TSIO360 -- loose turbo?

[whitepines]

I have a Seneca II with an unusual failure I haven't really found anyone else mentioning online. Long story short, ~35 hours out from what was supposed to have been a thorough annual (DAR signoff afterward), the number 1 engine failed (stopped producing power*) on climbout. On landing, it was eventually discovered that the turbocharger had wiggled loose from its mounting bracket, trashing both the turbo and bracket.

Has anyone else heard of anything like this happening before on these engines / airframes? The safety wire was still intact, but the bolts were bent pretty badly and torn out of the housing. Otherwise the turbo itself was fine, the engine apparently stopped producing power because the turbo pulled apart the slip joints on the exhaust system when it came loose (no boost).

I guess I'm curious as to how this happened and what I, as the owner / pilot, could have feasibly done differently to avoid the engine failure* and expense of a new turbo. I can't think of any easy way to inspect it on preflight, and honestly "wiggle the turbocharger on each engine to see if it's loose" isn't something that would have even entered my thoughts before this debacle.

* Yes, the engine was probably producing a *little* bit of power still, but with No. 2 at cruise climb (35") and No. 1 suddenly reverting to natural aspiration at ~10k, it sure acted like the real deal! I chose to cage it pretty quickly since I didn't know what was wrong and didn't want to make a perfectly controllable situation worse (fire, etc.).

 

[Bacho]

I have a Seneca II with an unusual failure I haven't really found anyone else mentioning online. Long story short, ~35 hours out from what was supposed to have been a thorough annual (DAR signoff afterward), the number 1 engine failed (stopped producing power*) on climbout. On landing, it was eventually discovered that the turbocharger had wiggled loose from its mounting bracket, trashing both the turbo and bracket.

Has anyone else heard of anything like this happening before on these engines / airframes? The safety wire was still intact, but the bolts were bent pretty badly and torn out of the housing. Otherwise the turbo itself was fine, the engine apparently stopped producing power because the turbo pulled apart the slip joints on the exhaust system when it came loose (no boost).

I guess I'm curious as to how this happened and what I, as the owner / pilot, could have feasibly done differently to avoid the engine failure* and expense of a new turbo. I can't think of any easy way to inspect it on preflight, and honestly "wiggle the turbocharger on each engine to see if it's loose" isn't something that would have even entered my thoughts before this debacle.

Needless to say I *won't* be going back to the shop that did the annual!

* Yes, the engine was probably producing a *little* bit of power still, but with No. 2 at cruise climb (35") and No. 1 suddenly reverting to natural aspiration at ~10k, it sure acted like the real deal! I chose to cage it pretty quickly since I didn't know what was wrong and didn't want to make a perfectly controllable situation worse (fire, etc.).

Any pictures of this?

 

[whitepines]

Any pictures of this?

Yep. Attached.

Unfortunately this is really not an easy part of the engine to look at, you have to pull the cowls to see anything at all.

 

[Bacho]

Looks like the bolts were over tightened and pulled out the threads. Probably would not have been obvious prior to running.

 

[Bell206]

~35 hours out from what was supposed to have been a thorough annual (DAR signoff afterward),

Curious why a DAR sign off? Was this an imported aircraft? New AWC issued? Also did the shop have the turbo or exhaust/waste gate off for the annual? As mentioned does look like threads pulled out.

 

[whitepines]

Curious why a DAR sign off? Was this an imported aircraft? New AWC issued? Also did the shop have the turbo or exhaust/waste gate off for the annual? As mentioned does look like threads pulled out.

Yes, an import. I had been wondering if the turbo might have been pulled for inspection or for overall repairs in the annual (I know a bunch of seals were replaced, for example, and could see the turbo or exhaust getting in the way), but haven't yet found out if that was the case.

 

[Bell206]

but haven't yet found out if that was the case.

Before fingering the shop on this, I'd see what they actually did. In all honesty, based only on your pictures, none of the turbo stuff looks to have been removed recently. Where did the aircraft come from?

 

[whitepines]

Before fingering the shop on this, I'd see what they actually did. In all honesty, based only on your pictures, none of the turbo stuff looks to have been removed recently. Where did the aircraft come from?

Yeah, and I was hesitant to go there, hence "might have been". Speaking to a couple people just now I don't think the turbo was removed at all since engine installation, as that would not be a normal procedure. I'm definitely learning a lot about these engines!

