Cracked Muffler solution

[Spiff]

I have an M4-180v with a lycoming O-360-C1F. I have had several cracked mufflers as a result of mechanics tightening the clamp that holds the exhaust. Maule says to leave it very lose so it can flop around in the breeze. That does work, but leads to another problem. The exhaust is held on by a pin in a hole. With everything loose the pin elongates the hole eventually. There are companies that custom build Maule exhaust systems for both the O-540 and the O-360. For the O-360 you have to send them your exhaust so they can measure it and it is a few thousand dollars. They do add a ball joint to the new exhaust. Last year at my annual my mechanic, used my existing Maule exhaust and just added ball joints. He then tightened the clamp. After over 100 hours, it seems to work very well. No cracked mufflers and no hole elongation. Here are photos.

[Flybrian1950]

Is the ball joint a simple clamp-on or was some welding involved?

[andy]

There are quite a few exhaust ball joints out there that don't require welding - like this one:
https://www.jegs.com/i/Jones+Exhaust/52 ... 5m--J1PJu0

You might need a STC or field approval though. I don't know if this would be considered a major airframe modification. Maybe one of the A&Ps on the forum can comment.

A ball joint doesn't help where the exhaust pipe connects to the slip joint at the cylinder exhaust port, though. That's a wear point that deforms the pipe and allows exhaust gas blow-by, especially on 4-cylinder engines that vibrate more than 6-cylinders. Nor does it help where the slip joint bracket is welded to the pipe. I've had both problems with my exhaust as well as cracking in the welds where the pipes connect to the muffler. Thicker stainless steel pipes, an outer pipe sleeve at the exhaust headers, a doubler plate on the slip joint bracket and redesigning the mufflers to avoid welds might solve these problems.

[Hottshot]

We've used ball joints on 180 horse for very long time, most of them were done Via Field approval. I will dig up some more paperwork to see if I have some left. It definitely adds longevity to the system

[Andy Young]

While we are on the subject, I find that I literally burn through mufflers. I use the Atlee Dodge units, and get about 500 hours out of them before the flame tubes burn out and depart the mufflers. Usually around that time I’m just starting to see failures in other areas as well, though I’ve largely solved the frequent vibration cracks. My failures all seem to be heat related.

Atlee Dodge has expressed surprise at the short life of my mufflers, stating that most people get about 1000 hours out of them. They feel that it’s because of the EGTs that I see, often in the 1400s. I have an IO-540-W1A5D. I am at right around or just under 75% power almost all the time, operating at 2200 rpm and 22” manifold pressure. At that power setting, I run AT peak EGT (not LOP or ROP) which is per Lycoming’s recommendation, found both in the engine operations manual and confirmed through several long conversations with the Lycoming tech support line. I have been running this way for many years, and thousands of operating hours.

I’m quite sure I could make the mufflers last to 1000 hours if a I ran rich enough to get my EGTs down to about 1200, which is what Atlee Dodge recommends, but I calculate that the additional fuel to go those extra 500 hours at that ROP setting would cost me about $6,000, whereas a pair of mufflers costs about $2500. I haven’t really explored running lean of peak, though I’m sure that would help, as the EGTs would be lower. I just haven’t been willing to give up the speed. Maybe I could run LOP and at a higher manifold pressure and break even? Can someone with LOP experience comment?

My believe is that much of the problem is due to the very short path from the combustion chamber to the flame cone, with those mufflers sitting right under each cylinder bank. I think that if the muffler(s) sat under the oil pan, like on a Cessna 170, or back at the firewall, like a Cub, the much longer exhaust runners would give the exhaust gases time to cool a bit and lose energy before impacting those expensive and somewhat delicate mufflers, thereby extending their life. Unfortunately, I’m not really up to getting approval for a whole exhaust system re-design.

I’m interested in the thoughts of others on these subjects. Happy to discuss operating at peak EGT, theories about exhaust life, etc.

