How to prevent cracked muffler on 912uls

[3kdscf]

200 hours on my engine and my muffler got a huge crack, clearly this is not ordinary wear. Talked to John McBean and his first guess was that the stinger was banging against the engine cowl on startup. I looked and sure enough, saw evidence on the stinger that it had been hitting the cowl. The distance from my stinger to the cowl is 3/8" which would seem like plenty but in a rough cold morning start, clearly that is not enough. Rather than buy another $995 muffler every 200 hours, I think I will increase the distance to 5/8" clearance :) And while I am at it, going to replace my engine mount vibration isolaters too in case those are shot and causing excess movement.

299872998829989

[Shadowrider]

Have you balanced your prop?

[3kdscf]

No, I have not. After reading your comment I headed to the Whirlwind RW-3B installation manual in a panic wondering if I missed something. It appears the prop is statically balanced at the factory because that is not even mentioned in the installation instructions but when it comes to dynamic balancing they say that it is recommended but not required.

I am fairly confident that the stinger hitting the cowl was my issue because the crack showed up after a particularly rough cold morning start. Agree that its definitely worth checking the dynamic prop balance because that could also be part of my problem but its hard to justify a $4000 prop balancer for one test. I just found this awesome thread (https://teamkitfox.com/Forums/threads/11085-Dynavibe-Available?highlight=dynavibe) where a TeamKitfox member is renting theirs out. I will have to contact them.

[Shadowrider]

Yes highly recommend getting the dynavibe and balancing your prop. Static balancing of your prop helps but when its spinning with a spinner is another story.

[avidflyer]

Last summer, I got a 912 muffler that was cracked up pretty good in the same place. Interestingly all the exhaust pipes were stuck in the muffler and wouldn't come out for anything. I ended up making a cut up the length of each nipple and was able to spread/expand the nipples that way and got the exhaust pipes out. Had it all welded back together, and I think it will last a good while in the next project. I ended up with the exhaust pipes being looser in the muffler than they probably were from the factory, but that will likely help things move around more and maybe not as likely to crack. Could be something to take a look at as well. A looser connection may leak a bit of exhaust, but not as much as a cracked muffler I'm thinking. Especially if the entire pipe breaks off. JImChuk

[3kdscf]

Interesting, my four pipes came out of the muffler very easily and also came off the block easily. When I installed my muffler I used loctite LB 8150 (https://www.amazon.com/gp/product/B00FX80LMQ/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1) anti-seize on the muffler end of each pipe as well as the engine block end of each pipe. It makes sense that if they became stuck that it could lead to cracking just from the thermal contraction and expansion.

[Dave S]

Interesting discussion - my experience mimics Dan's. The original installation involved a generous application of the silver/aluminum paste antisieze compound. In replacing the engine this winter (after a mere 500 hours) the muffler and pipes came apart with hand bumps. Also, the muffler has a brace going between the stinger and the muffler. Inspection after disassembly revealed no cracks.

[Delta Whisky]

If this is a contest, then I win! The shame of it all, I read this thread a couple of weeks ago but failed to look as closely as I should have before its last flight. You can see from the picture, I should have bent down and looked really close - it appears that there should have been some evidence of the impending doom. (Its there now!)

And, the last contact the vagrant exhaust pipe made as it left its welded home was one of the new fairings on the good ol Monster Shock.

3020330204

Now I need two things: the ability to rotate these pictures and a new exhaust muffler.

[Shadowrider]

Oh dang that stinks! Did you get to fly before it broke and test the new fairings out?

[avidflyer]

Wonder if you would dare to weld in another elbow? Add a patch on the end of the muffler with the correct size hole in it, weld in the new elbow, and then add the braces from edge of muffler to new elbow. I wish now I had put braces in mine when I had it at the welder. JImChuk

[Delta Whisky]

Dustin - your question touched on the irony of the planned flight. That day, Monday, showed hopes of being a relative calm day aloft and was the reason for the planned flight. I had previously made two flights in hopes of obtaining decent data but the turbulence makes the data suspect. None-the-less, I've started an analysis and, as soon as a head cold clears, I'll see if any part of what I have is useful.

Jim - Your idea has merit but one detail of what a close look into the muffler reveals gives me pause. I'll try to get a better picture the next time I'm at the airport but it appears the exhaust pipe/stinger is (or was) an extension of an internal pipe that runs the length of the muffler. This pipe contains numerous holes the exhaust gases have to pass thru on their way to freedom. If that is the case, four breaks were extremely clean and normal to the center line of the pipe.

