Talk to me about compression testing.

[alfadog]

Or cylinder leak-down testing, if you prefer.

So I consider myself a fairly green A&P. I've got now about 2 years under my belt of actual twirling wrenches and bucking rivets. For a while when I was doing a compression test, I would position the prop at what felt like top dead center against the 80 psi and take what I got as being the number. Meaning that I was kind of halfway between when it would snatch forward and when it would snatch backwards. But I found that I was getting a few low numbers but if I did it a different way I would get a better number. The different way being I would see what position of the prop in that narrow band before the pressure in the cylinder moves the prop one way or another, gave me the highest reading on the right-hand gauge. Meaning that I would rotate the prop in the direction of normal rotation and if I found a position where the prop would hold and the pressure was its highest, then maybe I should be using that number.

Question one, does that sound right to you? If not, how would you go about doing it differently?

Which brings us to question two. So, on the last engine I did a compression check on a few days ago, one cylinder would only give a decent reading when it was just about ready to snatch the prop forward. Like the slightest vibration and the prop would shoot forward. It was not in that sweet spot but was really far forward. Did I mention that it was far forward? So I took that number based on my thinking that that's how I should go about it. What could have caused it to act that way? A very small ridge or some rust in the cylinder at TDC or a sticky ring? This airplane has been sitting for awhile.

 

[Bell206]

This airplane has been sitting for awhile.

Normally you wouldn't perform a cold differential test on an engine that had been "sitting" for any length of time.

I always recommend to those who have limited experience on doing this test to follow the OEM procedures. Then after some experience is gained modify as they see fit.
https://www.jerrytemple.com/assets/Uploads/templesTips-page/pdf/Continental-Service-Bulletin.pdf
https://www.lycoming.com/sites/default/files/Cylinder Compression.pdf

 

[alfadog]

Normally you wouldn't perform a cold differential test on an engine that had been "sitting" for any length of time.

...

The engine is not cold. It runs fine and was warmed up before testing. The airplane has not been flown in a while, that is what I mean by "sitting".

Yes, I am looking at the Lycoming SB. It does not seem to preclude what I am doing to get the best number. IMO.

 

[Tom-D]

do the test when cold, or at normal run temps.
When I'm trying to find a cylinder that has morning sickness, I do them cold, because that is when the valve sticks open.

When the owner is tracking a trend, always do the test in the same manor.

keep in mind, the mag impulse snaps at TDC, that SHOULD be very close to where you need to be when adding pressure to the cylinder.

never allow the engine to rotate with full pressure applied.

 

[Clip4]

I am not an AP, but I sure have watched a lot of AP mechanics performing this test the way you describe. It always seemed like a bit of cheating, but honestly this test does not accurately simulate the operating temperature and what is occurring with the rings with the engine running. It is a static test.

 

[Tom-D]

I am not an AP, but I sure have watched a lot of AP mechanics performing this test the way you describe. It always seemed like a bit of cheating, but honestly this test does not accurately simulate the operating temperature and what is occurring with the rings with the engine running. It is a static test.

Have you ever seen a compression test done where each cylinder has a gauge and the test is done at idle? the engine is run one one spark plug.

 

[weirdjim]

At the risk of being burned at the heretical stake ...

I can tell more by the old "automotive" compression tester where you put an absolute pressure gauge in each cylinder and run the starter for about 6 or 7 revs. Of course, my clients want the "standard" 80/xx test in the logbook because they think that is the "gold standard" of engine performance. T'aint.

Call me being cynical, but if you had two (or three, or more) ways of telling what the health of a cylinder was, and you had your choice of half a dozen ways of doing it, and you were selling cylinder parts, which one would you choose?

Jim

 

[Clip4]

Have you ever seen a compression test done where each cylinder has a gauge and the test is done at idle? the engine is run one one spark plug.

Cars plenty of times, but not on an aircraft. Why the FAA does not require a dynamic test??

 

[BillTIZ]

Cars plenty of times, but not on an aircraft. Why the FAA does not require a dynamic test??

Something about being that close to a spinning blade.

 

[Checkout_my_Six]

I'd rather just have a look see.....

 

[Clip4]

Something about being that close to a spinning blade.

Turn the indicator facing aft.

 

[Tom-D]

Something about being that close to a spinning blade.

Never a danger, remove cowl, install all gauges in the one of the spark plug holes (they were threaded) clear the area, go start the engine and run at idle, shut down, have a look at gauges.

they will register the BMEP at idle, or what we know as dynamic compression. (nearly)

doing the compression test the way we do on our engines on a big bore radial with 28 cylinders would take a week

 

[Dan Thomas]

Or cylinder leak-down testing, if you prefer.

So I consider myself a fairly green A&P. I've got now about 2 years under my belt of actual twirling wrenches and bucking rivets. For a while when I was doing a compression test, I would position the prop at what felt like top dead center against the 80 psi and take what I got as being the number. Meaning that I was kind of halfway between when it would snatch forward and when it would snatch backwards. But I found that I was getting a few low numbers but if I did it a different way I would get a better number. The different way being I would see what position of the prop in that narrow band before the pressure in the cylinder moves the prop one way or another, gave me the highest reading on the right-hand gauge. Meaning that I would rotate the prop in the direction of normal rotation and if I found a position where the prop would hold and the pressure was its highest, then maybe I should be using that number.