[andy]

When I installed my JPI EDM 800 engine analyzer in 2002, JPI recommended that the EGT probes be placed 2 to 4 inches down from the bottom of the exhaust headers. I chose 3 inches. That produced an EGT at peak lean of about 1500 on my O-360-C1F. Lycoming recommends running 50F rich of peak on this engine so 1450 is typically what I look for. I haven't had any recurrent problems with the muffler flame traps. I've replaced a couple of JPI's EGT probes over the years since the hot, acidic exhaust gas corrodes them over time. I've replaced mufflers several times and occasionally the mechanics re-install the EGT probe at a slightly different location, which changes the EGT readings significantly. I think that it would be best to install the probes at the 4 inch point to reduce the corrosion and make the probe last longer. That would reduce the peak EGT reading but I'm not sure how much. It might get it down to 1400 and 1350 at 50F rich of peak. It wouldn't extend the life of the flame traps but it would give a more accurate idea of the EGT going into them.

Most of my problems with the exhaust system have been cracks in the welds at the mufflers and deformation of the pipes at the exhaust headers. I once had the welded bracket at the tail pipe hanger break off at the weld so now I carry stainless steel worm drive hose clamps and angle brackets as a temporary repair to get home. I also had slip joint brackets welded to the pipes break off at the welds.

A ball joint at the muffler input and output might relieve the vibration stress at the muffler welds. I think the problems with the brackets breaking at the welds are due to the use of fairly thin stainless steel pipes in those locations and too little metal in the weld joints. My solution would be to weld a thicker sleeve to the pipes at those spots so that there is more metal in the weld. That would save weight versus using thicker stainless steel pipes in the entire exhaust system. It would also strengthen the pipes at the slip joints and resist deformation. Even better would be a muffler design without welds where the pipes connect to the muffler. Pipe flanges bolted to thicker muffler end plates might do the trick. A pipe flange would have more metal and no weld joint inconsistencies at the right angle joint. Re-designing the muffler to have thicker end plates with threaded studs to connect to the flanges would strengthen the connection of the pipes to the muffler and make it possible to replace the muffler without replacing the pipes or flanges.

[Kirk]

Andy,

I’m not sure the service life you are experiencing is that much out of line. 500 hours does seem a bit short, but I’ve never made 1,000 hours without repairing flame tubes. Your analysis of the exhaust systems we run makes sense, not much we can do about that.

Best life I’ve gotten out of mufflers were from Dawley Aviation, top notch work. I wonder if Atlee Dodge and Dawley would tell what metal they use in their builds and what would be best for service life?

I can’t help you with LOP since the carbureted O540-J1A5D I’ve been running has too wide an EGT spread. My recollection of Lycoming’s SL is that they don’t recommend running at peak EGT until 70% or less, I ran my engine at peak EGT for over 1,000 hours but more in the 65% or less range since I cruise at 8-9,000’. Don’t know how high you cruise, but higher might be better.

A great series of articles on engine management was written by John Deakin in his column Pelican’s Perch. They are archived on AvWeb.com. You might get some ideas from a few of them I will reference here.

https://www.avweb.com/features/avweb-cl ... ne-part-1/
Pelican’s Perch #63 discusses idle mixture and setting full redline fuel flow. Could the problem be in takeoff and climb rather than cruise due to an excessively lean setting?

https://www.avweb.com/features/avweb-cl ... the-climb/
Pelican’s Perch #64 discusses takeoff EGT and a discussion of LOP in climb that makes a lot of sense to me.

https://www.avweb.com/features/avweb-cl ... -3-cruise/
Pelican’s Perch #65 discusses LOP in depth.
The J1A5D I have been running has the same rpm limits as the W1A5D you have. I mostly cruised at 2,000 rpm vs the 2,200 you have been using. Take a look at the “Motored Friction VS RPM chart which rather advocates lower rpm cruise. I’m speculating that lower rpm may help muffler life since you are pumping a lower volume of hot gasses through the exhaust.

Check out the “Outside the Box” paragraph for explanation of his (and George Braly’s) reasoning for their mixture use theories.

I think you are on the right track considering LOP. It’s hard to sort out anecdotal information from factual information. The articles I mentioned seem well thought out and logical so I choose to follow much of the advice they contain. Hope you find some useful nuggets in there.

[andy]

I haven't used Atlee Dodge mufflers but Dawley rebuilt my exhaust system a couple of times over the 26 years that I've owned my MX-7-180 and I've never had trouble with their flame tubes.