The big question I have is if the bolts were that loose, why wasn't this caught at annual. The engine was supposedly in excellent shape at pre-buy and is only mid-time. It's an honest question -- if this is just something that isn't checked and wouldn't be visible, I'll accept that, but I don't like the idea of engines just randomly dropping turbos either.

Aircraft was an import from Canada. Airframe is low time, in excellent condition, no corrosion, etc.

 

[Bell206]

The big question I have is if the bolts were that loose, why wasn't this caught at annual.

As mentioned above I don't think they were "loose" per se but were overtorqued which pulled the threads. Those pulled threads seem to around the bolt threads. Usually when this happens the bolt will remain wedged in the hole and appear tight. Then when pressure is applied or something else disturbs that area it pops loose as there is no clamping force remaining by the bolts. Also if those bolts had been actually loose in the holes you would see different damage and telltales. Since it came from Canada maybe check those records or ping Dan Thomas here for input as Canada has additional exhaust system checks required on a regular basis which may have been performed since the engine was installed. But definitely check with the shop to see if any of the exhaust was removed for the annual.

 

[whitepines]

As mentioned above I don't think they were "loose" per se but were overtorqued which pulled the threads.

Got it. So basically being overtorqued there would be no visible indication until one finally let go and a cascade failure occured on the rest?

Those pulled threads seem to around the bolt threads. Usually when this happens the bolt will remain wedged in the hole and appear tight. Then when pressure is applied or something else disturbs that area it pops loose as there is no clamping force remaining by the bolts. Also if those bolts had been actually loose in the holes you would see different damage and telltales. Since it came from Canada maybe check those records or ping Dan Thomas here for input as Canada has additional exhaust system checks required on a regular basis which may have been performed since the engine was installed. But definitely check with the shop to see if any of the exhaust was removed for the annual.

I'll do exactly that. My main concern right now is whether No. 2 is similarly affected, and if there's any way to verify before I have a repeat experience on the other engine.

 

[Tantalum]

Thanks for the write-up! Great work on the power loss on takeoff, and great to show that no, twin engine airplanes are not simply there to "fly to the scene of the accident" as so many idiots think

Sorry about the maintenance hassle, I wish you luck.

 

[Bacho]

Got it. So basically being overtorqued there would be no visible indication until one finally let go and a cascade failure occured on the rest?



I'll do exactly that. My main concern right now is whether No. 2 is similarly affected, and if there's any way to verify before I have a repeat experience on the other engine.

Those bolts would be easily over torqued. Bolts will need removed for inspection.You run some risk of the same thing happening on doing that though.

I would look pretty hard at a heli-coil repair before replacing the turbo. Heli-coils are quite common in aluminum parts.

 

[whitepines]

Since it came from Canada maybe check those records or ping Dan Thomas here for input as Canada has additional exhaust system checks required on a regular basis which may have been performed since the engine was installed. But definitely check with the shop to see if any of the exhaust was removed for the annual.

Looks like the last time the turbo was touched was when a new one was installed in 2019. I guess it could have taken several years for the bolts to finally fail?

Thanks for the write-up! Great work on the power loss on takeoff, and great to show that no, twin engine airplanes are not simply there to "fly to the scene of the accident" as so many idiots think.

Yeah, I never understood that. If you fly to a professional standard an engine failure in a twin is a non-event. That means proper preflight, proper W&B (no overloading, ensuring you still have OEI climb performance at altitude, etc.), proficiency in OEI operations, etc. I was more irritated that the engine let go than anything else when it happened, since I knew it was going to be a long drive to the destination!

 

[Bell206]

So basically being overtorqued there would be no visible indication until one finally let go and a cascade failure occured on the rest?

I guess it could have taken several years for the bolts to finally fail?

Yes it could happen that way. For example, everything could have been fine then during your last annual they removed a tailpipe or worked on something close that disturbed the turbo then it popped loosened up. These are mainly guesses since I can't look at the part but have seen similar issues. One issue that may cause a hardware torque issue is the over-use of anti-seize on exhaust components. Most torques values are for dry applications (no lube) and when torqued with anti-seize on the threads you can increase the applied torque vales considerably without knowing it.

My main concern right now is whether No. 2 is similarly affected, and if there's any way to verify before I have a repeat experience on the other engine.

Looks like the last time the turbo was touched was when a new one was installed in 2019.

