New Cylinder AD's released by FAA

[yr right]

Oldeee is it a io 360 tcm. if you like some help pm as mush info as you have and I'll tell you the fix for it.
Cheers

[Creampuff]

But, if you are suggesting that a 5/80 cylinder can run happily then how do you reconcile that with your statement about installation problems causing the cylinder to not make TBO?

My understanding is that the rules require a cylinder with those numbers to be repaired/replaced, because it’s not within specification.

The alternative is to get approval to run the aircraft in the experimental category. My understanding is Mr Atkinson did that with his 5/80 cylinder.

[Walter Atkinson]

That is correct. We were trying to prove (and did prove) that a static compression which would cause one to change the cylinder was not necessarily an indication that the valve was leaking while running. It clearly was not leaking while running. That was all there was to the issue. If one realizes that a static leak does not equal a dynamic leak then this is useful knowledge. If one is not willing to think about this in these terms and insists that ANY leak is a leak and do not care whether or not the problem is real or imagined, then this exercise was useless.

We use compression testing ONLY because for decades it was all we had. It is a suboptimal test and not nearly as accurate as engine monitor data. Eventually the FAA and CASA will catch up and change the rules for the poor mechanic who can't make the call on his own.

I have not, and do not recommend running an exhaust valve AT ALL which is leaking dynamically, it will soon fail. But I do know, intellectually, that just because it is leaking during a very poor test, that it does not mean that it is leaking while running.

"Those who insist that something is impossible, should not get in the way of those who are already doing it."

That is all. You may now return to your regularly scheduled programming.

[jdeakin]

Thanks Old Akro and others for trying to keep this thread on topic

Agreed!

John this is not an APS thread

I'm aware of that, but realize it's terribly difficult to avoid mentioning it from time to time, as that's where most of the research has been done. More correctly, GAMI/TAT does the research, we just teach it.

Yet again you blame the installers for problems with valves etc

YES! BECAUSE THAT'S WHERE THE PROBLEM IS! Factory (predominantly CMI) and overhaulers though, not the person who simply puts them in. Not always, but a large percentage. It is the rule, not the exception, on the failures. Not the overall production, which is pretty good.

The machining that takes place regarding valves is one of the most exacting and difficult tasks in the engine. The valve guide must be straight and true and precisely aligned with the seat, everything must be concentric, and the valve must end up precisely centered in the seat with the contact area exactly equal all the way around. The tiniest mistake in any of that machining will result in a valve destined to fail at 400 to 600 hours.

Take the matter of a valve where the machining process puts the valve microscopically off-center. The valve-to-seat interface will have a contact surface a bit narrower on one side. We used to think that's ok, because it does make contact, it passes a compression check fine, the Kerosene check is fine. A couple hundred hours, it works well. But all the time, the valve is NOT EQUALLY COOLED AROUND THE CIRCUMFERENCE, because most of the heat passes through the valve seat. Eventually, the unequal cooling causes the valve to actually begin to warp, and at 400 to 600 hours (approx) a tiny bit of the valve no longer seals properly, and now a bit of the 3500 degree combustion event can partially leak out during the time the valve should be sealed. At that point, the poor valve hasn't long to go. It takes about 75 hours of operation from the first sign of this to failure. We have borescope pictures, EMS data, pilot descriptions, and the remaining parts from these failures. All the way from first indications and prophylactic removal to failure and engine destruction.

This is just one of the failures, there are others.

We don't require your email address

I'm aware of that. I post that in case someone wants to contact me outside the forum (special format to evade the internet 'bots that collect addresses for spam). You don't require real names, either, more's the pity. It would be much more polite forum if you did.

Are you a sysop or forum administrator?

John Deakin

[jdeakin]

I have a suspicion that we have one cylinder location (TSIO 360) that causes a problem. (Its a recent hypothesis, the logs are 100km away and I haven't been back to check). I have assumed that this is a cylinder whose mixture is out of step with the others (no, the aeroplane doesn't have an analyser - yet). If this is the case, it could play straight into your "red box" area.

Yes, both cylinder position and unbalanced fuel flow can cause that. Baffling will help the first, GAMIjectors often solve the latter. An EMS will tell all (or nearly all.) That's the true value of an EMS. Not leaning.

I have looked at a number of removed cylinders and most have sunken valve seats.

AIRCRAFT ENGINES????? CMI AND LYCOMING???? Please, tell me more! You've got my full attention! I've read old reports of that, but those are due to either "soft seats" or timing issues related to below standard octane fuel. We should never see that today!

This gets us back to a point you made on the previous closed thread about an engine running happily with a 5/80 compression. Clearly you understand that the valve springs are there to make the valves follow the camshaft and not seal the valve. The cylinder pressure does that.

