Broken Studs

M

Magman

Pattern Altitude
Joined
Oct 30, 2020
Messages
1,979
Display Name
Display name:
Magman
Shortly after takeoff the engine sound changed and the pilot returned to the airport.

This is pic of #3 cyl on a Cherokee 140.

5 out of 8 studs are broken including 2 through studs.

Engine has about 1500 hrs and 40 years since MOH by noted shop .

There are no recent jug pulls, if any.

Prop was loaned to someone and he reported excessive vibration even

if track was good.

Amazing how 5 studs can break yet 3 hold it together.

Engine sent to Columbia and they have another with the same condition!
 

Attachments

  • D08D24C3-D486-4571-B372-FB734A493B00.jpeg
    D08D24C3-D486-4571-B372-FB734A493B00.jpeg
    178.8 KB · Views: 152
  • DF15218D-2ED6-4E78-9EFB-8F87395BB460.jpeg
    DF15218D-2ED6-4E78-9EFB-8F87395BB460.jpeg
    227.3 KB · Views: 143
I had stud break on my old motor. Doubt they all broke simultaneously.
 
5 of 8 studs broke on #1 cylinder of O360 with 100 hours on rebuilt engine. It cracked the case at the cylinder opening. Just now finishing the installation of the repaired engine.
 
"RTV can do that."

And I'll bet the referenced case involved a clicker type torque wrench rather than a bending beam type. If nothing else the clicker will under-torque in the presence of any squeeze-out in the joint. A bending beam wrench allows the operator to hold torque for a short time (maybe a couple of seconds), whereas the clicker type will simply trip at the torque peak - if it is even calibrated correctly.
 
nrpetersen said:
A bending beam wrench allows the operator to hold torque for a short time (maybe a couple of seconds), whereas the clicker type will simply trip at the torque peak - if it is even calibrated correctly.
Click to expand...
I've had more than one discussion with mechanics using a clicker type torque wrench when setting pre-load on a bearing (rolling torque). Use the correct tool was my comment
 
nrpetersen said:
"RTV can do that."

And I'll bet the referenced case involved a clicker type torque wrench rather than a bending beam type. If nothing else the clicker will under-torque in the presence of any squeeze-out in the joint. A bending beam wrench allows the operator to hold torque for a short time (maybe a couple of seconds), whereas the clicker type will simply trip at the torque peak - if it is even calibrated correctly.
Click to expand...
Yup. From the Lycoming Direct Drive Overhaul Manual:

upload_2021-9-11_14-22-33.png

From the Continental Standard Practices Manual:

upload_2021-9-11_14-28-30.png
 
Also, metal just….. breaks sometimes.
I broke a thrubolt on my IO-520, found on preflight a few years ago.
No RTV.
Torqued properly (following the current, appropriate, manual instructions.)
 
Might not be a bad idea to check cylinder nut torque every 100 hours like we do on the Jacobs radial.
 
Doug Reid said:
Might not be a bad idea to check cylinder nut torque every 100 hours like we do on the Jacobs radial.
Click to expand...

If there is anything 'bad' about it, (I agree with your idea...and I think the CAT manual actually says to check them).... the downside would be the 12 hours removal/replacement of intake/exhaust/baffle, and turbo accoutrement in the case of turbo-beasts. Only way to get to the nuts that I have found in large bore engines. Plus the MIFs ie when you loosen an exhaust stud.
Listening, in case others have a better way.
 
Interesting! The specified torques (600 in lbs for a 1/2 inch fine thread stud) are not particularly high compared to my industrial experience of 80 ft lbs on a 1/2-13 coarse thread grade 8 bolt (I'm a mechengr, not an A&P). This means that there must be flange-flex under the head of the cylinder hold down nut that is contributing a lot to these fatigue failures.
 
Anyone think the prop vibration could be a factor?

This is a major engine shop that did the o/h 40 years and 1400 hrs ago.

I use anonymous with things that could be taken as bad.

They have a LOT of experience though.

I haven’t gone through all log pages but I don’t think there have been

any jugs pulled.

You may note that Columbia has another one just like it!
 
Let'sgoflying! said:
If there is anything 'bad' about it, (I agree with your idea...and I think the CAT manual actually says to check them).... the downside would be the 12 hours removal/replacement of intake/exhaust/baffle, and turbo accoutrement in the case of turbo-beasts. Only way to get to the nuts that I have found in large bore engines. Plus the MIFs ie when you loosen an exhaust stud.
Listening, in case others have a better way.
Click to expand...

I never said checking cylinder nut torque was easy..it is a PIA on a radial engine as well. I have checked mine several times. IMO, if there is an issue, you can probably find it without looking at every single nut.