Question one, does that sound right to you? If not, how would you go about doing it differently?

Which brings us to question two. So, on the last engine I did a compression check on a few days ago, one cylinder would only give a decent reading when it was just about ready to snatch the prop forward. Like the slightest vibration and the prop would shoot forward. It was not in that sweet spot but was really far forward. Did I mention that it was far forward? So I took that number based on my thinking that that's how I should go about it. What could have caused it to act that way? A very small ridge or some rust in the cylinder at TDC or a sticky ring? This airplane has been sitting for awhile.

You got it. Moving the prop a bit coaxes the rings to seat better. As the piston came up to TDC there was no pressure on the rings so they weren't firmly seated on the piston lands. Might even be a few bits of carbon loose in there. Moving the prop with the cylinder under some pressure (doesn't have to be 80; move it at 40, maybe, then finish the pressure increase) will let the rings shift a bit to where they would actually be when the engine is running on a compression stroke.

I built my own compression tester. It has a squeeze-handle operated regulator, and I can bring the pressure up with one hand. If anything starts going wrong, just releasing the thing exhausts the pressure instantly. Never patented it; patents cost money. And it would cost a lot more than the typical cobbled-together affairs you buy from the aviation tool guys. It's way more complicated and involves a lot of machining.

 

[Kenny Phillips]

What so many here are calling a compression test should more properly be called called a leak-down test. (I still have my Snap-On tester, with its propeller warning!)
"Leak-down" is the more accurate term; a compression test is generally defined as putting a gauge in a plug hole, then cranking the engine and noting the highest reading. We always did both, and with the engine warm, if possible. As one is dynamic, you can get some information from both, and get a good idea of the condition of the rings and valves.

 

[airdale]

From race engine experience, not airplanes: It seems you are dealing with two variables in your measurements: Where you are relative to TDC and whether your last piston motion was downwards/away from TDC or upwards/towards TDC.

With the race engines, I made a "TDC finder" by welding a nut into a spark plug housing, then having a threaded rod secured by a jam nut poking into the cylinder. The length of the rod is not critical; its job is to touch the piston top a few degrees before or after TDC. Then simply mark the the touch points on the ring gear relative to some handy reference, and add a mark halfway in between them. That is TDC. (Valves closed of course). This takes longer to explain than it does to do.

Now you can turn the engine exactly to TDC with the piston moving upwards. You have eliminated both of your variables. You may learn something from that.

Re pressure on the rings, the air should take care of that ,especially since that upward piston movement seats them in the crank side of the grooves the same way they would be seated on the compression stroke when the engine is running.

 

[Tom-D]

Now you can turn the engine exactly to TDC with the piston moving upwards. You have eliminated both of your variables. You may learn something from that.

Bringing the piston to TDC with a low pressure in the cylinder, you will know exactly when TDC is reached, and the piston will stay at that position with out any help from some one hanging on to it. then simply stay clear of it, and bring the pressure up to 80 psi

 

[Dan Thomas]

What so many here are calling a compression test should more properly be called called a leak-down test. (I still have my Snap-On tester, with its propeller warning!)
"Leak-down" is the more accurate term; a compression test is generally defined as putting a gauge in a plug hole, then cranking the engine and noting the highest reading. We always did both, and with the engine warm, if possible. As one is dynamic, you can get some information from both, and get a good idea of the condition of the rings and valves.

That's the "differential compression test." A leakdown test involves putting compressed air in the cylinder and closing the inlet valve, then measuring the time it takes for the pressure to leak down to a specified pressure.

 

[Silvaire]

I had a cylinder like that on a Cherokee that had been sitting for a little over a year. Compression was around 65 but as I rocked the prop looking for that sweet spot it would drop to 40 right at TDC. There was what looked like corrosion on the top of the cylinder walls. Ran the engine and took it for a short flight, afterwards it was normal with a mid 70's reading at TDC. Eventually though that cylinder had to get replaced.

Doing a differential compression test gives you more information than just doing a cranking compression test and on an airplane it's easy to do because you can just turn the prop by hand to position it where you need it. If you do a cranking compression test and have a low number you're still gonna have to do a differential to find out why, so may as well just cut to the chase. Doing a differential or "leak down" on a car is a pita because you have to get on the crank pulley with a wrench and figure out where TDC is for each cylinder and how to hold it in place.

 

[Tom-D]

If you do a cranking compression test and have a low number you're still gonna have to do a differential to find out why,

Oh hell no,, knowing how dynamic compression works, if you see a running compression that is effecting the BMEP. you'll probably not even get a reading on the static compression gauge.

 

[UngaWunga]

Using a rod in the spark plug hole isn't a perfect method to find TDC. You'll get it withing +/-5deg or so due to the piston hanging out/floating near TDC when the crankshaft big end and connecting rod are in line. The best way to find TDC is the positive stop method with a degree wheel, but turning an aircraft engine backwards seems to be a no-no. Could this be why ignition timing tends to be off on aircraft engines?