True. I don't know how long Walter did that, so I don't how "happy" the engine would have been in the long term.

When I was young and building engines, I'd grind valves until my hands were sore.

Thanks for the memories - NOT! When I was a tadpole (15?) and a lineboy, the boss would make me "lap valves" just as you describe. I felt like an Indian, trying to start a fire! Only thing on fire were my hands. An evil chore.

Now the shop next door does all the head work (including porting & valve seats) on a 5-axis computer controlled mill. He says its better to leave the inlets a bit rough and that grinding paste does more harm than good. sigh.

One brand name is "Surda?" Spelling may be off, that's phonetic. I'm unsure why, but that may be the cause of our problems. I'd like to visit one of the shops that have a nice little sideline business of correcting new cylinders, and find out more.

But, if you are suggesting that a 5/80 cylinder can run happily then how do you reconcile that with your statement about installation problems causing the cylinder to not make TBO?

It ran 5/80 for some time, but NOT 400 hours, and strictly as a test. I've answered the rest in a message just before this one.

The other reason we have had cylinders replaced is that they "barrel" ie, the bore gets bigger in the middle. I had assumed that this condition could be improved by more attention to cooling management. Are you suggesting that it might be just a material issue and the luck of the draw?

I'd take a wild guess that your intial assumption is correct. Heat and Time are the enemies of our engines. The more we see, the more we like 380℉ as an operating maximum for long periods. We've got a LOT of data that shows these engines will take redline temps for short periods, even to 500℉ with no apparent damage, and go on to TBO. But at and above 500℉ we invariably see damage in the very short term, just a few minutes. 380, 400, maybe 420 seem to be the dividing line.

John Deakin

[No Hoper]

John D
No not doing mods job, you are making valued contributions here and would be a pity to lose you from the thread.
Reference valve and seat installation, are you suggesting that it is the overhaul shops not using the correct technique or the OEM?
CASA has given advice to the industry through ADs and AWBs about boroscope evaluation etc to assess cylinder condition when compression is below the 60/80 differential.
I have seen comp as low as 20/80 improved to an acceptable level by running the engine and rechecking.
Low compression from blow by can be simply a matter of the ring gaps aligning, well that was one theory.

[ericferret]

OLD AKRO

While a touch of thread drift is involved with this I would add the following story to your compression comments.

While carrying out diff comp checks on a Lycoming 0-320 installed on an R22 helicopter that had just flown in with no reported defects I was confronted with
0/80, 0/80, 0/80, 0/80.

Obviously the compression tester was u/s so I used a second one. 0/80, 0/80 e.t.c.

Obviously I was having a brain storm so I had a second engineer check the engine. Same result.

We dropped the exhaust and watched the exhaust valves closing and it was apparent that at slow rotation the valves were hanging in their guides due to excessive guide wear and with no inertia were not seating properly. That this amount of wear had no apparent effect on the engine performance when running was startling, a helicopter hovering at high gross weight will normally reveal any engine abnormalities.

[jdeakin]

Reference valve and seat installation, are you suggesting that it is the overhaul shops not using the correct technique or the OEM?

I think both. I think once in every 10 or 20 cylinders "something happens" to make one out of limits in one way or another. Perhaps a compounded error, where both parts are within tolerances, but when assembled together they're not. I don't honestly know, but we didn't see much of this on cylinders manufactured before about 1991. CMI (TCM then) had a strike, where managers went down on the floor, and it was also about then that some machinery upgrades were done. Some very old machinery that had been doing the work right might have gotten replaced by "modern" machinery that wasn't so good. All supposition, but the results are clear. Efforts to blame LOP are nonsense.

CASA has given advice to the industry through ADs and AWBs about boroscope evaluation etc to assess cylinder condition when compression is below the 60/80 differential.

Depending on how that's implemented, it's probably a good thing! Borescopes have been around a long time, but it's only in the last few years they've been cheap enough for even owners to have. Under $200, with digital picture-taking ability direct to a computer. If a shop or a mechanic doesn't have one, I'd look for another mechanic.

I have seen comp as low as 20/80 improved to an acceptable level by running the engine and rechecking.
Low compression from blow by can be simply a matter of the ring gaps aligning, well that was one theory.

YES! One of our slides shows compressions all normal (rough engine). A second compression check was done after flipping the prop through few times, and that test was 0/80. Sticky valve, of course, sticking on some strokes, but not others. A single compression check should NEVER be taken at face value. Always borescope it, if that's good, run or fly the engine, or fly it hard, and check it again. CMI did good, with the latest SB. It's now obvious that we've all pulled far too many cylinders based on compression alone for the last 100 years!

John Deakin

[Old Akro]

CASA has given advice to the industry through ADs and AWBs about boroscope evaluation etc to assess cylinder condition when compression is below the 60/80 differential.