[Dave S]

Found a photo which shows the small stinger brace I mentioned before that was part of the original design on the muffler that came with my kit. This is an earlier design ca. 2005. Don't know if that is currently used but it seems like a good idea. No cracks in the muffler body or the stinger weld to date.

[jiott]

My kit was purchased in 2012 and the muffler did not come with provisions for the stinger brace. John McB said the internal construction had been beefed up and the brace was no longer required. Mine has been flying for 1050+ hours now without the brace and no weld problems with the stinger-knock knock.

[Delta Whisky]

3021730218

Better (promised) pictures and some clarity of the internal "silencing" pipe. After looking more closely at the evidence today, I'm developing the theory that the fairly long stinger, in a vibrating environment, caused the cracking which in turn allowed it to then contact the cowl - not the other way around. I found an airflow deflector I had installed to better direct air to the oil cooler also cracked in two places. Clearly I have a vibration issue that has to be resolved and, yes, I did balance the prop during the first 10 hours of flight. (Now 180 hours.)

[avidflyer]

I wonder how long it would last if you were able to slide in an elbow that inside baffle pipe fit into fairly snug. Then weld the elbow and a patch on the end of the muffler. Might last for a bit till you get another muffler at least.

[AvDES LLC]

Do you have periods of excessive vibration with this engine? That vibration could be causing that pipe to smack that cowling and would explain a lot. My exhaust did the exact same thing as the first pictures in this thread and a buddy of mines exhaust did the same thing as the missing stinger photo. Appears this isn’t an uncommon issue.

[jrevens]

I may be wrong, but it seems to me that this issue could be due to the design &/or construction of the Kitfox-specific designed/built exhaust system. There doesn’t seem to be the same frequency of these kinds of failures with the genuine Rotax exhaust systems, for instance, used on many other designs.

[AvDES LLC]

you make a good point. I haven’t seen under the cowling of any planes with 912’s installed except kitfoxes. That said,it would be interesting to see how the cowl is shaped around the exhaust of other aircraft with the 912 as this would give us an idea as to how close other designers have placed the fiberglass.

It’s a bummer to see so many of us have had the same issue.

[avidflyer]

I have a different muffler that has ball joints near the muffler, and then shorter pieces go into the muffler. 2 pipes instead of one for each run. That would probably reduce some of the vibrations our Kitfox mufflers see. JImChuk

[AvDES LLC]

would you mind posting a photo?

[jrevens]

I have a different muffler that has ball joints near the muffler... JImChuk

Absolutely. Ball & slip joints are “standard”, and necessary on many aircraft exhaust systems. A ball joint between a tail pipe & the rest of the system is very common on Lyc’s & Continental’s. Also, for several reasons, the generally accepted “correct” stainless steel alloy is 327. I have to assume & hope that’s what Kitfox uses.

[avidflyer]

I'll try to get a picture tomorrow and post it. JImChuk

[jrevens]

Absolutely. Ball & slip joints are “standard”, and necessary on many aircraft exhaust systems. A ball joint between a tail pipe & the rest of the system is very common on Lyc’s & Continental’s. Also, for several reasons, the generally accepted “correct” stainless steel alloy is 327. I have to assume & hope that’s what Kitfox uses.

I don’t know where my head was, but I meant to say 321, not 327. There are others that are good too, but I believe the people who build systems for aircraft in the homebuilt world generally choose 321, in my experience. I’m interested in what knowledge others have about this.

[Delta Whisky]

As I removed my muffler and got to looking more closely at other items of interest under the cowl, I have become more of a believer in the "vibration caused this" theory. I am walking cautiously here because all too often I jump on the first theory I come up with and it takes a big hammer and crowbar to get me to step back and look elsewhere. Having come clean, I will share that four parts I designed and installed, made from aluminum, have to be replaced due to cracking - and they are located near the center of rotation. That got me to thinking about the muffler and how far it is off the center and that situation is exacerbated by the stinger that hangs off quite a bit more. I compared these parts to an RV-12 and saw a bit of difference in location and size of the exhaust components and am still wondering (based on how often this has now been reported) if there is a resonant frequency issue that might be playing in the equation and could explain why many but not all have seen the issue. Added to the thought process - I had about one finger's width of clearance between the exhaust stinger and cowl and my cowl (round) is fairly flexible where the stinger passes thru. (Born out by the discovery that at flight speeds the cowl has been pushed up against engine components - a situation that will also be addressed soon.) I'm given to a thought that if the contest is between the cowl and the steel exhaust system - at least once in a while the cowl should lose. One final observation - I'm running a Luga prop - because it is so light, it offers almost no dampening in the system. Anyhow, just some thoughts in hopes that someone will be able to put the pieces together and come up with a cause that leads to an solution.

