Bolts Came Off/Sheared on #2 Cylinde

[Lima Whiskey]

This is my first time using the forum. Last week, after doing several practice Instrument approaches each ending with missed approaches, with the two other partners in our 1979 M20J, we noticed the engine running rough when we were taxing back to the hanger. It didn’t seem too bad so we figured we would diagnose prior to the next flight. Also, we didn’t notice it when we were flying  

After a week of rainy weather, I pulled her out of the hanger, performed a preflight and fired her up. It was running super rough. When I throttled up the electronic rpm gauge would go crazy (running all the way up and then all the way down, even cutting out). Clearly I wasn’t going to fly it. 

When the mechanic on the field checked it out, he found the plugs were fine and so were the mags. What wasn’t fine was that 6 of the 8 bolts on cylinder two were missing and/or sheared off. Of the remaining, one was tight and the other finger loose. We bought the plane 18 months ago with an overhauled engine that only had 100 hours on it. We also had an annual done in August 2018 at a Mooney service center. The engine now has about 250 hours on it  

Any ideas how this happened? Did the overhaul guys mess up? Do the cylinder head bolts get checked on the annual? Is it possible that we have been doing something wrong with engine management that would have contributed to this? And most importantly, how do we make sure this doesn’t happen again? It could have been catastrophic if the remaking bolts would have left loose in flight.

[PTK]

Here's some good info...

https://youtu.be/CEI3UfkFWX8

 

[Guest]

Welcome to MooneySpace, sorry that your first post is a problem like this.  Generally this is very rare and in my experience caused by improper torquing of the cylinder hold down nuts.  Was the cylinder deck of the crankcase damaged?  If not, at very least all of the studs and nuts require replacement and the entire engine should be re-torqued.

Clarence

[Tony Starke]

Hello Lima Whiskey,

Glad you are OK and found the issue before the cylinder came off. There is a specific sequence to tightening the bolts when assembling a new engine. I have attached Lycoming's Service Instruction here. I'm guessing the two bolts that stayed are the through bolts?

SI1029D Bolts, Crankcase, Tightening Procedures for Thru-Studs and .pdf

[ELT]

Hi, 

I had the #2 cylinder on the right engine depart the airplane in flight on an Aztec.  3 bolts sheared and 3 backed off (narrow deck cylinders you have wide deck)  The investigation found that the overhaul facility failed to remove the paint on the case where the cylinders mated.  The paint cracked and the cylinder started working.  100 hours Something was not done right.  Check the engine closely.  You were lucky you were not inflight.

[jetdriven]

Look and see if there's paint on the flange under the cylinder base nuts. That is a major offender, the paint cooks off, the stud tension relaxes and the base stud fails in cyclic fatigue.

[ProprAire]

I had a similar incident back in 2012 that I reported on MooneySpace.  Here’s the excerpt...

Yesterday, I made a short VFR flight from Rochester to Syracuse (30 min at 3.5K ft). Run-up, take-off, and climb were all normal. After level off I reduced engine RPM to 2500 and closed the cowl flaps. I then proceeded to lean using the EDM-700 Lean Finder. Again, all seemed normal as the #2 cylinder was the first to lean (usually it's #2 or #3). I then adjusted the mixture to run at 100 deg ROP. All continued to go well through descent and landing (enriching the mixture along the way). After rollout I advanced the throttle to taxi to parking when I detected the first signs of something being amiss. I could smell burning oil and I noticed a 100 deg difference in CHTs on te EDM-700 (usually it's no greater than 70 deg). After parking, I dismounted and discovered both nose gear doors were coated with oil and more oil dripping on the tire. 

The Landmark ground crew towed the plane to their hangar where the mechanic (Jesse) and I pulled off the top cowl for a visual inspection of what could be the culprit. We discovered the trouble immediately. The #2 cylinder was attempting to separate from the case, having sheared off the top two bolts. Additionally, there was a 1 1/2 to 2 inch vertical crack in the case adjacent to the front edge of the cylinder...blown engine.

Subsequent inspection discovered all but one bolt remained intact.  It was believed there was a false torque measured after maintenance had attempted to use an epoxy to prevent oil leakage.  I carry that one remaining nut in my flight bag as a good luck charm!

 

Dan

N201NX

[kris_adams]

Who did the engine overhaul?  Sorry you are having problems.

[ProprAire]

I had a similar incident back in 2012 that I reported on MooneySpace.  Here’s the excerpt...

Yesterday, I made a short VFR flight from Rochester to Syracuse (30 min at 3.5K ft). Run-up, take-off, and climb were all normal. After level off I reduced engine RPM to 2500 and closed the cowl flaps. I then proceeded to lean using the EDM-700 Lean Finder. Again, all seemed normal as the #2 cylinder was the first to lean (usually it's #2 or #3). I then adjusted the mixture to run at 100 deg ROP. All continued to go well through descent and landing (enriching the mixture along the way). After rollout I advanced the throttle to taxi to parking when I detected the first signs of something being amiss. I could smell burning oil and I noticed a 100 deg difference in CHTs on te EDM-700 (usually it's no greater than 70 deg). After parking, I dismounted and discovered both nose gear doors were coated with oil and more oil dripping on the tire. 

The Landmark ground crew towed the plane to their hangar where the mechanic (Jesse) and I pulled off the top cowl for a visual inspection of what could be the culprit. We discovered the trouble immediately. The #2 cylinder was attempting to separate from the case, having sheared off the top two bolts. Additionally, there was a 1 1/2 to 2 inch vertical crack in the case adjacent to the front edge of the cylinder...blown engine.

