Compression tests are BS - an example

Eric Pauley

Eric Pauley

Pre-takeoff checklist
Joined
Apr 1, 2021
Messages
233
Location
Madison, WI
Display Name
Display name:
Eric Pauley
Rant time. Nothing that hasn't been said before.

About two years ago I bought my plane. I had a pretty thorough prebuy done, and as part of that the mechanic told me the compressions were:

74 68 72 75

with leakage past the rings. Oh no, better prepare to trash that cylinder! I took this info to the seller and took $4k off the purchase price. For reference, this engine had 2800 SFRB. Good deal, enough to replace that bad cylinder!

The next year, had the plane in for annual (2900 SFRB) and got the following compression readings:

76 77 75 78

That bad cylinder was now my second best! For reference when the engine came from the factory the first compression reading was:

73 75 75 74

That's right, all compressions at annual were as good or better than new, at 2900 hours.

I've attached every compression test since rebuilt. The prebuy test is pretty obvious.
Compression tests are military-grade BS. Prove me wrong.:popcorn:
 

Attachments

  • Screenshot 2023-05-19 at 10.41.47 PM.png
    Screenshot 2023-05-19 at 10.41.47 PM.png
    92.4 KB · Views: 29
Compression testing isn't total BS, it just not a sole arbitrator of condition. Compressed air hissing into the exhaust or intake manifold is a sign of real trouble.

Continental agrees with you that a negative compression test is not the final determiner of trouble, but might be a piston's ring gaps lining up to prevent seal. A situation which can be remedied with further flight and retesting.

High compressions are a good sign and low compression is a trigger to investigate further. See Continental's Service bulletin SB03-3, which is now incorporated into Continental's Standard Practice Maintenance Manual M-O.
 
Last edited:
If my memory is right, Smokey Yunick didn't believe compression tests were a great indicator of performance, either, and preferred a leak down test. That info is old, and automotive, and he was a pretty opinionated guy at the time, but he had a pretty good reputation of making engines work well. Oh, and if the NY Times obit is right, he was also a B-17 pilot in ww2.
 
Albany Tom said:
If my memory is right, Smokey Yunick didn't believe compression tests were a great indicator of performance, either, and preferred a leak down test. That info is old, and automotive, and he was a pretty opinionated guy at the time, but he had a pretty good reputation of making engines work well. Oh, and if the NY Times obit is right, he was also a B-17 pilot in ww2.
Click to expand...

You should read his autobiography. Pretty interesting.
 
Out of curiosity, when the pre buy reading was taken, had the engine sat for a long time? In non aviation engines I’ve seen many times that engines that have sat (many times for years) unused will show low compression readings/high oil consumption initially, then after a short time of heavy load use “heal” themselves. I have a couple possible theories as to why, but would not be surprised if this is the case.
 
I had a similar experience with a IO-540 at prebuy. One cylinder at 65psi with the other five in the upper 70’s. My broker had the price reduced 4K to compensate for a cylinder change. He took the plane up the following morning and ran it hard for 30 minutes then checked the compression again…..all in the upper 70’s. Carved a chunk out of the JPI 830 I had his shop install at the annual a few months later.
 
Flatiowa said:
Out of curiosity, when the pre buy reading was taken, had the engine sat for a long time? In non aviation engines I’ve seen many times that engines that have sat (many times for years) unused will show low compression readings/high oil consumption initially, then after a short time of heavy load use “heal” themselves. I have a couple possible theories as to why, but would not be surprised if this is the case.
Click to expand...

Not especially so (150h in the year before the test) but previous years it did 400 so relatively? It may have sat for a few months before. Since then I’ve done 100/yr though.
 
I see compression tests, like bore scoping and oil analysis, as just one more tool in the tool kit. I wouldn’t, solely based on numbers, automatically do anything particularly as drastic as a cylinder change. But a bad number, and by bad I mean one that’s close to Lycoming’s published minimums would motivate me to explore further.
 
Eric Pauley said:
Compression tests are military-grade BS. Prove me wrong.:popcorn:
Click to expand...
Simple. Engine compression tests are not a finite, go-no-go test for the serviceability of a cylinder. Those compression readings are merely a benchmark value for that particular time and ambient condition of the engine which require additional interpretation and actions, if needed. Nothing more.

You’ll also find the established guidance states a 60/80 cylinder is still serviceable with some OEM guidance allowing values well below 60/80 for cylinder serviceability. So your 68/80 cylinder is hardly a candidate to be “trashed” or even to be considered “bad” at this point. And your subsequent compression tests proved that.

