What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Exhaust Leak

S

SHIPCHIEF

Well Known Member
Today was the 8th flight of N89SE, my Mazda Turbo Rotary powered RV-8.
It was also the first cross country flight, meaning I intended to land at another airport.
The target airport is just shy of 50 miles from the home drome, so it was a pretty big first leap, and it wasn't without cost.
The turbo outlet flange nuts backed off and searing exhaust blew out the gasket then through the turbo housing insulation jacket, through the stick on reflective heat barrier, and charred the fiberglass cowl. Smelled bad too. :(
I had remained high approaching the destination, so I had the luxury of pulling power and even S turning to bleed off some energy for a straight in to a no wind landing :)
God must have been watching, I had planned a longer route, but fog kept me from going there, so the flight time from leak to landing was thankfully brief.
The fix is simple, correct material science & better locking, and the cowl repair will probably include an aluminum inspection hatch.
 
Just a follow up:
The Turbo exhaust flange - to - exhaust pipe flange is assembled with studs in the turbine housing and I used tall brass exhaust nuts with hardened washers. The brass nuts softened from the heat, forming a kind of flange at the end contacting the washer. Once the tension was lost, heat & vibration allowed the nuts to back out farther.
 
Good update. This is a scary thing, glad it turned out ok.

When I am flying a new alternative engine, I spend the first 10 hours directly over the airport and the cowling comes off after every flight for a thorough inspection of every nut, bolt, clamp, hose etc. First flight limited to 20 mins and no flight over 45 within that 10 hours. You simply can't be too careful on a new setup.

With regards to turbo hardware I use Grade 8 hardware and never studs and nuts, always bolts in blind holes and these are drilled and lock wired on the turbine outlet. On the turbine inlet, bolts and double nuts as there is usually no room to lock wire.

When it comes to gaskets, I use none on the outlet and embossed stainless shim types on the inlet. Never use composition gaskets as they can fry or take a set, allowing the fasteners to come loose.

Great to see you putting hours on it! :)
 
Ross;
you pretty well summarized the problem.
Dave Leonard over on the flyrotary email list had valuable input as well.
I cleaned up the mess, installed grade 8 bolts and grade C high temp lock nuts on the turbo mount flange. I used the stainless embossed shim type gasket.
I assembled the turbine exhaust - to - exhaust pipe flange W/O gasket, using 1" grade 5 bolts that I had on hand. I used grade 8 washers, stacked based on the hole depth, they are not all the same depth.
I removed the Kevlar exhaust pipe wrap, cut off an undamaged section, and stuffed it into the burned thru hole in the turbo blanket.
I made exhaust pipe heat shields out of cut offs from galvanized building studs, left over from building my hangar.
N89SE ran like a top, and made the 32 minute flight back to the Home Drome with no trouble.
Today I shopped at the local specialty fastener store and picked up more Grade 8 hardware and some 1/16" cobalt drill bits, no one seems to carry drilled head grade 8 bolts around here.
I brought the top cowl home to repair the heat damaged spot over the turbo. The repair might be an aluminum inspection door.
Oh, one last thing; today I post flight inspected the repair, and all eight fasteners took 1/12th of a turn to re-tighten. I figure that's just thermal setting. Even if safety wired, they would need to be un-wired and re-torqued at least once?
 
Without Knowing How Much Heat This Application Sees..

I would only offer that you might want to consider that even grade 8 fasteners are typically tempered at about 800 degrees F. Prolonged exposure in a high heat environment can lead to further tempering of the fastener, which decreases both hardness and tensile and increases ductility. Grade 8 cap screws, for example, are typically not used in applications over 450 degrees. The manufacturer of the turbo should be able to tell you what material/grade is appropriate for both male and female fasteners, as well as application torque(s).
As far as the 1/2 turn "relaxation" that you saw after your last flight, you should check repeatedly as suggested above. If, for example, you're using a fastener with 20 threads per inch, then the joint has loosened by .025 inches, you've lost clamping force, and fatigue failure can occur. As above, you need to make certain that you're using proper fasteners.
Terry, CFI
RV9A N323TP
 
SHIPCHIEF said:
Ross;
you pretty well summarized the problem.
Dave Leonard over on the flyrotary email list had valuable input as well.
I cleaned up the mess, installed grade 8 bolts and grade C high temp lock nuts on the turbo mount flange. I used the stainless embossed shim type gasket.
I assembled the turbine exhaust - to - exhaust pipe flange W/O gasket, using 1" grade 5 bolts that I had on hand. I used grade 8 washers, stacked based on the hole depth, they are not all the same depth.
I removed the Kevlar exhaust pipe wrap, cut off an undamaged section, and stuffed it into the burned thru hole in the turbo blanket.
I made exhaust pipe heat shields out of cut offs from galvanized building studs, left over from building my hangar.
N89SE ran like a top, and made the 32 minute flight back to the Home Drome with no trouble.
Today I shopped at the local specialty fastener store and picked up more Grade 8 hardware and some 1/16" cobalt drill bits, no one seems to carry drilled head grade 8 bolts around here.
I brought the top cowl home to repair the heat damaged spot over the turbo. The repair might be an aluminum inspection door.
Oh, one last thing; today I post flight inspected the repair, and all eight fasteners took 1/12th of a turn to re-tighten. I figure that's just thermal setting. Even if safety wired, they would need to be un-wired and re-torqued at least once?
Click to expand...

