Van's Air Force
Don't miss anything! Register now for full access to the definitive RV support community.
Superior cylinder rocker boss failure
- Thread starter Greg Arehart
- Start date
Greg Arehart
Well Known Member
Update
Friends,
As of this afternoon (3 pm), no cylinder. I called Eagle, who have been trying to call ECI but nobody's home at ECI. Daniel at Eagle pulled some info off of their website (tracking info) and it appears that the cylinder was shipped 12/27 (after being ordered 12/21 and told it would be shipped 12/22) for delivery 1/3. It's not clear what happened but Daniel is still trying to find somebody at ECI to query.
Weather here sux until Friday anyway.....
Frankh - as far as I can tell, this is on my dime. I've already paid for the new cylinder. Maybe down the road I might see something, but this is a cylinder from the "old" Superior that went bankrupt so they probably have no liability.
Steve,
Sounds like an interesting theory. I can clearly see the scratches on the bosses from the press-fitting of the bearings. The bearings essentially fell out of the broken boss when I opened up the rocker cover. Would this type of failure likely take 335 hours to develop (that's the time on the engine since new)? I will be taking parts to my friendly neighborhood metallurgist after the repairs are done in hopes of getting some answers.
cheers,
greg
Friends,
As of this afternoon (3 pm), no cylinder. I called Eagle, who have been trying to call ECI but nobody's home at ECI. Daniel at Eagle pulled some info off of their website (tracking info) and it appears that the cylinder was shipped 12/27 (after being ordered 12/21 and told it would be shipped 12/22) for delivery 1/3. It's not clear what happened but Daniel is still trying to find somebody at ECI to query.
Weather here sux until Friday anyway.....
Frankh - as far as I can tell, this is on my dime. I've already paid for the new cylinder. Maybe down the road I might see something, but this is a cylinder from the "old" Superior that went bankrupt so they probably have no liability.
Steve,
Sounds like an interesting theory. I can clearly see the scratches on the bosses from the press-fitting of the bearings. The bearings essentially fell out of the broken boss when I opened up the rocker cover. Would this type of failure likely take 335 hours to develop (that's the time on the engine since new)? I will be taking parts to my friendly neighborhood metallurgist after the repairs are done in hopes of getting some answers.
cheers,
greg
frankh
Well Known Member
Interesting
Not sure of the circumstances but when you buy a compNy you also buy it's liabilities. I would have thought that if Superior wants to do business with a tightly knit group of customers (us) they would be delighted to fix this at their cost. If they don't I certainly won't be buying an engine from them.
Frank
Not sure of the circumstances but when you buy a compNy you also buy it's liabilities. I would have thought that if Superior wants to do business with a tightly knit group of customers (us) they would be delighted to fix this at their cost. If they don't I certainly won't be buying an engine from them.
Frank
Mike S
Senior Curmudgeon
Any of those scratches in the immediate area of where the part broke?
Greg Arehart
Well Known Member
Yes, there are scratches everywhere on the inside of the bosses. There are enough of these scratches that one could perhaps suggest that they were the start of the fracture, but one could also argue the other way as well. I can see no indication of beach marks (as DanH suggested) anywhere using a 10X lens. Here are a couple photos of the bosses:
Central boss looking at the thin side of the boss down:
One end boss with bearing scratches:
(I have higher-definition photos if anyone wants them - these were degraded a bit to fit in the space available)
greg
Greg Arehart
Well Known Member
Mike,
I agree that the break seems to go along this scratch. Really needs a close look by someone like DanH who is familiar with failure modes etc.
cheers,
greg
I agree that the break seems to go along this scratch. Really needs a close look by someone like DanH who is familiar with failure modes etc.
cheers,
greg
I've never installed a rocker shaft bushing. It isn't something generally done in the field. Fresh cylinder assemblies come with the bushings already installed; we just check the shaft fits.
However, the numbers are in the books. My Lycoming manual says the bushing OD to boss ID press fit should be between 0.0022" and 0.0038" interference. The actual boss ID is 0.7380"-0.7388", so you would need bushing OD between 0.7410" and 0.7418". Not much room for error.
It's a pretty tight press fit for a hole less than 3/4". The manual says nothing about heating; just drive in the bushings using the specified drivers. So yeah, it would be easy to gall the bores, even with correct fits. Nobody can tell you what the interference fit may have been by looking at a photo of the galling.
I tend to discount the "bushing too big" theory. If it was so big that it cracked the boss during installation, it should have grown some beach marks between then and now (partial crack), or show contact damage at the faces of the break (fully broken). If it didn't crack at installation, the bushing oversize shouldn't make any difference. I think the additional hoop stress would act like bolt preload, reducing the fatigue effect of cyclical loading.
Given clean crystaline faces in the break, I think you're still looking at a one-shot overload. A galling gouge could serve as a "break here" stress concentration.
Obviously I like the subject, but again don't depend too much on me. I'm not a professional.
POSTSCRIPT: No point in leaving stones unturned. You can easily measure the bushing OD's. It may be difficult to get a good measurement of the boss ID's (given that they're broken into half-moons), but they were cut with a piloted reamer anyway. It's tough to get an undersize hole with a reamer.
However, the numbers are in the books. My Lycoming manual says the bushing OD to boss ID press fit should be between 0.0022" and 0.0038" interference. The actual boss ID is 0.7380"-0.7388", so you would need bushing OD between 0.7410" and 0.7418". Not much room for error.
It's a pretty tight press fit for a hole less than 3/4". The manual says nothing about heating; just drive in the bushings using the specified drivers. So yeah, it would be easy to gall the bores, even with correct fits. Nobody can tell you what the interference fit may have been by looking at a photo of the galling.
I tend to discount the "bushing too big" theory. If it was so big that it cracked the boss during installation, it should have grown some beach marks between then and now (partial crack), or show contact damage at the faces of the break (fully broken). If it didn't crack at installation, the bushing oversize shouldn't make any difference. I think the additional hoop stress would act like bolt preload, reducing the fatigue effect of cyclical loading.