But at a minimum, at no time should initial compression readings be used on their own to make any serviceability determinations until the recommended re-tests and additional actions are performed. Unfortunately, these results have a history of being misunderstood and have led to a number of un-necessary cylinder removals based on compressions alone. And by association, the same is happening with mis-identified borescope results as well.
 
Compression (leak-down) tests are valuable as a trend. Also cold and hot will give different readings. As I have said elsewhere, I do a compression test at each 50 hour oil change and plot by cylinder. The numbers certainly bounce around, but trends do show up--at least they did in my last O-320. As others have mentioned compression was not the only factor in my decision to replace the old engine. Oil consumption, oil analyses (trends), 2925 hours, increasingly wetter lower plugs, and loss of personal confidence in the engine were all factors.(Yes, I could have gone for top overhaul, but the engine had already been re-manufactured by Lycoming.)

Also test technique plays a part. I have my technique. It is not the same as others' techniques. The easiest illustration is to GENTLY rock the prop and observe the pressure change.If you're not a rocker you will get different results than a rocker.

Due to a stupid accident requiring 7 staples in my scalp, I now pressurize to 40 psi, rock the prop to the maximum reading, then increase to 80 psi inlet pressure. And I only rock using the up blade.
 
Last edited:
Eric Pauley said:
74 68 72 75

with leakage past the rings. Oh no, better prepare to trash that cylinder! I took this info to the seller and took $4k off the purchase price. For reference, this engine had 2800 SFRB. Good deal, enough to replace that bad cylinder!
Click to expand...
Based on this description, it seems to be that you had a deal for at least $4000 more than what the seller was prepared to accept, so he took that off the price to close the deal. I don't know why anyone would consider 68 to be a trash cylinder; I've never heard that. But I would consider an engine at 2800 to probably be in need of overhaul soon and price it accordingly.
 
Sounds more like a lesson learned and a rant, than what you actually believe.

If you were buying a plane and a cylinder had a compression reading of 60, would you say "pfft... compression readings are BS", and move forward without concern?

Willing to bet you wouldn't buy that plane for anything until they dropped the price 4k, or proved it was fine.

At least that's how I'd feel ;)
 
Best readings seem to be obtained immediately after flight although not easy to do.

Cold can vary a lot.

Short ground runs generally wind up inconsistent. Maybe uneven heating?

Auto Tech did these completely different.

No plugs were removed and I think it measured Starter draw.

How many folks pay attention to orifice size?
 
455 Bravo Uniform said:
I guess it’s the same way that people call internal combustion engines “motors”.

This is an actual simple compression test:
View attachment 117421
Click to expand...

While you are correct (obviously), roaming around the aviation community correcting everyone on this is a fool's errand. Yes, plenty of people know you are right, but words and definitions are purely instrumental and in this case calling it a "compression test", while wrong, is what gets you the result you want in aviation.
 
Last edited:
Eric Pauley said:
While you are correct (obviously), roaming around the aviation community correcting everyone on this is a fool's errand. Yes, plenty of people know you are right, but words and definitions are purely instrumental and in this case calling it a "compression test", while wrong, is what gets you the result you want in aviation.
Click to expand...
I now understand where you’re coming from. I’m not trying to correct folks, I’m trying to educate. My satisfaction doesn’t come from showing somebody was wrong, but rather that they learn. I plan to retire in three years, and become a university professor. I’m sorry I came off sounding the way I did. I could’ve taken your all caps response using the word “novel” and the emoji’s as coming off as a smart ass. Gotta love context, or lack there of, in written Internet communications. Fly safe.
 
455 Bravo Uniform said:
I now understand where you’re coming from. I’m not trying to correct folks, I’m trying to educate. My satisfaction doesn’t come from showing somebody was wrong, but rather that they learn. I plan to retire in three years, and become a university professor. I’m sorry I came off sounding the way I did. I could’ve taken your all caps response using the word “novel” and the emoji’s as coming off as a smart ***. Gotta love context, or lack there of, in written Internet communications. Fly safe.
Click to expand...

No worries. This thread has no point but to start smart *** fights anyway! ;)
 
Bell206 said:
No need to. The seller probably had the aircraft over-priced for the 2800hr engine and came out ahead when he only had to give up $4K for one cylinder.;)
Click to expand...

All the people can’t be all right all of the time.