I am also not convinced that the failure came from a bit of gasket sealer on the mating surface. I remember a lawsuit where a twin flew directly into a cumulogranite cloud and the family got a big settlement because the seat belt bolts broke on impact. 40 years and 1500 hours since cylinders installed ?
 
nrpetersen said:
This means that there must be flange-flex under the head of the cylinder hold down nut that is contributing a lot to these fatigue failures
Click to expand...
In my experience, you'll find most of these types of stud/bolt failures is due to improper torque and out-spec studs. It's not really about "flange-flex" as it is "hardware-flex" in tension preload. Use of improper sealants, etc. itself can also cause the loss of this preload. Two of the biggest culprits that lead to this type failure is the use of dry torque vs the required wet torque requirement plus the fault of not re-torquing the other end of the through bolts after mx. The second issue is the use of used hardware that are stretched beyond limits. As I'm sure your understand, the integrity of the tension preload value is maintained partly due to the stretch of the fastener. There's been a few studies on this over the years.
 
Last edited:
Detonation is also known to cause cylinder base failures. But detonation in an O-320 is very unlikely. Something else was at work here, and it's not likely we'll get at it by guessing, especially since we don't know what the overall hours (or number of overhauls) are on that case and its studs. And we don't know that those nuts were tightened to the correct values at the last overhaul either. Both overtightening and undertightening can cause problems.

If one stud breaks, the studs on either side of it are now loaded up more, and they'll fail. It's a cascade effect. Losing one stud out of eight is not losing 1/8 of the strength; you're losing much more. Same thing goes for a broken wheel stud on any motor vehicle.
 
Both of the OP's pix show the crack originating on the cylinder side of the flange, with final failure being on the outboard side of the stud. This would be expected with flange flex. My guess would be under-torquing (under-preloading) but that is only speculation. The area of final failure is quite small compared to the overall cross sectional area of the stud.

It could also have been detonation at work, which could be expected with tiny amounts of jet fuel contamination into the AvGas or MoGas.

The failures are not from propeller vibration.
 
nrpetersen said:
with final failure being on the outboard side of the stud.
Click to expand...
You can also have failures in the case halves that run parallel to the cylinder flanges due to through-bolt torque loss. As explained to me, it's the torque preload on the through-bolts that handles the primary forces of combustion and if any portion of the through-bolt torque process is not followed or fails, it will allow those forces to transfer to the other areas like cases, etc. which are not designed to handle those higher forces.
 
Magman said:
...You may note that Columbia has another one just like it!
Click to expand...

Yea but we need to prevent ourselves from concluding that "just like it" means it has to be for the same reason.
 
Let'sgoflying! said:
Also, metal just….. breaks sometimes.
I broke a thrubolt on my IO-520, found on preflight a few years ago.
No RTV.
Torqued properly (following the current, appropriate, manual instructions.)
Click to expand...
What did you do in your preflight where you caught this? Was it during run-up?
 
Report from electron microscope:

“ Clear fatigue failure initiating from multiple points.

No manufacturing or material defects involved.”


Info will be trickling in. Stay tuned.
 
nrpetersen said:
Interesting! The specified torques (600 in lbs for a 1/2 inch fine thread stud) are not particularly high
Click to expand...

I noticed that too, however these are clean and lubricated. My experience of non aviation stuff always specified clean and dry threads for assembly.I would guess that this will increase the stud tension from a specific torque by a few times. Oil works :)
 
Spring Ford said:
I noticed that too, however these are clean and lubricated. My experience of non aviation stuff always specified clean and dry threads for assembly.I would guess that this will increase the stud tension from a specific torque by a few times. Oil works :)
Click to expand...
The engine manuals give the specifics on torquing the nuts and they often specify lubricated threads. Rules-of-thumb can get you into trouble. And non-aviation stuff does not apply.

I posted the relevant pages from the manuals in post #7. Lubricated threads.
 
Spring Ford said:
I noticed that too, however these are clean and lubricated. My experience of non aviation stuff always specified clean and dry threads for assembly.I would guess that this will increase the stud tension from a specific torque by a few times. Oil works :)
Click to expand...
Except that as a manufacturer of fatigue testing systems, we used industrial grade 8 fasteners lubricated with moly-disulphide, which is a more powerful lubricant than oils and will give more consistent preloading. I'll have to admit though that the general policy was to not reuse fasteners in critical applications. Most important though was the design details of a joint to avoid cyclic fastener strain, and we used of long fasteners. We never used lock wire or lock washers.

Engine manufacturers have to make do with a lot of compromises that we didn't have to observe. Most, but not all, of our equipment was not weight critical. But there is a lot to be said for the more-consistent fatigue spectrum that an aircraft engine experiences, and the net experience and recommendations of the factory service manual would be carefully recommended by any of us. Only occasionally did some of us get to diagnose things that didn't make obvious sense.
 
You must log in or register to reply here.
Back
Top