Engine Trouble

[Mark Dickens]

Thanks for your kind words everyone.

Whee - the engineers noticed a contaminant on the cylinder flanges too. The #4 & #5 cylinders have been taken to South Air in Dunedin for inspection and crack testing. They will decide if they can go back on the engine.

Mark - the engine has 361 Tachometer and 458 Hobbs. I spoke to Bob yesterday and he was very helpful. Bob said they use a torque of 60 ftlbs on the large studs and 30 ftlbs on the smaller ones. He felt the Lycoming specc'd torque of 50/25 ftlbs was not quite enough.

bkyser - some of the remaining nuts were loose, but the studs/ through bolts were all tight. We had to remove the #5 cylinder to extract one of the through bolts. We found it tight and secure, but it requires replacement due to the #4 (left side) end being sheared.

The oil filter is being removed and checked today, and the remaining cylinders are being Boroscoped.

Not to start a criticism of Bob, but if your logbook entry has the words "as per the Lycoming Overhaul Manual", and he's knowingly over-torquing the cylinder hold down studs, yeah, I think that's an issue. Lycoming is the authority here if he's rebuilding an old Lycoming engine.

[Bissetg]

Agreed, if your logbook entry says that then you should be able to rely on the Lycoming standard being met.

[Battson]

Hold on, does the log book entry have those words?

[Mark Dickens]

Not sure if you're asking me, but yes, my logbook from Bob has those words

[Bissetg]

Further to this conversation I’ve asked a retired engineer to explain how bolt torque works with respect to engines.

I am unqualified and keen to learn and understand, hopefully a qualified forum member can confirm or correct his view.

His position is that engine bolt torque is critical. Reason being that the bolts and the case and cylinder flanges are made of different materials and therefore have different expansion rates so when your engine gets hot the tensile load can be amplified pinging the heads off the studs.

The combination of the increased tensile load caused by the heat expansion and the reduction of tensile strength of the bolt due to thermal material strength degradation poses a significant risk.

That all sounds logical but as an unqualified and uneducated metallurgical chap like myself it would be good to have an experts view.

[Battson]

Not quite right

[Battson]

It's quite unintuitive. Bolts experience stress in an interesting way.

If a bolted connection is torqued up tight, so the preload is higher than the stresses the bolts are experiencing, then there's is very little cyclic stress or strain inside the bolt when the stresses are applied and removed (generalizing dangerously to make it easy to explain). This means the bolt doesn't really experience stress cycles to the same degree. Its almost like a properly torqued bolt is immune to stress, up to a certain point.

As you can imagine, cyclic fatigue is a huge factor in internal combustion engines.

As preload in the bolts (or studs and nuts) is lost, they experience stress and strain with each firing of the engine, this is a huge problem. From this point onwards, failure is a ticking clock, depending on the largest "critical flaw" (biggest micro crack) leftover from manufacturing. Once the stress / strain cycles reach tens or hundreds of millions, the cracks will have grown large - to the point the bolts cannot hold the stress any longer, and they snap.

At some point during this process, the remaining fasteners may also be overloaded, and the problem spreads.

[Nev]

Hold on, does the log book entry have those words?

Here's the actual wording on mine:

6A3CF77C-2059-4EF1-9941-FCA5297A557E.jpg

[Bissetg]

My experimental engine Logbook entry was similar, new cylinders etc, except they weren’t.

The important thing in my mind is that we need to be vigilant and keep a close eye on our engines. Certified engines can fail too.

Knowing what to look for and what the warning signs are is important to us all so please keep sharing the info!

[Bcone1381]

The important thing in my mind is that we need to be vigilant and keep a close eye on our engines. Certified engines can fail too.

Knowing what to look for and what the warning signs are is important to us all so please keep sharing the info!

How can we make this easy to access, and convenient. What items should be on the list?

To be specific in the case of broken studs. These probably failed because Bolts are elastic and the ignition/power stroke event places them in a high tension 40 times a second. I think The stretch - compress stops with a clamping pressure higher than what the compression forces are on the joint or fastener. so i think if they are stretched a bit every event due to a sealant on the cylinder base or a less than adequate clampdown by inadequate torque on the nuts, then they fail due to fatigue.

So, how can I see if mine are torqued properly? I think I need a special tool for my stud bolts that are like Nev's in post #1. Edit. Here is a link to Narrow Deck Cylinder wrench. Will the failure mode of hardware that is not torqued to spec just stretch the and cycle and fatique itself to death without giving any warning? Will Torque Seal indicate a loosening of the bolt?

Simple checks like that and understanding failure modes do more than just make us feel good.