I think Abraham Lincoln said that…
 
Eric Pauley said:
Rant time. Nothing that hasn't been said before.
Click to expand...
How many compression tests have you personally performed? A rant by someone who has no experience is just noise.

The engine compresses air while the piston is moving. A compression test puts pressure againt a stationary piston. Big difference. As Domeick said, the prop must be rocked a bit. That's to get the rings to seat properly against their lands and be in the same position they're in in an actual compression stroke. The restrictor in the compression tester does not pass the volume of air necessary to dislodge the rings from their sticking positions in the cylinder, and they won't be against the lands if the prop was turned backward even a little bit before applying pressure. I have had low readings thousands of times that suddenly fixed themselves when I rocked the prop.

Removing a sparkplug for the test dislodges bits of carbon that can fall onto an open valve and hold it open a bit when it closes. Not common, but rapping the rockers with a soft-face hammer fixes that. Cylinder pressurized for this.

It's not a compression test. That's the auto engine test, done while cranking. It's not a leakdown test, either. That one requires pressurization, then shutting off the air and timing the fall of the pressure.

We are using the differential pressure test, which measures the supply pressure (80 psi) ahead of a defined restriction, and the downstream pressure after the restriction. The difference in the readings tells us how much leakage there is past the rings or valves. More leakage allows more airflow that the restrictor increasingly holds back to create the lower downstream pressure. It's a very accurate test provided that it is done correctly.
 
Dan Thomas said:
How many compression tests have you personally performed? A rant by someone who has no experience is just noise.

The engine compresses air while the piston is moving. A compression test puts pressure againt a stationary piston. Big difference. As Domeick said, the prop must be rocked a bit. That's to get the rings to seat properly against their lands and be in the same position they're in in an actual compression stroke. The restrictor in the compression tester does not pass the volume of air necessary to dislodge the rings from their sticking positions in the cylinder, and they won't be against the lands if the prop was turned backward even a little bit before applying pressure. I have had low readings thousands of times that suddenly fixed themselves when I rocked the prop.

Removing a sparkplug for the test dislodges bits of carbon that can fall onto an open valve and hold it open a bit when it closes. Not common, but rapping the rockers with a soft-face hammer fixes that. Cylinder pressurized for this.

It's not a compression test. That's the auto engine test, done while cranking. It's not a leakdown test, either. That one requires pressurization, then shutting off the air and timing the fall of the pressure.

We are using the differential pressure test, which measures the supply pressure (80 psi) ahead of a defined restriction, and the downstream pressure after the restriction. The difference in the readings tells us how much leakage there is past the rings or valves. More leakage allows more airflow that the restrictor increasingly holds back to create the lower downstream pressure. It's a very accurate test provided that it is done correctly.
Click to expand...

More of nothing that hasn't been said before, I see.
You and @455 Bravo Uniform will have to hash things out on what you want everyone to call this thing. :popcorn:

FWIW there's a little-known organization called the FAA that calls it a compression test: https://www.faa.gov/documentLibrary/media/Advisory_Circular/ac20-105b.pdf
There's also an old sewing machine factory that calls it a "compression check": https://www.lycoming.com/sites/default/files/attachments/Cylinder%20Compression.pdf
 
Eric Pauley said:
More of nothing that hasn't been said before, I see.
You and @455 Bravo Uniform will have to hash things out on what you want everyone to call this thing. :popcorn:

FWIW there's a little-known organization called the FAA that calls it a compression test: https://www.faa.gov/documentLibrary/media/Advisory_Circular/ac20-105b.pdf
There's also an old sewing machine factory that calls it a "compression check": https://www.lycoming.com/sites/default/files/attachments/Cylinder%20Compression.pdf
Click to expand...
But how many have YOU done?
 
Dan Thomas said:
But how many have YOU done?
Click to expand...

0. Zero. I'm also not saying anything that hasn't been said by very experienced mechanics. You don't need to trust me in the slightest because I'm not making any claims.

That being said I don't really see what authority physically performing this test gives you. I've seen trivially disprovable falsehoods espoused by people that have done 1000s of them.
 
When someone has to put their name, reputation and possessions on the line the perspective changes.

Signing that the determination was I/A/W the manufacturers directives helps a lot.

Broken rings will show up here; but they will also put metal into the oil.

A valve that is leaking is likely to burn in time.

Rather find out in flight?

Often owners will push a Tech to let a low reading slide by stating “ It will get better after flight”.