Anyways, leakdown tests in motorcycle and car engines gives you a better view of their health, and should hold true for aircraft engines as well.

 

[airdale]

Using a rod in the spark plug hole isn't a perfect method to find TDC. You'll get it withing +/-5deg or so due to the piston hanging out/floating near TDC ...

Not to argue, just to clarify. The rod is not used to find TDC. It is used to find points on either side of TDC where the piston is the same distance before and after TDC. It is actually most accurate when the piston is 90deg from TDC (because that is when the piston if moving at maximum speed per degree) but to do that the rod needs to be too long. I have never looked to see how many degrees either side I am using but counting ring gear teeth would be an easy way to figure it if one cared. The actual degrees from TDC doesn't matter either. You just mark the two points on the ring gear, then mark halfway between them. That is TDC to a very high degree of accuracy.

No degree wheel needed and there is no issue with the piston floating because it always approaches the measuring rod from below and never gets near TDC. As far as turning the engine backwards is concerned, if that is prohibited then this method won't work.

 

[Tom-D]

These rods that are mentioned are called "piston stops" there are stops made for each type of engine because of the spark plug hole location differences.

use the wrong one, get a wrong result.

 

[Checkout_my_Six]

really?....guess I got the wrong one.

 

[Tom-D]

really?....guess I got the wrong one.

Probably.

 

[Checkout_my_Six]

Probably.

Model E-25 TIMING INDICATOR
INSTRUCTIONS
1.) Remove all top spark plugs. Install piston stop into no. 1 cylinder top spark plug hole.
2.) Install timing disc indicator on aircraft propeller spinner or hub using the supplied
rubber bands.
3.) Turn propeller slowly in direction of rotation until piston lightly touches piston stop.
4.) Rotate disc of timing indicator (while preventing the aluminum flower pot from turning)
until Top Center (T.C.) mark is under the point of the weighted pendulum pointer.
5.) Slowly turn propeller in opposite direction until piston again lightly touches piston stop.Observe reading on the disc under the pointer and rotate the disc (whi
le preventing the aluminum flower pot from turning) to exactly one half of the number of degrees toward the Top Center (T.C.) mark.
6.) Remove piston stop from cylinder and find the compression stroke of no. 1 cylinder by placing finger over spark plug hole and rotating propeller until blow by is felt by finger, continue rotation until pointer is under T.C. You have now found top center on the compression stroke.
7.) To either check the magneto timing or to time the magnetos to the
engine, move the propeller opposite direction of rotation past the specified magneto timing setting and then back in direction of rotation until the desired setting before top center is under the pointer (this removes the factor of gear back lash).
8.) The breaker points should just be starting to open at this setting. Breaker points
opening should be checked with an Eastern Model E-50 Synchronizer.
Note:
A.) A light tap of the finger against the dial will insure absolute correct timing.
B.) An occasional drop of oil at the pointer bushing is all that’s required to keep the timer in good operating condition.

 

[Tom-D]

Have you tried it in your P&W 985?

 

[alfadog]

Have you tried it in your P&W 985?

I used the TimeRite

 

[Tom-D]

I used the TimeRite

I do Too.

But here is a question, How do you compensate when dealing with auto advanced mags.

 

[alfadog]

I do Too.

But here is a question, How do you compensate when dealing with auto advanced mags.

You got me there. What I'm really good at, is reading and understanding maintenance manuals so if I run into one like that, I guess that's what I'll have to do.

 

[Tom-D]

You got me there. What I'm really good at, is reading and understanding maintenance manuals so if I run into one like that, I guess that's what I'll have to do.

me too, I had to call Al at Savage magneto to get my Warner timed correctly.

 

[JAWS]

These rods that are mentioned are called "piston stops" there are stops made for each type of engine because of the spark plug hole location differences.

use the wrong one, get a wrong result.

A time rite is not the same thing as the flower pot.

 

[Tom-D]

A time rite is not the same thing as the flower pot.

We know, one is for timing your radial, the other is for growing your weed.

 

[Tom-D]

my Time rite and Warner piston stop. If you need them, they can be bought.

 

[Bell206]

my Time rite

I wish I still had my original Time Rite with the wooden box. I think it had every rod and scale made. Old mechanic gave it to me and was one of those things that unfortunately disappeared out the hangar one day never to return.

 

[Tom-D]

I wish I still had my original Time Rite with the wooden box. I think it had every rod and scale made. Old mechanic gave it to me and was one of those things that unfortunately disappeared out the hangar one day never to return.

This is what I have collected over the years as I needed it.
It has all rods made, and all scales plus the new version instructions and the original

 

[Dan Thomas]

I've used my digital inclinometer, taped to a prop blade, to find TDC using the threaded piston stop. Much more accurate than the flower-pot protractor and its sticky needle, but the inclinometer takes a bit of thinking through as you use it.

 

[Checkout_my_Six]

You can get a free iPhone app to do that with the piston stop to find tdc.

 

[Dan Thomas]

You can get a free iPhone app to do that with the piston stop to find tdc.

Got that, too.