Edit. Here is a link to Narrow Deck Cylinder wrench. https://www.aircraft-tool.com/Detail?id=6494-SET

[Bcone1381]

Edit. Here is a link to Narrow Deck Cylinder wrench.

https://www.aircraft-tool.com/Detail?id=6494-SET

[Steve W]

I built my 0-360 angle valve 6-7 years ago and recently had reason to torque the case nuts. 60 foot pounds was in the hard drive of memory. I think the AME who helped me torque the rod bolts mentioned to use 60 lbs for the case.
I went through my OH manuals and couldn’t find anything that indicated more than 50 ft lbs for the cylinder base nuts and nothing on line researching Lycoming SI’s. It’s obviously something being done in the field without mention from the manufacturer.
Parts being used for experimental engines CAN have an abundance of unknown hours. During OH, the through bolts are suppose to be replaced but there is no requirement for the short case studs to be changed out. I believe these studs only have a life of X amount of hours considering the cycles of stress. Engines for their first or second OH should be safe but there are parts out there with over 10k of hours accumulated and nothing more is considered about the high cycled studs being put back into service. My 2 cents worth…..

[Battson]

Will the failure mode of hardware that is not torqued to spec just stretch the and cycle and fatique itself to death without giving any warning?
Will Torque Seal indicate a loosening of the bolt?

Yes, and no, respectively.

Yes: Short of using NDT techniques, you aren't going to find cracks in a torqued in-service bolt unless the bolt fails, or you regularly disassemble and look for them - which is inadvisable, in this case.
No: The bolts don't have to rotate to loosen, and rotating doesn't always mean they got looser (although it would probably indicate they are loose!)

Retorquing a bolt (or stud and nut) that has already started to literally "stretch" (i.e. plastic deformation) or crack doesn't help your situation either, in fact it will probably make it worse. Once the cracking process has started and the crack has grown a meaningful amount, you'd need to reduce the stresses to stop it continuing to grow (which isn't going to happen in this use-case).

As far as I can tell, the main thing is getting the engine assembled correctly in the first place. There is a lot to know here, in terms of doing it correctly, so I won't dare to summarise the Lycoming manuals. But you need to know what you're doing e.g. dry vs wet torque, etc., which is why I choose to get my engine serviced exclusively by experienced certified professionals. I don't know enough about it.

Beyond that, I think looking for the signs and symptoms of impending failure is the best bet. As Nev has already said. e.g. Check your filters, recognize changes in vibration, oil consumption / leakage, exhaust emission products, engine temperatures, always borescope cylinders once a year, monitor plug wear, fuel pump pressures at run-up, etc. the list goes on and on. Know your aircraft intimately and you'll have the best chance of seeing it coming.

I fully take Grant's point about experimental engines. As Nev's experience makes painfully clear, whether you are flying over water, everglades, forests, mountains, scrublands, a city, vineyards... etc. There are not always great options available, if you need to land suddenly. This is another reason why I get some peace of mind from the -540 engine and other 6-cylinder engines, they are well balanced and often keep running "well enough" if a cylinder fails somehow. Of course we all have a range of stories and experiences, so generalising is dangerous.

[Battson]

Parts being used for experimental engines CAN have an abundance of unknown hours. During OH, the through bolts are suppose to be replaced but there is no requirement for the short case studs to be changed out. I believe these studs only have a life of X amount of hours considering the cycles of stress. Engines for their first or second OH should be safe but there are parts out there with over 10k of hours accumulated and nothing more is considered about the high cycled studs being put back into service. My 2 cents worth…..

I am not so sure, I mean anything is possible, but if I recall correctly:

Cyclic fatigue has a critical stress limit. Below that stress limit, cyclic fatigue cracking cannot occur as the crack cannot grow, because the stress isn't high enough. Crack growth is a function of the critical flaw size (largest micro-crack), as well. If the crack is growing, it will not last very long on an engine, 10^8 stress cycles is an average fatigue life for instance - which is about 700 hours of flight time, But it can vary through a wide range depending on the stress level and crack size. The rate of cycle accumulation is just too fast for parts to last for along time on an engine, if fatigue cracking processes are in place.

[Steve W]

Well, that’s comforting to know.
And this is obviously way beyond my pay grade!

[African Bear]

Nev,
What bad luck for this to happen, but good luck and good airmanship got you and your family safely on the ground.
Interesting reads on all the posts and a good jog to remind us all to look for even the smallest tell-tales.

[Nev]

How can we make this easy to access, and convenient. What items should be on the list?
So, how can I see if mine are torqued properly?

Brooks, this has been on my mind too. With the benefit of hindsight I can now see there was one warning sign that I missed, namely a small oil leak that appeared several weeks beforehand. I had previously had an oil leak from the prop governor, so I attributed it to that, not realizing that it was a new leak. Secondly, I'll now be taking a look at the cylinder studs/nuts where visible on the preflight.

Once we get the aircraft airworthy, I'll fly it back to my home airfield, and at the recommendation of my aircraft engineer and the engineers at South Air in Dunedin we will be inspecting the other cylinder flanges for contaminants, and torquing them to Lycoming specifications. This will take a day or two to accomplish, but given the flow on effects of the past week it should provide simple peace of mind.

If you and your engineer wanted to perform a similar check on your engine, and if it was still on the workbench during the build process, then it's relatively straightforward to do this and would probably take half a day. If the engine is already installed and flying, then it requires the removal of the baffles and a few other components - which is what we will be doing next week to mine. The baffles also need to be removed to gain access for the specialist tool you mentioned.