Given clean crystaline faces in the break, I think you're still looking at a one-shot overload. A galling gouge could serve as a "break here" stress concentration.
Obviously I like the subject, but again don't depend too much on me. I'm not a professional.
POSTSCRIPT: No point in leaving stones unturned. You can easily measure the bushing OD's. It may be difficult to get a good measurement of the boss ID's (given that they're broken into half-moons), but they were cut with a piloted reamer anyway. It's tough to get an undersize hole with a reamer.
Last edited:
E. D. Eliot
Well Known Member
My opinion
Of course it's only my opinion - and I am not a aircraft engine mechanic, etc., but if this structural failure was from an engine that I put together or if from a company that I owned, I would want very much to 'purchase' your engine from you so that my engineers and mechanics could study it and so that you wouldn't fly it any more. If it was my company (and product reputation), I would offer you a new engine - not just parts of an engine - in trade.
I think - and am not qualified in any way as an expert - but I think that there are too many unknowns here - I am very concerned about further flight using that engine as it is at this time. Sorry, I don't want to be critical but I know how else to put it. Please be careful.
Of course it's only my opinion - and I am not a aircraft engine mechanic, etc., but if this structural failure was from an engine that I put together or if from a company that I owned, I would want very much to 'purchase' your engine from you so that my engineers and mechanics could study it and so that you wouldn't fly it any more. If it was my company (and product reputation), I would offer you a new engine - not just parts of an engine - in trade.
I think - and am not qualified in any way as an expert - but I think that there are too many unknowns here - I am very concerned about further flight using that engine as it is at this time. Sorry, I don't want to be critical but I know how else to put it. Please be careful.
Last edited:
Ron Lee
Well Known Member
Greg, Mr Eliot made a critical, and correct in my opinion, point. You can make a valid case that the entire engine is suspect. That may mean just the other three cylinders. But another incident like this could end up far worse.
I do not understand all the issues made here about benching and other new terms to me, but my suggestion is that when you return the plane to service, if you do, that you fly over expressways or suitable terrain at high altitudes to get home.
The same incident may never happen again. It could have been a one time aberration with that cylinder. I have no idea. This seems like a time to be extra cautious. I suspect that you already came to the same conclusion.
I do not understand all the issues made here about benching and other new terms to me, but my suggestion is that when you return the plane to service, if you do, that you fly over expressways or suitable terrain at high altitudes to get home.
The same incident may never happen again. It could have been a one time aberration with that cylinder. I have no idea. This seems like a time to be extra cautious. I suspect that you already came to the same conclusion.
Greg Arehart
Well Known Member
Yes, if/when I return this to Reno, I will be flying high and above terrain suitable for making an emergency landing if necessary (probably along highway 95). I also plan to have an escort plane.
Certainly I will be contacting Superior to see what can be done, but my immediate goal is to get the airplane back safely to my hangar in Reno. I agree that the other cylinders may be suspect, but understanding that will have to wait on a determination of the cause of failure. If there is no obvious one-off cause (like a stuck valve), I am likely to replace all the cylinders.
Al, you are correct - the RV in a puddle of oil (hoping that the rains wash some of that away
). Thanks for the offer of help and I'll put you on my call list.
cheers,
greg
Certainly I will be contacting Superior to see what can be done, but my immediate goal is to get the airplane back safely to my hangar in Reno. I agree that the other cylinders may be suspect, but understanding that will have to wait on a determination of the cause of failure. If there is no obvious one-off cause (like a stuck valve), I am likely to replace all the cylinders.
Al, you are correct - the RV in a puddle of oil (hoping that the rains wash some of that away
). Thanks for the offer of help and I'll put you on my call list.cheers,
greg
Zero4Zulu
Well Known Member
0.0022" and 0.0038" is a very tight fit. That .003" has to move somewhere. Thin wall bushings do require more interference and will collapse as they are pressed in. But the alum. will move too without cracking initially and still have an enormous amount of pressure in that bore. Maybe not relevant here at all, I don't know.....
Compared to the other pictures posted and the set of cylinders I have right here, those particular castings may have never been intended to have the bushings at all. They are so thin.
One thing I haven't seen mentioned is whether the two end bosses would be strong enough to do the job without the center boss. From a purely TLAR standpoint, it looks as though a thin/broken center boss might not lead to all three failing. On the other hand, I think if an end boss failed, the center would go right away, followed by the other end. I've also noticed that all the bosses appear to have failed along a diameter of the bore, except for the one on the left in the original photo. In particular, the lower left looks like it broke well off of the diameter at an area that was actually pretty thick. I'm suggesting that perhaps this was the location of the initial failure and everything else broke as a result. Just a thought.
rocketbob
Well Known Member
I doubt you will get any sort of help from Superior because A. they currently can't supply replacement cylinders and B. since the cylinders were produced before they went Chap. 11, they no longer have to honor the warranty.
With that said, and this is my opinion, your cylinder is entirely repairable. If this happened to me I would have weld repaired the bosses myself. Weld and set the cylinder up on an angle plate and re-bore the rocker pin hole using oversize rocker pin bushings. A cylinder shop can weld up the bosses back together and they could easily build up the wall thicknesses to restore them to better-than-new strength. In fact this can be done to all of your cylinders if the wall thicknesses are similar.
As far as engine welds go, this would be child's play. Similar repairs are done by engine shops routinely. If the other bosses are welded back up you won't need oversize bushings since the boss ID's would shrink after welding. But the holes themselves would have to be cleaned up with a reamer.
With that said, and this is my opinion, your cylinder is entirely repairable. If this happened to me I would have weld repaired the bosses myself. Weld and set the cylinder up on an angle plate and re-bore the rocker pin hole using oversize rocker pin bushings. A cylinder shop can weld up the bosses back together and they could easily build up the wall thicknesses to restore them to better-than-new strength. In fact this can be done to all of your cylinders if the wall thicknesses are similar.
