwoTired]

Some people go as far to calculate the hose volume out, but I think that's where splitting hairs occurs. It would take someone with a PhD in physics to fully answer the question if the hose adds cylinder volume with a check valve at the fitting.

You don't need a PhD.  I had this stuff put in the Engine FAQ.
http://forums.sohc4.net/index.php?topic=369.msg476032#msg476032

Compression ratio is the starting volume relative to the final volume.  Volume at piston BDC is cylinder volume plus chamber volume.  Volume at TDC is just chamber volume.   Since the piston's actual displacement doesn't change adding volume to the compression chamber (compression tester hose volume) lowers the compression ratio and final compression.  It makes quite a gauge reading difference if you change the CR from 9 : 1 to 5 : 1.

You can make an argument for subtracting the volume beyond the one way check valve, as it will equalize over successive pressure peak cycles, or even the spring pressure applied to the check valve the the pressure must overcome.  But, on the cylinder side of the check valve the pressure is lost and compressed from zero at each individual compression cycle.  It cannot cumulatively build up within the hose between check valve and cylinder anymore than it can in the cylinder itself.

Imo, if you want to know true cylinder health use a differential gauge.  This will also tell you where the leaks are by listening to the hissing air location.  The compression gauge is still useful for locating gross failures, or differential problems among cylinders.


Regarding the altitude pressure variation: note the barometric pressures vary with invading weather systems ranging from 980mb to 1050mb (13.7 PSI to 15.2 PSI).  Might be useful in your quest for "most accurate".

I submit, that a differential set up is far cheaper than a calibrated compression tester, a laptop running an Excel equations (or Matlab), and sensitive baro pressure gauges, which are used to extrapolate true compression values.


...Not, that mental masturbation can't be fun, as an alternative to actually repairing or diagnosing a motorcycle.   

Now for something else useful, here's how to turn a page for the newspaper.   

The Page Turner

[fendersrule]

Two Tired, you are awesome and a wealth of information.

So for someone that has very little mechanical knowledge and wants to use a compression gauge at 2700 feet, what advice would you give to get more accurate numbers? Keeping in mind that I've never stated that I wanted "perfect" numbers, but I want to get as close as I can with minimal effort.

Would you agree that using a certified Kawasaki service gauge and certified Kawasaki adaptors (3cc of hose volume) is a good step? In the FAQ, you mention this exact adaptor that I'm getting (has a long thin tube), so any additional chamber volume is going to be minimal...

I'm not sure what you meant by the altitude part of your post, but would you agree that  factoring in altitude at 2700 is the right move? According to the charts posted online, the correction for 2700 feet to convert it to sea-level measurements would be about 1.067, on the conservative side.

Take a look at the adapters that I will be receiving (I've ordered both a 14mm and a 12mm adapter). They are OOP, and are NOT cheap. This is an expensive test that I will be doing.  :

http://forums.sohc4.net/index.php?topic=133013.msg1489177#msg1489177

[2wheels]

  Save the money you will spend on an ultimate compression tester, spend it on the best torque wrench available. 

I think calibrating the pressure gauge makes sense, no matter how much you spend buying it.
Torque wrenches should be calibrated too, probably will cost more than the wrench did

[fendersrule]

I'm covered for Torque wrenches....

I've got two Craftsman wrenches, one in inch pounds and one in foot pounds. I also got an old-school lever-dial torque wrench.

[Markcb750]

Craftsman T.W. are good ones.  they are warranted to be +/-5% full range.  Stored well they will function a long time.

When I Calibrated Wrenched for the Air Force The best were Snap-On.


I have Craftsman also FWIW.