I have a very high grade borescope. We have used it for a number of automotive investigations (mainly catalytic converters and chassis cracks inside a box section). I have fooled around with it with aircraft engines, both complete engines and (with my favourite LAME), diassembled engines so that I can look through the borescope then remove the pot and look directly inside to compare.

In my opinion it requires a very high degree of skill & proficiency to diagnose developing issues with a borescope. If you have part of a valve missing, its pretty easy - but you don't really need a borescope for that. I'm prepared to bet that the guy at CASA who wrote this has never done it. I'm interested to hear from Walter or John, but at 60/80 I don't believe you are likely to get any useful information from a borescope.

And back on the compression test, the ones I have witnessed rarely seem to get the same number twice. LAME's do it by feel a bit, rocking the prop until they get a number they like. Any test that is this inconsistent is questionable. Its also highly dependant on piston ring orientation. Like yr right's valve seats, piston rings rotate also. Every now and then all the piston ring gaps line up and the compression goes to hell. Run the engine or just leave it till next time and you get a completely different number.

[Old Akro]

AIRCRAFT ENGINES????? CMI AND LYCOMING???? Please, tell me more! You've got my full attention! I've read old reports of that, but those are due to either "soft seats" or timing issues related to below standard octane fuel. We should never see that today!

The first 100 hourly after I bought into the Seneca we replaced 6 pots. I wasn't part of the decision and I was new to the joys of Continentals. But, I took all the old pots back to my workshop and took them next door to my engine guy. Both the seats on the valve looked horrible and worn with the contact part of the seat worn into the valve and the seat in the head looked similar. With the valves in place they looked "sunken". Unfortunately, I've thrown them all out now.

At the time, I had quite a lengthy discussion about valve seat design and we compared the valve seat geometry with a number of modern car engines. It was quite interesting. The view was that the CMI design is quite old fashioned and not as effective at heat dissipation as modern designs. The CMI design was as I remember valve seats when I used to hand lap them. Angular and geometric, whereas now they tend to be more like compound curves.

This aeroplane has had the same owners since 1985 and (from memory it has done 2 engine rebuilds in that time). No-one remembers what brand these cylinders were. It may be that they were a bad batch. I have a recollection that CMI cylinders were not available and that they were another brand. We replaced them with CMI cylinders.

BTW, the 5 axis CNC machine is made by Centronic in the USA. Its spooky to watch it doing a cylinder head. It will change the cylinder head shape, do a port job, skim the head and cut valve seats all in one go.

One of the advantages of race engines is that they get torn down a lot. So its easy to try a change, then see how it performs.

I spent some time with a drag race guy a year or so ago. The drag guys use no data logging at all. They don't need to because the pull the engine down after every race and they can look directly at the parts. They also use the almost forgotten art of "plug cuts" and reading insulator colour to set mixture.

[No Hoper]

Old Akro
The engine is "pumped up" with cylinder pressurized to ensure the rings are seated. Also ensures piston is at TDC as this affects comp as well.
Boroscope is an art but not that difficult on piston engines, more skill required for turbine engine inspections though.
Do you know the chap who wrote the AWB?

[Walter Atkinson]

I ran the low compression cylinder for over 250 hours. The only reason I stopped was that the engine was overhauled (not for that reason).

It is pretty easy and straightforward to see a valve that is leaking while the engine is running through a borescope (not using the borescope while it's running!!). A heathy valves will have a very even, donut appearance to it where the heat is evenly distributed as it leaves the valve for the head. A leaking valve will show a very definite change in the circular pattern. CMI (TCM) put out a SB on this with pictures, etc.

ANYONE can see this change with even a cheap borescope. Any LAME who does not have a borescope should get one-- they are very inexpensive and an invaluable tool. There are many A&Ps in the US who now take a quick peak at the valves every time they gap the plugs. We are not removing cylinders nearly as often as we used to as a result. According to CMI's SB, compression below 60/80 is no reason to remove the valve unless the borescope shows the burning pattern. In addition, as the OEMs catch up to the A&P/pilot population's knowledge about how to use the engine monitor for this (it is even better) we will likely begin to see recommendations that we have been teaching for over 14 years adopted about how to determine whether or not a valve is leaking form inside the cockpit during flight. We've been doing this quite accurately and successfully for well over a decade.

We will be happy to have the OEMs, engine builders and FAA join the party of knowledge. More and more are doing so.

Our recommendation:
1) NEVER run a cylinder with a known leaking exhaust valve--it is soon to fail.
2) Verify this with a borescope or engine monitor data and, if it is NOT leaking, no need to stop running it.

The above has been our recommendation for well over a decade, so those who have been misquoting my recommendation might want to read the above again.