As far as engine welds go, this would be child's play. Similar repairs are done by engine shops routinely. If the other bosses are welded back up you won't need oversize bushings since the boss ID's would shrink after welding. But the holes themselves would have to be cleaned up with a reamer.
Last edited:
rvmills
Well Known Member
As Greg's bud and possible chase plane for the trip home, one of the things we have discussed is pulling off the other three rocker covers and having a look-see at the bosses. An obvious (visible) impending failure would be an instant no-go, of course.
However, there may be nothing obvious or visible. There will be some trained eyes helping Greg, but this may be more specialized territory (boss metalurgic failure, or whatever it really is).
I doubt one should or could get advice on the internet for an acid test that says fly it or don't (and that's no slam on VAF!!) but perhaps we can get a lead on a good "who ya gonna call" person for some "how & what to inspect" gouge before making a go/no-go call, or who to have look at it for Greg. Greg, my first thought would be to call Mahlon, or Ken at Lycon, and just talk about it a bit.
Thoughts?
I know Greg is using an abundance of caution!
Cheers,
Bob
However, there may be nothing obvious or visible. There will be some trained eyes helping Greg, but this may be more specialized territory (boss metalurgic failure, or whatever it really is).
I doubt one should or could get advice on the internet for an acid test that says fly it or don't (and that's no slam on VAF!!) but perhaps we can get a lead on a good "who ya gonna call" person for some "how & what to inspect" gouge before making a go/no-go call, or who to have look at it for Greg. Greg, my first thought would be to call Mahlon, or Ken at Lycon, and just talk about it a bit.
Thoughts?
I know Greg is using an abundance of caution!
Cheers,
Bob
Last edited:
Has anyone called Superior yet? It has been reported that they have honored some warranty work that dated before the new owners took over. Just a thought, it might be worth a call.
Rather than debating how many teeth the horse on the other side of the field has it is easier, but less fun, to just go over there and count them
Rather than debating how many teeth the horse on the other side of the field has it is easier, but less fun, to just go over there and count them
rvmills
Well Known Member
Has anyone called Superior yet? It has been reported that they have honored some warranty work that dated before the new owners took over. Just a thought, it might be worth a call.
Rather than debating how many teeth the horse on the other side of the field has it is easier, but less fun, to just go over there and count them
I think Greg has found that none of the horses on the other side of the field are answering their phones...at all! Could be the season...could be no more teeth...dunno!
Cheers,
Bob
Dye Penetration?
I was thinking about how I wouldget comfortable with other jugs myself Bob. I figure that taking off the rocker covers for a viual would at least give a gross check. I am wondering if a field-type Dye-pen kit might be used on the other bosses to see a little more detail.
I'll admit that I haven't used one for decades, so may have forgotten the limitations.
I was thinking about how I wouldget comfortable with other jugs myself Bob. I figure that taking off the rocker covers for a viual would at least give a gross check. I am wondering if a field-type Dye-pen kit might be used on the other bosses to see a little more detail.
I'll admit that I haven't used one for decades, so may have forgotten the limitations.
Greg Arehart
Well Known Member
Update
Spoke with Bill at Superior this afternoon. He has been talking with Lou at Eagle Engines as well. He volunteered to send me a cylinder but it has no other parts with it (like valves, guides, etc.) and I already have an ECi on the way. Superior is not making cylinders again yet. We discussed a couple of the potential failure causes, and need more data to assess, but apparently this is the only one out there with this failure mode.
Doesn't sound like any fiscal help on the way at this point.
Oh, and the ECi excuse for the delay in sending the cylinder was "holidays, lots of people out of the office" which still isn't excuse enough for telling me one thing and doing another.
greg
Spoke with Bill at Superior this afternoon. He has been talking with Lou at Eagle Engines as well. He volunteered to send me a cylinder but it has no other parts with it (like valves, guides, etc.) and I already have an ECi on the way. Superior is not making cylinders again yet. We discussed a couple of the potential failure causes, and need more data to assess, but apparently this is the only one out there with this failure mode.
Doesn't sound like any fiscal help on the way at this point.
Oh, and the ECi excuse for the delay in sending the cylinder was "holidays, lots of people out of the office" which still isn't excuse enough for telling me one thing and doing another.
greg
Last edited:
rvmills
Well Known Member
Paul,
Dye-Pen was a thought I had as well, but haven't done one since a test kit in college, and no real training, so not something I'd trust (me doing). Perhaps opening them up will reveal that the other bosses look different (the hole is drilled in the center of the bosses, the metal looks less granular, etc., but who knows if those are even issues at this point). This is a tough call.
Greg,
I'd recommend getting the Superior cylinder, even if you don't use it. Would be a good comparison piece for the old one (and the other 3), and it would be good to see what the rocker bosses look like off the shelf. Just a thought.
Cheers,
Bob
Mike S
Senior Curmudgeon
An old trick I have used successfully before, clean things with
Brake Clean, then heat the part with a hair dryer. If there are any cracks there, the oil which has wicked into the cracked part will start to wet the surface where the crack is.
I would stuff the pushrod tubes with CLEAN rags to keep the Brake Clean out of them.
Good luck.
Brake Clean, then heat the part with a hair dryer. If there are any cracks there, the oil which has wicked into the cracked part will start to wet the surface where the crack is.
I would stuff the pushrod tubes with CLEAN rags to keep the Brake Clean out of them.
Good luck.
Greg Arehart
Well Known Member
Mike,
Thanks for the idea - I think we'll give this a try.
Bob,
The cylinder they have is one of a different series (I can't remember what it's called now, but different casting technique etc. from the ones I have). Bill also said that the wall thickness on the center boss is essentially irrelevant - all of the strength is in the thick end parts of the boss.
cheers,
greg
Thanks for the idea - I think we'll give this a try.