Subsequent inspection discovered all but one bolt remained intact.  It was believed there was a false torque measured after maintenance had attempted to use an epoxy to prevent oil leakage.  I carry that one remaining nut in my flight bag as a good luck charm!

 

Dan

N201NX

[jetdriven]

Whoever put silicone glue or epoxy under the base nut needs to be shown this. This could have easily killed someone. I’m really surprised people do things like this and think there won’t be severe consequences. 

[jaylw314]

12 hours ago, PTK said:

Here's some good info...

https://youtu.be/CEI3UfkFWX8

 

Dang it, you beat me to the link!

[MB65E]

It’s a frustrating common issue with the lycoming 360 cases. I’ve had #2 cyl/case issues in the past when I know everything was done correctly. It’s a flawed design. Lycoming uses the cheapest pot metal on earth for their castings. Divco then makes you believe their cases are made of GOLD, they’ll sell a welded case, then the same case cracks in 200hrs. It sucks!

The same cracks are then missed by an MSC, cylinders are changed,  some then get the bright idea to use a sealing compound. The engine is put together with the existing cracks in the case all for me to discover.

Fun stuff! 

-Matt

[Piloto]

In addition to previous comments the bolt failure could be caused by poor alloy during bolt manufacture. Bolt fatigue due to excessive time on the bolt. To minimize bolt stress avoid low RPM at high power settings. Any time you see oil drip on the nose wheel doors check the engine leak source.

José

Edited May 26, 2018 by Piloto

[Cyril Gibb]

Perhaps there might be a spun, or partially spun main bearing shell as a result of the lack of case tightness if the thru bolts were loose?

I’ve wondered if at annual time, checking the thru and hold-down cylinder bolts to 80% of the torque would provide an early warming of incorrect assembly.  Not suggesting full retorquing, just testing for obvious looseness.

[N201MKTurbo]

1 hour ago, Cyril Gibb said:

Perhaps there might be a spun, or partially spun main bearing shell as a result of the lack of case tightness if the thru bolts were loose?

I’ve wondered if at annual time, checking the thru and hold-down cylinder bolts to 80% of the torque would provide an early warming of incorrect assembly.  Not suggesting full retorquing, just testing for obvious looseness.

The problem is re-torqueing the cylinders requires removing a significant amount of the baffling, the pushrods and tubes and the intake and exhaust systems so you can swing the cylinder base wrenches. This would take a full day to do.

[MB65E]

I believe the threads in the case take the brunt of the beating from the cylinders.  The threads loosen up in the case, The cylinders then start slapping the case causing cracks. The thru bolts hold torque, but the other half of the cylinder is moving slightly. The front half of the left side of the case has less material and is weaker causing more movement. This process starts small and then ultimately the case will fail. The cracks I have seen have only been on the Io-360 cases. Lycoming has instructions for installing helicoils in their engines, I believe the helicoils would handle the stresses better than the base pot metal. This would hold the torque of the studs in the case preventing failure. Neat video of the strengths of verious inserts. Also, All Porsche horizontally apposed engines have helicoils for their studs! Just thinking out loud. Any comments about helicoils on the case studs would be helpful. 

-Matt

 

[Cyril Gibb]

2 hours ago, N201MKTurbo said:

The problem is re-torqueing the cylinders requires removing a significant amount of the baffling, the pushrods and tubes and the intake and exhaust systems so you can swing the cylinder base wrenches. This would take a full day to do.

Not suggesting fully  retorqueing precisely because of what you said.  A lot of effort for little benefit to the vast majority of engines.

 I think that the majority of cylinders coming loose are caused by either insufficiently torqued cylinders when installed, or unclean mating surfaces like paint.  What I was questioning was if just sampling some of the easily accessable fasteners could give an early warning before failure.

I’d really be interested in the other cylinders on these failures.  I assume paint or inadequate torque would be common to the other cylinders.  Has anyone tried the other cylinders to see if they are way below spec torque?  If so, maybe torque sampling might be an easy safely check. (Of the easily accessible fasteners) 

 

[N201MKTurbo]

1 hour ago, Cyril Gibb said:

Not suggesting fully  retorqueing precisely because of what you said.  A lot of effort for little benefit to the vast majority of engines.

 I think that the majority of cylinders coming loose are caused by either insufficiently torqued cylinders when installed, or unclean mating surfaces like paint.  What I was questioning was if just sampling some of the easily accessable fasteners could give an early warning before failure.

I’d really be interested in the other cylinders on these failures.  I assume paint or inadequate torque would be common to the other cylinders.  Has anyone tried the other cylinders to see if they are way below spec torque?  If so, maybe torque sampling might be an easy safely check. (Of the easily accessible fasteners) 

 

It is normal while inspecting an engine to look at all the cylinder nuts to see if any have backed off. 

[jetdriven]

Torque is set while moving. IE a properly torqued fastener will take significantly more than rated torque to get it moving again. So you really can’t retorque them. Also If you have studs loosening you can’t just tighten them again.  You need to investigate the reason for the loosening and also replace all the studs and through bolts beucause they’ve been overstressed. 

A friend of mine’s Bonanza had a cylinder work loose but he caught it before it departed the case. He put it back on but less than a year later the engine came apart inflight and he crashed, killing himself and a passenger. I’m not afraid of these engine but loose cylinder hardware is a VERY serious failure and can not be tolerated under any circumstance.