It may; but will YOU fly it when out of Annual?

The smart owner will start the inspection with this task and before the Annual expires.

Just in case!
 
Eric Pauley said:
More of nothing that hasn't been said before, I see.
You and @455 Bravo Uniform will have to hash things out on what you want everyone to call this thing. :popcorn:

FWIW there's a little-known organization called the FAA that calls it a compression test: https://www.faa.gov/documentLibrary/media/Advisory_Circular/ac20-105b.pdf
There's also an old sewing machine factory that calls it a "compression check": https://www.lycoming.com/sites/default/files/attachments/Cylinder%20Compression.pdf
Click to expand...
If it's all been said before, and it's all ********, and you know it, why did you soak the seller for four grand?
 
Lindberg said:
If it's all been said before, and it's all ********, and you know it, why did you soak the seller for four grand?
Click to expand...

Greg Bockelman said:
The seller should already KNOW that.
Click to expand...

Bell206 said:
No need to. The seller probably had the aircraft over-priced for the 2800hr engine and came out ahead when he only had to give up $4K for one cylinder.;)
Click to expand...

Lindberg said:
If it's all been said before, and it's all ********, and you know it, why did you soak the seller for four grand?
Click to expand...

Can't win!! Who soaked whom!? FWIW the seller was a branch of a major legacy US air carrier. I don't lose a lot of sleep over it either way.
 
Another amusing but somehow instantly vitriolic thread about aircraft maintenance.

So I guess the real question is: what A&P made time with your girl/stole your lunch money/peed in your cheerios? :D

I would ditch the <whatever we're agreeing to call the thing with shop air pushed into the spark plug hole, measuring the held pressure, and trying not to get smacked by the propeller while cocking three ears in the direction of exhaust, air filter, and crankcase breather> test in favor of a clever eye with a borescope.
 
43 appendix D…

(d) Each person performing an annual or 100-hour inspection shall inspect (where applicable) components of the engine and nacelle group as follows:

3) Internal engine—for cylinder compression and for metal particles or foreign matter on screens and sump drain plugs. If there is weak cylinder compression, for improper internal condition and improper internal tolerances.
 
Sorry but this rant is complete BS. Experienced mechanics know exactly what a compression test is and what it isn't and yes, we call them "compression" tests and there is no ambiguity or confusion about what that means. Sure you can do a regular cranking compression test but when you get a low number then what? You've got no idea where the leak is so you still have to do the leak-down to figure that out and with that big old propeller sticking out there the leak-down test is easy as hell to do so why not just start there? No need to wear out your starter for an inconclusive test.

Now as for the numbers - think of it more like the canary in a coal mine, a first indicator and not a conclusion but armed with the information it gives you such as how much leakage and where it's coming from you definitely have more knowledge than you would if you just didn't do it. Aside from that what is your proposal to replace this time proven procedure that we've been using for the past 100 years? Please don't tell me borescope because if you stick a scope in there and see visible evidence of a burnt valve or busted ring I can guarantee you from experience that the compression results on that cylinder are gonna be damn near zero. I've had cylinders that had 20/80 results with exhaust valve leakage that even after removal and looking right up the bore you couldn't see anything visible that would convince you that it needed to be pulled off but yet, based on the compression test, it most definitely did.

Don't obsess over the numbers, there are many factors that can effect results and mechanics know what they are. The manufacturers have detailed guidelines on how to proceed if they are below the acceptable standard and in fact that standard is probably way lower than what you're thinking it is.
 
schmookeeg said:
Another amusing but somehow instantly vitriolic thread about aircraft maintenance.

So I guess the real question is: what A&P made time with your girl/stole your lunch money/peed in your cheerios? :D

I would ditch the <whatever we're agreeing to call the thing with shop air pushed into the spark plug hole, measuring the held pressure, and trying not to get smacked by the propeller while cocking three ears in the direction of exhaust, air filter, and crankcase breather> test in favor of a clever eye with a borescope.
Click to expand...

None, but I have been (unsuccessfully) pitched expensive maintenance that would not have improved safety or dispatch reliability. In some cases on the basis of compression test results.

I completely agree on the value of borescoping. Obviously the compression test has value, but its relative importance compared to actually looking in the cylinder has been painfully over-emphasized but far too many people I’ve talked to. The undeniable blind obsession with these numbers throughout piston aviation is the primary thing that I consider BS.
 
You must log in or register to reply here.
Back
Top