[jrevens]

Have you dynamically balanced the prop?

[Delta Whisky]

Yes and balanced the carbs ------- and get complements on how smooth the engine is. Although, it is smoother at 5200 than anywhere else - but you have to pay close attention to see any difference.

[avidflyer]

Sorry guys, I forgot to get a picture of my muffler. I was even in the hangar. (probably was wondering why I had walked up there....) I'll try again tomorrow. JImChuk

[jrevens]

I don't really have any good pictures of the hole/clearance where my tailpipe exits the cowl, but maybe you can get a good idea from these, FWIW. The clearance is not a lot, but is probably a little more than 1/2". I haven't seen any evidence of the pipe contacting the cowl.

30297303003030130302

[avidflyer]

Well I got a couple of pictures of my muffler with the ball joints on the pipes. It actually has two ball joints per pipe. Lots of springs to wear out, but maybe the stinger won't vibrate off. I got this muffler quite a while back, don't remember what it came off of. There is a cabin or carb heat source attached to the muffler as well. JImChuk

[AvDES LLC]

Thank you for those photos. That helped explain a lot. When I get the cowling back on babe I’ll be sure to take some photos of clearance as well since my exhaust tip broke the same way and I never payed attention to that clearance.

[Rodney]

I thought the muffler was made of a 304 stainless - but am not absolutely sure. What I do know is that I welded my muffler cracks with 308 stainless wire on a MIG welder and it has held so far.
I just got a TIG welder and am going to weld in an Air/Fuel bung into the muffler this week. I'll still use some 308 filler rod for the weld.

Regards
Rodney

[AvDES LLC]

good to hear you’re taking the leap of tuning faith with the AFR. I actually spent the evening putting together a bit of code to help make this process a bit more user friendly and cost effective. Please keep us posted with the initial readings as well as any adjustments made and their results!

[AvDES LLC]

Anyone have an old exhaust they will soon be replacing or sitting on a shelf?

Im half tempted at this point to get an old exhaust can, cut the muffler in half and install a plate in the middle to divide the two engine banks. Then installing an AFR meter on each side will allow for tuning of each carb independently of one another for some very very fine adjustments to see what kinda power can really be pulled from these stock engines - I think tuning to an average is good, but independently would really allow for an understanding as to what each side of the engine is doing with the fuel being delivered without the crossover tube.

[Eric Page]

I have a new, unused exhaust system for a 912UL/ULS, including down-pipes with cylinder flanges and muffler with welded-on stinger. It dates from 1995. No springs or other hardware. I'd take $775 plus shipping. Not interesting in parting it out.

[AvDES LLC]

thank you and if I end up not finding an older clapped out exhaust to cut up, this may be an option. Trying to not destroy new/functional parts if at all possible

[Delta Whisky]

Anyone have an old exhaust they will soon be replacing or sitting on a shelf?

Im half tempted at this point to get an old exhaust can, cut the muffler in half and install a plate in the middle to divide the two engine banks. Then installing an AFR meter on each side will allow for tuning of each carb independently of one another for some very very fine adjustments to see what kinda power can really be pulled from these stock engines - I think tuning to an average is good, but independently would really allow for an understanding as to what each side of the engine is doing with the fuel being delivered without the crossover tube.

Have you given any thought to putting an O2 sensor in one of the headers - one on each side of the engine? Probably would have to go in the forward cylinder exhaust pipes to meet the distance requirements (exhaust valve to sensor) - if they are long enough.

[AvDES LLC]

i have, however you hit the mail in the head with my worry. The sensors have a warning that is very clear to keep them below 1600*F. I’d be worried I’d kill sensors with a quickness.

[Delta Whisky]

Interesting. Just checked a few of my logs and have yet to see any temps above 1450. I'll measure (maybe) tomorrow and check the distance from the end of the exhaust pipe/cylinder head.

[AvDES LLC]

would appreciate that. Interestingly the only time I’ve ever approached the 1400 temps is when I was seeing that extreme lean condition tapping out at around 14.9 to 15.1 to 1 on the AFR. You at all planning to check your ratios during flight and seeing where your engine is at as far as mixture goes?

[Delta Whisky]

Ahh, such a creature of habit. I'm so used to thinking there is an airplane at the airport that I spoke before thinking - the airplane is at the hangar, but the headers are at Kitfox helping in the build of a new muffler. So, measurements aren't available - at least from me. I'll guess that there is a good possibility that my temps are higher because I took them at a much lower altitude. (about 2,000 ft msl)

And I do plan to install an AFR gauge when the muffler makes it back. I am curious to compare my experience and numbers to yours.