Bob,
The cylinder they have is one of a different series (I can't remember what it's called now, but different casting technique etc. from the ones I have). Bill also said that the wall thickness on the center boss is essentially irrelevant - all of the strength is in the thick end parts of the boss.
cheers,
greg
rvmills
Well Known Member
Update from Greg
Greg asked me to pass this update from today (1/5):
We met today in Jean and accomplished the cylinder removal and replacement...with much help, and a little weather glitch at the outset. Greg and Dan (Guccidude1) Ross headed to Reno-Stead at 0630, but were stuck there till noon waiting out freezing fog at the airport (and only at the airport...go figure!).
In the meantime, I reclaimed my plane from now-snowless Agua Dulce, and headed to Jean to start taking things apart till Greg and Dan arrived. I was joined there by Rich (AI), Terry (VAF, RV-7A) and Brian (VAF, future RVer).
Everything came off/apart easily, and the pics in the OP represent it well...though I will say, to my untrained-in-metallurgy eyeball, the bosses' granularity looks even larger than in the pics. Also, to my untrained eye, the position of the rocker arm shaft holes in the bosses looked the same on all four cylinders...and no glaring external defects were noted on the bosses on the other 3 cylinders. There was no visual evidence of either valve hitting the pistion either. Those are just my FWIW comments, of course. Rich did a thorough visual inspection as well.
One thing we found was that the exhaust valve push rod was not true. It didn't roll well on the window on the airport building (the flattest thing we could find). Not radically bent, but the bend was slightly visible, and we could all feel the bend when rolling it. The intake valve push rod did roll well. Both valves did seem to move a bit when tested, but that was with the springs still on, so no definitive level of movement or stickiness could be determined with the tools on hand...of course that cylinder is going into a packing box and getting a ride home, then inspected.
When Greg and Dan showed up, I had to launch for Stead to beat sunset (gotta work tomorrow), so we did a hand-off, and by the time I got to Reno, they had the cylinder back on, test run, and test flown. Greg and Dan will camp at the casino hotel at Jean tonight, then make another test flight to a nearby airport to get gas (the pump at Jean is down), then will fly home to Stead with Dan as chase for Greg. Once home, Greg will be able to inspect and work all issues more deeply and effectively, and I know he will continue to post info as he peels the onion.
I know he's very grateful for all the VAF support on the forum, and for the help from the gang at the airport. I know he'll be writing more when he gets home (he didn't want to post from the iPhone tonight).
He sounded bushed, but I told him to be sure to let it all ride on double-zero...and get this engine work paid for in a few spins!
More to follow from Greg.
Cheers,
Bob
Greg asked me to pass this update from today (1/5):
We met today in Jean and accomplished the cylinder removal and replacement...with much help, and a little weather glitch at the outset. Greg and Dan (Guccidude1) Ross headed to Reno-Stead at 0630, but were stuck there till noon waiting out freezing fog at the airport (and only at the airport...go figure!).
In the meantime, I reclaimed my plane from now-snowless Agua Dulce, and headed to Jean to start taking things apart till Greg and Dan arrived. I was joined there by Rich (AI), Terry (VAF, RV-7A) and Brian (VAF, future RVer).
Everything came off/apart easily, and the pics in the OP represent it well...though I will say, to my untrained-in-metallurgy eyeball, the bosses' granularity looks even larger than in the pics. Also, to my untrained eye, the position of the rocker arm shaft holes in the bosses looked the same on all four cylinders...and no glaring external defects were noted on the bosses on the other 3 cylinders. There was no visual evidence of either valve hitting the pistion either. Those are just my FWIW comments, of course. Rich did a thorough visual inspection as well.
One thing we found was that the exhaust valve push rod was not true. It didn't roll well on the window on the airport building (the flattest thing we could find). Not radically bent, but the bend was slightly visible, and we could all feel the bend when rolling it. The intake valve push rod did roll well. Both valves did seem to move a bit when tested, but that was with the springs still on, so no definitive level of movement or stickiness could be determined with the tools on hand...of course that cylinder is going into a packing box and getting a ride home, then inspected.
When Greg and Dan showed up, I had to launch for Stead to beat sunset (gotta work tomorrow), so we did a hand-off, and by the time I got to Reno, they had the cylinder back on, test run, and test flown. Greg and Dan will camp at the casino hotel at Jean tonight, then make another test flight to a nearby airport to get gas (the pump at Jean is down), then will fly home to Stead with Dan as chase for Greg. Once home, Greg will be able to inspect and work all issues more deeply and effectively, and I know he will continue to post info as he peels the onion.
I know he's very grateful for all the VAF support on the forum, and for the help from the gang at the airport. I know he'll be writing more when he gets home (he didn't want to post from the iPhone tonight).
He sounded bushed, but I told him to be sure to let it all ride on double-zero...and get this engine work paid for in a few spins!
More to follow from Greg.
Cheers,
Bob
Last edited:
He sounded bushed, but I told him to be sure to let it all ride on double-zero...and get this engine work paid for in a few spins!
More to follow from Greg.
Cheers,
Bob
No kidding, a few good rolls at the craps table and this could be a non-event!
Amazed to hear all the support, just plane awsome (misspelled pun intended). Lets see the Bonanza owners association come together like that.
Glad to hear everything went smooth. In 6 months this whole thing will just be an after thought and "let me tell you about a time" story.
Greg Arehart
Well Known Member
N7965A is now safely in my hangar at RTS after an uneventful flight home.
When Bob left we had just arrived and brought the new cylinder. Rich Mier and Terry Frazier did the majority of the work and it was incredible to see how fast someone who knows what they are doing can put stuff back together! As Bob said, the exhaust pushrod is clearly bent slightly, so that is part of the picture.
We examined the camshaft, other cylinder heads, and everything in general and could see no other issues, so we buttoned it up and did a ground run on the engine. No leaks or other issues, so I went flying around the pattern a couple times and all seemed normal (except for all the leftover oil blowing around). It was getting dark so we called it a day.