Was the other engine turbo changed at the same time in 2019? If not you may not have a problem. But may still want to check it. There are ways to check the security of the bolts without removing the safety wire, but I would defer to your mechanic how he would check them. Perhaps a good tactile inspection of all exhaust components would be a measure at this point.

 

[whitepines]

Quick update -- we just found broken rocker arm bolts in the No. 5 cylinder. Mechanic and I are both starting to think this was a cascade failure initiated by the cylinder failing on climbout, which surprisingly didn't cause much in the way of additional vibration in the airframe itself prior to the power loss.

My best guess at the moment is that when the cylinder failed the engine suddenly started vibrating badly enough to cause the previously overtorqued turbo bolts to fail, and I'm wondering if the reason I didn't feel much vibration is that the subsequent turbo bolt failure and power loss may have been very rapid (on the order of seconds). I know the runup was fine and the takeoff was normal, and I don't see how a failed cylinder would have allowed normal takeoff power, so it has to have failed sometime after entering cruise climb.

Now it's a waiting game for a new Continental cylinder. Sigh...

 

[Bell206]

we just found broken rocker arm bolts in the No. 5 cylinder.

Were the push rods bent as well?

 

[whitepines]

Were the push rods bent as well?

Not sure, I can ask. I know the lifters were damaged. Thankfully the cam etc. are fine, so the damaged parts will all be replaced as part of the new cylinder assembly.

Digging back in the logs again, the failed cylinder is not original, it was replaced at the same time as the turbo. Several other cylinders were replaced at the same time, all due to low compressions.

I've had an engine monitor on order for several months now, but the parts still haven't come in due to the ongoing supply chain problems. All I can say is that the monitor would have really paid off here, since it should have caught the cylinder issues before the cascade failure occurred...

 

[Bell206]

Airframe is low time, in excellent condition, no corrosion, etc.

he failed cylinder is not original, it was replaced at the same time as the turbo. Several other cylinders were replaced at the same time, all due to low compressions.

Were these items replaced just prior to the aircraft going up for sale?

 

[whitepines]

Were these items replaced just prior to the aircraft going up for sale?

No, looks like it was flown for a couple of years on those parts before it went up for sale.

 

[Bacho]

So are you willing to take it back to the same A&P per the OP?

 

[whitepines]

So are you willing to take it back to the same A&P per the OP?

I did edit that out as we learned a lot more about the failure since. I'll chalk this one up to learning about the engines and how small (imperceptible) issues can cascade into huge problems, which is why an engine monitor is a good idea when keeping track of 12 cylinders and two turbochargers. The most frustrating part of this is that the engine monitor has been on order basically ever since I bought the plane, I just didn't expect an issue so quickly.

As to whether I'd take it back to the shop in Canada that installed the parts that failed? No, not a chance.

For anyone that's curious, pics of the #5 cylinder rocker arm studs attached. This is what probably initiated the cascade failure.

 

[whitepines]

Just to wrap things up here, new turbo, new turbo mount, new cylinder, new exhaust system parts (header etc.), and the plane is back in service. All told around a $17k unexpected bill and a bunch of driving around the country vs. flying (extra time off work).

My biggest takeaways are...
1.) Engine monitors can save a ton of cash when installed and used (i.e. ground the airplane if any sensor readings don't seem normal, better to delay a day or two than be down a month and out $$$)
2.) In a twin, any degree of split throttles can be a sign of a sick engine. This one's a lot more nebulous since some twins normally split the throttles a bit (why I didn't catch it), but I was surprised when I got it back that the little bit of split was effectively gone.
3.) It's entirely possible for damage (e.g. overtorqued bolts) done years prior to exist and wait for the proper opportunity to cause havoc, all while appearing normal at inspection time. This is the most frustrating part for me -- I prefer to spend money up front to prevent problems than spend it afterward to fix them, but you can't fix what you can't see in the first place.

 

[NordicDave]

@whitepines Awesome you're back in the air. As a fellow turbo owner, the torque settings are critical. Shocking how many A&P's are "tight plus a 1/4 turn" or the Harley Davidson method... "2 grunts and a pull at the wrench".

Turbo mounting hardware takes a thermal expansion beating. The AN hardware is low cost and I change mine out every few hundred hours or on-condition.

A great reminder for pilots changing oil to be curious about what might appear wonky.