[AvDES LLC]

this is awesome to hear. I STRONGLY suggest avoiding the innovate motor sports AFR gauge. I’ve killed 3 in just this tuning event - even my wife was looking at me asking WTH I was doing because there is no way 3 die in the same fashion only in my hands. Well… I returned it and the gentleman behind the counter told me he was getting a lot of these returned. Take that for what you will. I went with the AEM unit and it was actually 6 bucks cheaper and didn’t have a single issue during yesterdays almost 4 hours of flying / engine time.

[Delta Whisky]

That's great feedback - the AEM unit has been on my buy list for a few days now.

[Delta Whisky]

Since losing my exhaust stinger I’ve been hunting the culprit hiding behind the scenes. Some felt that it was most likely engine vibration while others leaned towards the hole in the cowling was too small and startups and/or shutdown shakes caused the stinger to hit the cowl and caused the breakup of a fine marriage. I leaned towards the vibration theory as there was plenty of evidence supporting the same but outside the scope of this report.

While waiting for delivery of a new muffler, Nickolas’ postings on managing air fuel mixture conditions caught my attention as mixtures on the extreme edges of acceptability will often cause vibration.

Once the muffler arrived an AFR gage was installed and I found the only adjustment needed was at idle and a new and improved idle was noticeable.

I next attempted to see if there was any way to get a feel for the “exhaust pipe hitting the cowl” theory and the best I could come up with was the trusted blue tape approach seen here:


30611


After no fewer than 50 start-and-stops (from a fairly wide range of rpm’s) the pipe has yet to make a mark on the tape and the clearance is about ½ of the original clearance.

On to the vibration theory - and here’s where the report gets interesting.


Background: after about 10 (original) hours on the plane, I attempted to balance the prop/engine but couldn’t do better than its original set up. Meaning: no added weight except for the dynamic balancer. BUT – my attempt to balance was performed at 4,000 RPM only and the results were for the most part unrepeatable, run-to-run.


Now, with the loss of the stinger my attention was more focused and I started over again. A base line run was made with the original set up – with the still installed dynamic balancer and spinner and runs were made at 2,000, 2,500, 3,000, 3,500 and 4,000 rpm. Then the same runs were made with the spinner removed (remember, zero additional weight). Next, following the DynaVibe instructions, I balanced the engine/prop to a fairly respectable level. All in all, the results looked like:


30609



Clearly, my label of “respectable” is for the higher rpm’s in the runs following the baseline runs. Nothing I did would eliminate the weird waviness and I started thinking the dynamic balancer may not be up to the task. So off it came.

Tests and set ups were rerun at the same data points and it became clear that something is still happening at 3000 rpm (about 1200 at the prop) but all of the traces show a respectable amount of similarity. (I stopped after run 8 when noticing how it was starting to flatten out across the tested rpm range.)

Finally, a flight was made and the results are presented here on the same graph.


30610




Bottom line, I’m more convinced 1. that it was vibration that caused my issue, 2. the dynamic balancer influenced my vibration survey (favorably) quite a bit but not enough, 3. the dynamic balancer has its own set of issues, and, 4. I’m going to check propeller balance at each condition inspection. BTW, the very minor vibration I used to feel in the pedals is now gone.


Oh yeah, any and all reasonable offers for a slightly used dynamic balancer will be kindly received; the highest bidder will be notified of acceptance NLT 1200 noon this coming Monday.

[AvDES LLC]

this is outstanding news to hear that yet another person has seen drastic improvements as a result of the AFR tuning process. Thank you for reporting your results.

[styree]

I have a new, unused exhaust system for a 912UL/ULS, including down-pipes with cylinder flanges and muffler with welded-on stinger. It dates from 1995. No springs or other hardware. I'd take $775 plus shipping. Not interesting in parting it out.
Do you still have this? I’m looking for an exhaust for my 4-1200 round cowl w 912uls

[Eric Page]

PM sent...

[avidflyer]

The other day I pulled my cowl off on the Kitfox 4 and look what I found. A little bit of fabing up some support pieces, a trip to the welder and $50 and it was all good again. Probably for good this time with the bracing. JImChuk

[AvDES LLC]

The other day I pulled my cowl off on the Kitfox 4 and look what I found. A little bit of fabing up some support pieces, a trip to the welder and $50 and it was all good again. Probably for good this time with the bracing. JImChuk

good thing you found it before the stinger decided to detach in flight. Keep your eyes on the corners of those welds for a while. Typically welds round a corner in an effort to reduce the stress concentration at the weld interface.

[Slyfox]

check your gearbox. static should not fall under 300. just rebuilt my box it was about 250. now it is smooth on shut down and start up. that will crack your exhaust right now.