This morning Dan and I had a leisurely breakfast waiting for the heavy frost (yes, frost in Las Vegas!) to dissipate. From there we flew to HND for fuel (and because some of our new mechanic friends are there!). All instruments normal, though the new cylinder has elevated CHTs as expected before breakin. Ground check just revealed more oil blowing around. From there we launched for RTS and had an uneventful flight north - some of the smoothest air I've seen. Arriving in Reno around 1 pm we spotted the dreaded ice fog again. Fortunately, it was (finally) starting to break up and we were able to get into RTS. 15 minutes earlier and we would have had to stop short. By the time I taxied to the hangar, it was sunny and getting warm.
I have the cylinder and (with the help of some other folks) will be trying to figure out just what caused the failure. I will post whatever I find. I will also be looking more closely at the remaining cylinder heads for any suggestions that they might follow suit.
My thanks go out to all of the RV community for helping me out on this. I cannot say enough good things about this group.
Finally, a shout-out for Rich Mier, who really oversaw the replacement process. I hope he doesn't mind my posting his number here in case anyone needs a good mechanic: 702-275-7114. I only wish it had gone a bit slower so I could have learned more!
Unfortunately, I didn't have the energy to roll the dice last night and try for a brand-new engine. I'm happy enough to be back in the air again even if it is with one grey cylinder and three gold ones.
cheers,
Greg
p.s. changed my signature back again
When Bob left we had just arrived and brought the new cylinder. Rich Mier and Terry Frazier did the majority of the work and it was incredible to see how fast someone who knows what they are doing can put stuff back together! As Bob said, the exhaust pushrod is clearly bent slightly, so that is part of the picture.
We examined the camshaft, other cylinder heads, and everything in general and could see no other issues, so we buttoned it up and did a ground run on the engine. No leaks or other issues, so I went flying around the pattern a couple times and all seemed normal (except for all the leftover oil blowing around). It was getting dark so we called it a day.
This morning Dan and I had a leisurely breakfast waiting for the heavy frost (yes, frost in Las Vegas!) to dissipate. From there we flew to HND for fuel (and because some of our new mechanic friends are there!). All instruments normal, though the new cylinder has elevated CHTs as expected before breakin. Ground check just revealed more oil blowing around. From there we launched for RTS and had an uneventful flight north - some of the smoothest air I've seen. Arriving in Reno around 1 pm we spotted the dreaded ice fog again. Fortunately, it was (finally) starting to break up and we were able to get into RTS. 15 minutes earlier and we would have had to stop short. By the time I taxied to the hangar, it was sunny and getting warm.
I have the cylinder and (with the help of some other folks) will be trying to figure out just what caused the failure. I will post whatever I find. I will also be looking more closely at the remaining cylinder heads for any suggestions that they might follow suit.
My thanks go out to all of the RV community for helping me out on this. I cannot say enough good things about this group.
Finally, a shout-out for Rich Mier, who really oversaw the replacement process. I hope he doesn't mind my posting his number here in case anyone needs a good mechanic: 702-275-7114. I only wish it had gone a bit slower so I could have learned more!
Unfortunately, I didn't have the energy to roll the dice last night and try for a brand-new engine. I'm happy enough to be back in the air again even if it is with one grey cylinder and three gold ones.
cheers,
Greg
p.s. changed my signature back again
frazitl
Well Known Member
Glad you had a good flight home.
Linda and I watched you on the APRS just to be sure.
I talked to an experienced mechanic yesterday afternoon and he was convinced that the bent push rod was almost certainly caused by a sticking exhaust valve. Sticking could be caused by a number of issues including manufacturing clearances or carbon buildup at the top of the list. His recommendation (and mine too having thought it over) was to do a wobble test on the other three cylinders valves. I did feel a small amount of sticking just off the seat on the exhaust valve of the failed cylinder, but that might have just been me hoping to find a logical cause for the failure. A wobble test on the failed cylinder would be very interesting...
Think I'm going to do this test on my engine during the next annual.
Linda and I watched you on the APRS just to be sure.
I talked to an experienced mechanic yesterday afternoon and he was convinced that the bent push rod was almost certainly caused by a sticking exhaust valve. Sticking could be caused by a number of issues including manufacturing clearances or carbon buildup at the top of the list. His recommendation (and mine too having thought it over) was to do a wobble test on the other three cylinders valves. I did feel a small amount of sticking just off the seat on the exhaust valve of the failed cylinder, but that might have just been me hoping to find a logical cause for the failure. A wobble test on the failed cylinder would be very interesting...
Think I'm going to do this test on my engine during the next annual.
Last edited:
Mike S
Senior Curmudgeon
Yep
This is exactly what I thought as I read about the push rod being bent.
Just a WAG here, but I suspect a stuck valve was the genesis of the problem-----and the rocker boss sacrificed itself to prevent further damage to the push rod.
This is exactly what I thought as I read about the push rod being bent.
Just a WAG here, but I suspect a stuck valve was the genesis of the problem-----and the rocker boss sacrificed itself to prevent further damage to the push rod.
Doug
Well Known Member
I'd be interested to learn if the wall thickness of the pushrod 'tube' differs in any way to the regular Lycoming part.
An increase in thickness (or material type) would significantly increase the buckling force required, resulting in a corresponding increase in the load on the rocker boss when the valve sticks.
Even with the reduced area across the failure face of these bosses I would have thought the material loading would not have incressed by more than a factor of say 1.5 yet we see what appears to be an atypical single event failure mode.
Something is not right (stating the obvious), I hope you can discover what is going on.
Also, someone suggested that an exhaust valve is most likely to stick at the initial power reduction? If this is true - why?
Doug Gray
An increase in thickness (or material type) would significantly increase the buckling force required, resulting in a corresponding increase in the load on the rocker boss when the valve sticks.