As I've discovered, the downstream effects of such an incident are not much fun and there's a fair amount of stress involved. My aircraft was tied down outside for 5 nights at an airfield 3 hours drive from home. Gale force winds loosened the control locks and have apparently done some further damage that we need to quantify. Additionally, my flying companion Sarah has lost her enthusiasm to fly in my aircraft. The overall cost I haven't begun to calculate, but fortunately it's only financial. No one was injured, and no one died.

If I had known of these issues beforehand, I would have gladly spent a day rectifying them in order to avoid the events of the past week.

[Bcone1381]

Brooks, this has been on my mind too. With the benefit of hindsight I can now see there was one warning sign that I missed, namely a small oil leak that appeared several weeks beforehand........ I'll now be taking a look at the cylinder studs/nuts where visible on the preflight.

Lycomings leak. I don't know. Do the stud nuts reveal anyting to us? I really dont know. I want to know. I like a good preflight!!! I love our engine cowls .... that we can see them!

To pick up a new small leak and find a broken stud, the nut may not have even been loose. I don't have enough skill to pick that up. I have an idea....Seems to me like a vibration meter might be handy. One could know his engine's base vibration level, document it and other parameters once every 25 hours or so, then watch and see if the vibration changes over time.

Nev, Others may not know that you probably did engine trend checks in your career, where we documented all engine parameters (including vibration on turbine aircraft) in cruise flight so our Maintainers could watch and see engine wear and tear trending over time. For example, A bird strike might not have been noticed, but the vibration might spike up to reveal that something suddenly changed.

If a single stud broke, I suspect a vibration indicator might spike up (measurable, but yet small enough I might not feel or notice it.) Someday our engine monitors might include this data point.

I want to just trust my engine builder. I don't want to tear into it, tearing off all that baffling, to verify the engine that was built iaw Lycoming Manuel. Yet I feel your experience and the results it has had on your companion. Thats significant to me.

https://www.amazon.com/HOJILA-Digital-Vibration-Analysis-Mechanical/dp/B0BZCSPQZT/ref=sr_1_2_sspa?c=ts&dib=eyJ2IjoiMSJ9.v-ITaoJBYFCgDOx5ekRlFsNisRYUgt7aioEuzNv0UuapSGAoQ55EkNZqbDRzLfL8MMoY1YqU1W27dPZoCJV6dxB0BQq26KtvQzOF9ZKMW_c91cyTgQuliowlTWuPwzAbJCPZ1SfKnk6WUwm1GugY4WYQ1UnNw-CbGw6dh_UWLv_CQ558qnzzRjPoOYyX2yS2edL6gm2HpQLAmVYnn1kdTDiOuJ-OILbF1JUYgDFQ93OYNMa2onPoWUQ7f9fh6crA6qq8HHpb4C0JSO3FT1ZY0b4MEL3qGD3rJVap1ulLPtU.BXMIw5Z2nH0FZHosMQ6fNkcIoKrmzbEhfLr90EFQ7-E&dib_tag=se&keywords=Mechanical+Vibration+Meters&qid=1731495483&s=industrial&sr=1-2-spons&ts_id=16413791&sp_csd=d2lkZ2V0TmFtZT1zcF9hdGY&psc=1

[Nev]

I wonder how a vibration monitor would go on a piston engine. They were certainly the biz on the turbines and jets. Maybe someone has tried it.

[whee]

Battson is always so at explaining things is ways most folks can understand. Wish I had that knack.

IMO, the best way to prevent this from happening is to check your cylinder bases for signs that they were assembled with sealant on the base flange. Typically this is easy to see with a visual inspection. If they were assembled with sealant on the flanges then your engine was not assembled in accordance with the Lycoming or Continental overhaul manual and you are at high risk of loosing preload on the studs/through bolts which will lead to their eventual failure.

[svyolo]

I wish I had something, gained from experience, to add. But my only "add" is that Lycoming and Conti have changed so much, over so many decades, what is real any more? What is the bible of their engines? We exist in a tiny market, with tiny manufacturers, that are trying to stay in business. Lycoming recommends something now that they didn't for the last 50 years. So they were wrong for 50 years?


Does anybody at Lycoming or Conti even know the standards by which the original engines were designed or built? How many AD's have come out over the decades for their screw ups?

The best we, as users, can come up with are "best practices". What is that?

I haven't worked my way up to the novice level yet.

[Nev]

Battson is always so at explaining things is ways most folks can understand. Wish I had that knack.

IMO, the best way to prevent this from happening is to check your cylinder bases for signs that they were assembled with sealant on the base flange. Typically this is easy to see with a visual inspection. If they were assembled with sealant on the flanges then your engine was not assembled in accordance with the Lycoming or Continental overhaul manual and you are at high risk of loosing preload on the studs/through bolts which will lead to their eventual failure.

Check both sides of the cylinder flange too - between the flange and the hold down plates which is where sealant was found on mine.

[svyolo]

I checked mine today and they look OK. I broke in my engine on the ground, and now the prop doesn't cycle, which it did in the beginning. Pulled the filter, and there were bits of red sealant. Not large, maybe 1 - 3 mm. Taking apart the Johistral governor tomorrow.