[yr right]

OK lets have a think here'


We do a leak rate test iaw manufactures procedure's lyc and tcm are different.


They say no leak past the valves is excepted period.
Regulators enforce this in LAW


Now you have four places for a leak to occur
Rings evident via the breather for a leak
Exhaust valve via the exhaust
Inlet valve via the inlet track
Crack head via a leak in the head


All give a different result. And there are techniques that we use when a valve is leaking to try and get it to reseat before we remove a cly. There are also limits via SB for the wear on a valve guild. A worn guild can lead to a dropped valve due to uneven loading on the valve during closing
.
Next and this is important.
We do this test at the end of a m/r and before issue of a new m/r unless there is a problem the engine may develop a problem 1 hour into a m/r our 1 before the end of the m/r.


Now they ran a 5/80 after finding it. So how long was it running before this was found and in the other thread they said how long they ran it for.
Now it can be safely assumed that they where trending the engine before this had taken place and it wasn't pick up . So why not
Instrument wasn't calibrated
Instrument wasn't fitted
They didn't know how to read the instrument
The instrument didn't pick the defect up
What did the instrument say when they knew of the defect
or a small bit of carbon under the seat
Did you do retesting during this period
And the list can continue.


Now this leads me into thinking why or why would you run the risk maybe to yourself but the risk to others. And unless you flying over nowhere like here you pose a risk to someone on the ground however small that risk is.
So what was the wrong with the cly in the end did you change it or just leave it for Devine intervention or send it to the healing room.


Now despite them saying a leak rate tester is not a really good tool its a very powerful tool and used correctly can and will lead to stop an engine failure.
Boros are a great tool even the little cheap ebay one I had a selection of small mirrors now I don't, I ha at my disposal a $1000 one to a $75k one all singing all dancing But the leak rate is still a better tool to see if a valve is leaking.


Cheers

[yr right]

Sorry forgot to add
Only leaks passed the ring is allowed.
Cheers

[No Hoper]

Yr Right
Can't fault your what you just posted, except to add that with leaks I have also found it benificial to throw some soapy water at the cylinders, on two or three occasions have found cracks around the head as well as ring leaks

[yr right]

Yeah soap in water to find bubbles just didn't put it all in sorry. I know tcm say to do that in there SB as we'll
Cheers

[Walter Atkinson]

Now they ran a 5/80 after finding it. So how long was it running before this was found and in the other thread they said how long they ran it for.

Watched it descend below 60/80 and kept an eye on it every 25-50 hours.


Now it can be safely assumed that they where trending the engine before this had taken place and it wasn't pick up . So why not

It was picked up.


Instrument wasn't calibrated

Yes it was.


Instrument wasn't fitted

Yes it was.


They didn't know how to read the instrument

Yes we do.


The instrument didn't pick the defect up

Yes it did.


What did the instrument say when they knew of the defect

The correct values.


or a small bit of carbon under the seat

No.


Did you do retesting during this period

Constantly. Every 25-50 hours.


And the list can continue.

No it can't. We were doing research. We were very careful. We learned a lot. Sorry you don't "get this."

[yr right]

Oldee I was a drag racer. We even made our own fuel injection. Now whilst a lot dont use data loggers a lot do. That's why you can have a N/A 500 ci suspension car running faster than nitro TF cars in the earlier 80s
And btw I also can pretty much tell you how to fix your plane. Is it a turbo model ?

Cheers

[yr right]

So research then hey. We'll let's look at that again. You claim data I've already said it's not data it's obverations. Now 5/80 is not research at all. Un less it carried out on a calibrated dyno how can you ever say what power the cly is producing.
Compression = power
A low cly will not make as much power as a cly running at 70/80. That's why we have limits. That's why tcm and lyc have different methods of performing a lake rate test. And different tesster one with a calabrated orifice and one with out. You can use a tcm on a lyc but you can't use one that has not got a cal with out useing the external orifice on a tcm. Why is that then.
And whist a low comp generally will not be picked up by a pilot it will how ever be picked up at
So now this low cly all thought making power isn't making what it should be making.
The whole reason for doing this is to make the engine safe. So now what is your option of an un safe engine.
And of course it's all in the name of research but I'm sorry I don't subscribe to this theory as I have people's life's in my hands and I take that responsabity extremly seriously.
And now you are say I give miss quotes we'll no you given out dangerous information
Cheers

[yr right]

I also said at the start of this thread that the AD from what I read seam quite a normal response to a problem that has occurred. Then we going the Nstb report which supported the action of the FAA but did say we'll maybe not all the cly need to be recalled.
Everyone said what about Oem cly and I said compare apples to apples.
It would appear what I said in the start to be true.
Cheers