Even with the reduced area across the failure face of these bosses I would have thought the material loading would not have incressed by more than a factor of say 1.5 yet we see what appears to be an atypical single event failure mode.
Something is not right (stating the obvious), I hope you can discover what is going on.
Also, someone suggested that an exhaust valve is most likely to stick at the initial power reduction? If this is true - why?
Doug Gray
Doug
Well Known Member
Dan, Yes I see you did. Your observation about the power reduction seems to have been spot on so I was wondering what you had in mind.
What do you suppose is the significance of the power reduction in the onset of the failure?
Knowing now that a stuck exhaust valve is a possible candidate cause - what does that add to the discussion?
I have in mind that lead condensing and solidifying on the valve stem may be the mechanism for a sticking valve but this would require a massively fast reuction in temperature.
I found this interesting:
http://www.avweb.com/news/maint/182894-1.html
Doug Gray
What do you suppose is the significance of the power reduction in the onset of the failure?
Knowing now that a stuck exhaust valve is a possible candidate cause - what does that add to the discussion?
I have in mind that lead condensing and solidifying on the valve stem may be the mechanism for a sticking valve but this would require a massively fast reuction in temperature.
I found this interesting:
http://www.avweb.com/news/maint/182894-1.html
Doug Gray
Mike S
Senior Curmudgeon
Remember "shock cooling"???
Air cooled engine------
Air is cooling the exterior of the engine, and residual heat is still inside-----
Steel cools slower than alum.
Power reductions stop the balancing effect of adding heat to the engine innards, but the plane is still going through the air, so the cooling is still happening on the outside.
Net result, things tighten up inside.
This is an over-simplification, but basically the info applies.
Air cooled engine------
Air is cooling the exterior of the engine, and residual heat is still inside-----
Steel cools slower than alum.
Power reductions stop the balancing effect of adding heat to the engine innards, but the plane is still going through the air, so the cooling is still happening on the outside.
Net result, things tighten up inside.
This is an over-simplification, but basically the info applies.
Temperature change.
I'm waiting for somebody to remove the springs and report an actual sticky valve.
Not neccessarily. We tend to think in terms of CHT, but actual component temps are different. There's a chart in Taylor's "Internal Combustion.." with temperature measurements at the mid point of an exhaust valve guide for a large air-cooled aircraft cylinder The range is about 425F to well over 600F. The valve itself is 1000F to 1200F at the face and stem temperature would be a gradient. Elemental lead has a melting point of 622F. Liquid to solid is a range of just a few degrees, and the neighborhood temps range all around the melt point.
Me too.
az_gila
Well Known Member
Doesn't the sodium filling of our exhaust valves even this gradient out a lot - as well as explaining why exhaust valves cost so much...
Greg Arehart
Well Known Member
More data. Here is the set of downloaded data from the Dynon engine monitor (larger version at http://my.imageshack.us/v_images.php). Data were taken in 5-second intervals. Cylinder #2 is the red lines. CHT & fuel flow are plotted on the right-hand scale, and RPM and EGT on the left-hand scale. Looking at this in detail indicates that the dropoff in EGT (and start of dropoff in CHT, though it is buffered by the thermal mass of the cylinder) occurred 10-15 seconds after I reduced power.
For the record, I routinely run premium autogas in my engine. When I left Reno the day before, I had a full tank of autogas. I filled up with avgas in Ely, and then again in Jean before this flight occurred, so the fuel in the tanks was probably >90% avgas at the time of failure.
I have not had the opportunity to look at or do tests on the valves at this point (too cold in the hangar!).
In looking back over data from the last 5 months I notice that when I pull back power in the pattern to land, the #2 and #3 EGTs typically drop more than #1 & #4 (not always, but 75% of the time). I have not yet gone back to older data to look at patterns. With the new cylinder, I see the same drop, so am thinking that it was not an indicator of imminent failure (or maybe #3 is about to fail). Here's a typical pattern when I was doing some landing practice:
cheers,
greg
For the record, I routinely run premium autogas in my engine. When I left Reno the day before, I had a full tank of autogas. I filled up with avgas in Ely, and then again in Jean before this flight occurred, so the fuel in the tanks was probably >90% avgas at the time of failure.
I have not had the opportunity to look at or do tests on the valves at this point (too cold in the hangar!).
In looking back over data from the last 5 months I notice that when I pull back power in the pattern to land, the #2 and #3 EGTs typically drop more than #1 & #4 (not always, but 75% of the time). I have not yet gone back to older data to look at patterns. With the new cylinder, I see the same drop, so am thinking that it was not an indicator of imminent failure (or maybe #3 is about to fail
). Here's a typical pattern when I was doing some landing practice:
cheers,
greg
Last edited:
rvmills
Well Known Member
Greg and I just talked on the phone, and are meeting at the airport this morning to take the cylinder over to our AI bud's hangar and pull the valve springs to see how sticky the ex valve is. I don't think the AI has the tool for the wobble check, but the nearby engine shop may. I'm sure Greg will post an update after we look into things today.
Good discussion on power reduction, temps and valves. I found the article interesting as well, especially the discussion of second order impacts of sticking valves, which included rocker boss failure. Too early to tell, of course, but it was good info.
With reference to the power reduction and temperature changes, I talked to Greg about this to confirm the phase of flight that day. Greg had just departed Jean (2000' MSL) and was leveling off at about 4000' MSL (probably 4500' for the short VFR flight he was on). He had just started a slight power reduction from about 2500 RPM to about 2300 RPM for cruise, when the engine started running rough. That would not seem to be such a radical power reduction as to cause rapid cooling, as Mike was referring to. Certainly internal temps were changing, and it would be hard to determine how much valve stem temps were changing...but from the graphs above, you can see fuel flow drop at the power reduction, and CHTs peak, then drop a bit, while EGTs climb a bit...except for #2.