Scott,

You really need some high temp locking nuts for that turbo. Talk to your CAT mechanics or dealer parts guy, they use some silver-plated all metal locking nuts for the turbos that will withstand high temps. I don't know if they take the 1800 F that a rotary can dish out, but they are much much better than we can find through typical industrial suppliers.

I did not see them in this catalogue, but the "One Safe Source" is a very good reference that is used for most common parts. Note there are good oring tables there too.

I used to make the selection of high temperature turbo bolts as a design engineer, but have long forgotten the material specs. The cautions of grade 8 above are valid. Even though stainless galls badly, with copper anti seize it still may be the best material readily available. One would have to look at high temperature creep and yield strength to know definitively. Maybe turbine fasteners, but I am not familiar with them.

Here is a link to the book for download.
http://www.whayne.com/parts/Documents/OneSafeSource2013-14_PECP9067-04.pdf

Thanks for your posts, I enjoy hearing about your development program.
 
Last edited:
I ended up using 18-8 stainless steel hardware, available from the local hardware store. I checked 18-8's material properties, it's good to about 1500F. Exhaust temps run to 1750F at power, but the manifold & fasteners are open to surface airflow around the engine.
I drilled the bolt heads for the exhaust flange bolts & safety wired them.
the Turbo mount flange was more difficult. One hole wouldn't take a bolt, I had to make a stud by cutting off a stainless carriage bolt. Then feed the stud through the manifold flange into a nut that fits so snug against the manifold that it's effectively locked. The I set the turbo over the stud and put a fender washer, cut away to fit the turbo. Then installed the top nut and bent the washer to be a lock tab.
The remaining base holes accepted fine thread bolts and center punch lock nuts.
I have one half hour flight on this, a full heat cycle, including a brief period with 1800F exhaust temp. Next chance I have to work on it, I'll check tightness on the base flange bolts and lock nuts prior to flight.
 
I think you'll find this stainless material will not only gall badly (stainless on stainless) it will also break easily after a few thermal cycles. Its fatigue resistance is poor. I have also found that fine thread fasteners don't last long here either. Finally, don't use washers under the bolt heads or nuts, they tend to deform and things come loose. I've been doing turbo stuff for 35 years and tried most solutions out there...

Any parts subjected to true 1800F temps really needs Inconel fasteners.

Clearance air within the cowling can keep turbo stuff alive by keeping piping and fastener temps down somewhat. Tightly cowled and ceramic wrapped exhaust usually spell short life unless 321 stainless or Inconel piping is used. 304 stainless has no place on a Wankel aircraft engine IMO.

I've been flying my turbo for 10 years with grade 8 fasteners and they've been on my turbo shop car for 16 years but neither of these see much over 1400F. Race car and race plane stuff might see 1600-1800F. The Grade 8 fasteners have also stood up to that over the years better than anything other than Inconel. I've done dozens and dozens of turbo systems with 3/8 NC grade 8 turbo hardware over the years and have rarely had any fasteners break.

Rotaries are a different beast. The high acoustic signature of the exhaust combined with the higher temps are really brutal on exhaust parts- one of the major drawbacks of the Wankel IMO.

Be very careful that flanged joints are not being subjected to thermal expansion loading- use slip joints or bellows to prevent this.

I'd be making short flights and checking this stuff every time until you have a good number of hours on it all. This is critical stuff- a red hot turbo lying in the bottom of the cowling will ruin your whole day...
 
Last edited:
Ross;
Well taken, I will check every flight. Which is probably not more than one flight before my next week of work. Inconel 600 shows a high temperature service limit of 2000F.
Do you have any leads on a good supplier I can call, who won't mind selling such a small order? I could possibly receive the parts before my next days off.
I have replaced this hardware a few times, so there is a growing pile of used parts and a list of good parts that fit. Now I can make a proper order of the exotic material parts with no waste.
 
SHIPCHIEF said:
Ross;
Well taken, I will check every flight. Which is probably not more than one flight before my next week of work. Inconel 600 shows a high temperature service limit of 2000F.
Do you have any leads on a good supplier I can call, who won't mind selling such a small order? I could possibly receive the parts before my next days off.
I have replaced this hardware a few times, so there is a growing pile of used parts and a list of good parts that fit. Now I can make a proper order of the exotic material parts with no waste.
Click to expand...

I see lots of suppliers on the net but many might not want to sell you a handful of bolts and nuts. Just have to send an email to a few and see I guess...
 
You must log in or register to reply here.
Back
Top