[Andy Young]

This is exactly the kind of thoughtful discussion I was hoping for. I read all the three of these linked articles carefully. He makes some very good points, but not all of it rings true to me. I think he’s mostly focussed on big Continentals, and from my experience, those are much more sensitive to operate. I’m particular, the red box at 75% power doesn’t seem to hold for a Lycoming 540 with an 8.5:1 compression ratio. Certainly the issue with max fuel flow being set too low, which he mentions frequently, is specifically a Continental problem. The fuel injection systems are VERY different in concept and operation between these two manufacturers. Continentals (which I have flown for work) do have an easy adjustment for max fuel flow that is often set incorrectly. Anyway, according to his charts, at 22 squared, I’m deep in the red box. And yet, I’ve been doing exactly this for thousands of hours, and never see CHTs over about 320°. It’s worth noting that while Continental says to run at peak EGT only when at 65% power and below, Lycoming recommends doing so at 75% and below.

To answer your question, I typically cruise between about 1000’ and 2500’ msl, higher only as needed to clear terrain, which can push me up to about 12,000’ now and then. However, in my personal flying, I spend the vast majority of my time below 5000’. Very different from your operation by the sounds of it.

I don’t think my problem is being too lean in the climb, as (contrary to what the articles recommend) I reduce power shortly after breaking ground, and am at 75% power or less in the climb. I’m typically only climbing to a few thousand feet anyway, and reducing power lets me lean right away. Even if I’m going up to say 5000 or 6000’, I usually have at least 25 miles or so to get there, so no need to climb aggressively; I prefer a gradual cruise climb, matching the terrain as it rises. My EGTs in the climb have always been about what they are in cruise. No surprise, as I’m usually doing cruise climbs. I’ll pay closer attention to what the EGTs look like during take-off and initial climb (before reducing power).

In any case, good point about running lower RPM to reduce frictional losses and possibly reduce the impact on the flame tubes. I have tried 2000 RPM, but it just sounds wrong at this RPM to me. Maybe that’s just because I’m so used to what it sounds like at 2200. I’ll try it again and see if I can get used it, while paying careful attention to all the numbers to see what benefits there are (if any significant).

Atlee Dodge does use a different (they claim better) stainless alloy from what Maule uses, and it’s thicker. They’ve told me the alloy number, but I can’t remember it at the moment. Not sure if Dawley is using the same as either of them. The Atlee mufflers do last a lot longer than the Maule units.

Andy: Interesting ideas on redesigning the system. Too bad it takes an act of congress these days to get approval to make changes. I’d like to see a system with the muffler held securely somewhere as far as possible from the exhaust flanges, and flexible pipes (ball joints in all the right places, etc.) from flanges to muffler.

[Kirk]

Andy,

I agree the articles are largely Conti centric so I try to sieve the useful nuggets. It is also centered on injected engines, though many principles can be universally applied. I too am a Lycoming guy and am thankful for the robust durability.

The famed “red box” concept is at the center of George Braly’s seminars and articles. I think it does have some validity for all engines in terms of long term wear. As Deakins says, “cylinders are cylinders”. It really should be an amber box since we all operate in it for some periods during a flight.

That being said, I do think you are on the right track considering 50 LOP. I played with cruise rpm a lot when I first started with my O540-J1A5D. Being a curious sort, I explored the lowest rpm settings. It gave me the best speed and range at 2,000 rpm and I liked the reduced noise.

I vaguely recall reading somewhere that best propeller efficiency is around 1400 rpm but engine efficiency drops off considerably before then. I tried 1800 rpm (respecting the MP limit placarded for that Hartzell prop) and got greater vibration. Interestingly, it was hard to get a definitive peak EGT. That told me 1800 wasn’t good for a carbureted engine with a 2 blade metal prop. It may actually work for you, but more in the 65% range.

What Lycoming SL did you find that recommended peak EGT at 75%? I thought peak or lean of peak wasn’t recommended until 70% or less. I searched and didn’t find it. But is 5% really a meaningful difference.

Kirk

[Andy Young]

The 75% recommendation is found on page 3-5 of the Lycoming Operator’s Manual for the O-540 and IO-540 series. The manual is part # 60297-10.

[Kirk]

Got it. I was looking in Service Instructions. SI 1094D being most current and all I could find is even more vague than the Operators Manual.

Kirk