Not sure where the mixture was that day, but Greg and I talked a little more about leaning, and given that we fly out of Reno at 5K, with 6-7K DAs on the ground not uncommon in the summer, we are aggressive leaners on the ground, and use EGTs and fuel flow to ensure we are not over- or under-leaned during the climb (think Deakin). I use my plugs as indicators of whether I have been aggressive enough on the ground, and even with aggressive leaning, I still sometimes find deposits. We are constantly working this!!
So were lead or other corrosive deposits a factor in a sticky valve? We'll look it over good to see. It will be interesting to see how the clearances at the valve guide look and feel, and just how clean or dirty things look. Sounds like Greg just pulled up, so back atcha with more data later.
Cheers,
Bob
Good discussion on power reduction, temps and valves. I found the article interesting as well, especially the discussion of second order impacts of sticking valves, which included rocker boss failure. Too early to tell, of course, but it was good info.
With reference to the power reduction and temperature changes, I talked to Greg about this to confirm the phase of flight that day. Greg had just departed Jean (2000' MSL) and was leveling off at about 4000' MSL (probably 4500' for the short VFR flight he was on). He had just started a slight power reduction from about 2500 RPM to about 2300 RPM for cruise, when the engine started running rough. That would not seem to be such a radical power reduction as to cause rapid cooling, as Mike was referring to. Certainly internal temps were changing, and it would be hard to determine how much valve stem temps were changing...but from the graphs above, you can see fuel flow drop at the power reduction, and CHTs peak, then drop a bit, while EGTs climb a bit...except for #2.
Not sure where the mixture was that day, but Greg and I talked a little more about leaning, and given that we fly out of Reno at 5K, with 6-7K DAs on the ground not uncommon in the summer, we are aggressive leaners on the ground, and use EGTs and fuel flow to ensure we are not over- or under-leaned during the climb (think Deakin). I use my plugs as indicators of whether I have been aggressive enough on the ground, and even with aggressive leaning, I still sometimes find deposits. We are constantly working this!!
So were lead or other corrosive deposits a factor in a sticky valve? We'll look it over good to see. It will be interesting to see how the clearances at the valve guide look and feel, and just how clean or dirty things look. Sounds like Greg just pulled up, so back atcha with more data later.
Cheers,
Bob
Mike S
Senior Curmudgeon
Agreed, sounds like that is not a significant change......
Doug
Well Known Member
Interesting plots - especially the incident flight. Can you expand the data for the first 30 datapoints and turn off any plot smoothing? I suspect this event occurred very rapidly and the smoothing may mask some temperature variations.
Some observations:
1. I would have characterised your engine as a cool running engine. I wish my temps were as low as these.
2. Prior to the failure the Temps on Cylinder 2 are marginally cooler than the other cylinders. This is unlike the temps seen on the earlier plot.
3. The failure occurred at the highest if not the peak EGT temperature.
4. Difficult to be certain with the interaction of temperatures, fuel flow and rpm but it appears that the engine was leaned prior to reducing rpm/power (FP prop). Alternatively some short term restriction to fuel flow.
5. Following cyl #2 failure - the rpm drops to less than 2000, Cyl #1 EGT falls as if fuel had been cut off to that cylinder. Later the same happens to cylinder #3 but at a lower rpm. This is not seen in the earlier plot.
Perhaps there was some form of FI, flow divider blockage.
Very odd!
Doug Gray
Some observations:
1. I would have characterised your engine as a cool running engine. I wish my temps were as low as these.
2. Prior to the failure the Temps on Cylinder 2 are marginally cooler than the other cylinders. This is unlike the temps seen on the earlier plot.
3. The failure occurred at the highest if not the peak EGT temperature.
4. Difficult to be certain with the interaction of temperatures, fuel flow and rpm but it appears that the engine was leaned prior to reducing rpm/power (FP prop). Alternatively some short term restriction to fuel flow.
5. Following cyl #2 failure - the rpm drops to less than 2000, Cyl #1 EGT falls as if fuel had been cut off to that cylinder. Later the same happens to cylinder #3 but at a lower rpm. This is not seen in the earlier plot.
Perhaps there was some form of FI, flow divider blockage.
Very odd!
Doug Gray
Last edited:
Greg Arehart
Well Known Member
Doug,
Here is more detail at the beginning. No smoothing, data taken in 5-second intervals (original post also is not smoothed). You can see at point 21 that I began leaning as indicated by the fuel flow. A couple points later (24) is where I started to pull power and the failure occurred at point 25. My recollection of events is that the engine actually coughed the first time before I touched either the mixture or throttle.
With regard to your mention of cool-running, I took off when temps were still around freezing, and did not climb aggressively. On most warmer days when taking off my CHTs will push 400 degrees, particularly if I am climbing. But, I don't really have any trouble keeping CHTs in a reasonable range if I climb with rich mixture for cooling. In cruise, CHTs have typically been between 325 and 375 depending on OAT and to a lesser extent, altitude.
I'll be interested in any thoughts you have on the data.
cheers,
greg
Here is more detail at the beginning. No smoothing, data taken in 5-second intervals (original post also is not smoothed). You can see at point 21 that I began leaning as indicated by the fuel flow. A couple points later (24) is where I started to pull power and the failure occurred at point 25. My recollection of events is that the engine actually coughed the first time before I touched either the mixture or throttle.
With regard to your mention of cool-running, I took off when temps were still around freezing, and did not climb aggressively. On most warmer days when taking off my CHTs will push 400 degrees, particularly if I am climbing. But, I don't really have any trouble keeping CHTs in a reasonable range if I climb with rich mixture for cooling. In cruise, CHTs have typically been between 325 and 375 depending on OAT and to a lesser extent, altitude.
I'll be interested in any thoughts you have on the data.
cheers,
greg
Last edited:
At 100 seconds into the flight (point 20) it's at WOT and full rich. EGT's are ballpark 1250-1300. Fuel flow is about 15 GPH.
At 105 seconds leaning begins. By 120 seconds FF is down to 10 GPH with the throttle still at WOT; RPM is actually climbing. EGTs spike to a ballpark of 1450-1500, peak or very close (the EGT curves are flattening). At 125 seconds or just after, #2 fails.
Jean is at 2832 MSL so let's assume 4000 feet on downwind. The RPM and standard day pressure would suggest something over 80% power, maybe 85%. Leaning to peak was, mmmmm, very aggressive.
At 105 seconds leaning begins. By 120 seconds FF is down to 10 GPH with the throttle still at WOT; RPM is actually climbing. EGTs spike to a ballpark of 1450-1500, peak or very close (the EGT curves are flattening). At 125 seconds or just after, #2 fails.
Jean is at 2832 MSL so let's assume 4000 feet on downwind. The RPM and standard day pressure would suggest something over 80% power, maybe 85%. Leaning to peak was, mmmmm, very aggressive.
Last edited:
john.schwaner
Member
Stuck Valve Check?
Check for valve sticking. A stuck valve that does not open can blow the supports.
Check the push rods for any bending by rolling them on a flat surface. Push open the valves with a heavy push with your thumb. Feel for any dragging as the valve pops open and shut.
Used to see this often on the O-200 cylinders
Check for valve sticking. A stuck valve that does not open can blow the supports.
Check the push rods for any bending by rolling them on a flat surface. Push open the valves with a heavy push with your thumb. Feel for any dragging as the valve pops open and shut.
Used to see this often on the O-200 cylinders
Hmm
Hmm, diplomatically said. I think leaning to anything over 125 ROP at anything above 75% power is not prudent (Deakin's Red zone).
Certainly not a good idea but - By what mechanism could the high EGT's and assumed high cylinder pressures at such a power setting (for such a brief time) contribute to the failure seen?? Could it "stick" a valve or otherwise overload the valve train?? What are the valve guides made of?? Hard to believe that these engines are that intolerant to such momentary inattention.
Thanks Greg for sharing this data. The value of such a discussion is great and probably unprecedented outside of internal discussions of engineering staffs.
And thanks John for expert advice from a world of experience.
Hmm, diplomatically said. I think leaning to anything over 125 ROP at anything above 75% power is not prudent (Deakin's Red zone).
Certainly not a good idea but - By what mechanism could the high EGT's and assumed high cylinder pressures at such a power setting (for such a brief time) contribute to the failure seen?? Could it "stick" a valve or otherwise overload the valve train?? What are the valve guides made of?? Hard to believe that these engines are that intolerant to such momentary inattention.
Thanks Greg for sharing this data. The value of such a discussion is great and probably unprecedented outside of internal discussions of engineering staffs.
And thanks John for expert advice from a world of experience.
Last edited:
rvmills
Well Known Member
Concur, but I wanted to ask Greg if he leaned first or reduced throttle first. EGT and RPM graphs makes it look like the former. When I mentioned aggressive leaning, I was talking about on the ground. 4K on a cold day is early (low DA) to lean first, but I don't know how much he actually leaned. That peak seems to coincide with the failure, so was he leaned to peak (red zone), or did EGTs start to drop becuase of the failure and subsequent rough running. We'll discuss it today some more...headed out to see if the engine shop guy is there with the valve spring remover, as we couldn't meet up with him yesterday.
Exhaust valve push rod is bent a bit, intake valve push rod is straight (rolled them on glass). Tried the thumb press...will eat some wheaties today and push harder...and hope the engine guy is in today!
Mo later!
Cheers,
Bob
Last edited:
Mike S
Senior Curmudgeon
Welcome to VAF!!!!
John, welcome to VAF.
Nice to have another voice of knowledge in things aircraft here.
Good to have you aboard.
John, welcome to VAF.
Nice to have another voice of knowledge in things aircraft here.
Good to have you aboard.
I've been following this thread with interest from the start.
The exhaust valve head and stem would heat quickly from the high EGT. The stem will expand a bit and if the guide to stem clearance is on the tight side of the tolerance or lead deposits are present on the stem or guide- presto, stuck valve is possible.
I'd expect the pushrod to give long before the rocker boss would fracture. IMO the design thickness of these are marginal looking at the photos.
I'm curious though what the valve head to piston crown clearance is on one of these engines at full valve lift? Anyone know? To bend the pushrod, the valve had to be hit by the piston yet no mark was evident on the piston. It is unusual in my experience to see a valve stuck fully open. We can't judge how far the valve might have been open based on the degree the pushrod is bent since it may have bent only a bit before the boss failed. The pushrod seemed barely bent. Usually when catastrophic valve piston contacts occurs for whatever reason, the pushrod is seriously bent and it takes a lot of force to break a rocker or boss on most engines. Maybe we need to examine other possibilities more closely.
The exhaust valve head and stem would heat quickly from the high EGT. The stem will expand a bit and if the guide to stem clearance is on the tight side of the tolerance or lead deposits are present on the stem or guide- presto, stuck valve is possible.
I'd expect the pushrod to give long before the rocker boss would fracture. IMO the design thickness of these are marginal looking at the photos.
I'm curious though what the valve head to piston crown clearance is on one of these engines at full valve lift? Anyone know? To bend the pushrod, the valve had to be hit by the piston yet no mark was evident on the piston. It is unusual in my experience to see a valve stuck fully open. We can't judge how far the valve might have been open based on the degree the pushrod is bent since it may have bent only a bit before the boss failed. The pushrod seemed barely bent. Usually when catastrophic valve piston contacts occurs for whatever reason, the pushrod is seriously bent and it takes a lot of force to break a rocker or boss on most engines. Maybe we need to examine other possibilities more closely.
Last edited:
The piston can't hit the valve at full lift , at least in a flat piston engine, pretty sure. What happens is the valve sticks at an intermediate or closed position and the lobe comes up and bends the pushrod. Amount of pushrod bend is determined by how far open the valve is. Unless the rocker boss fails before you get to the top of the lobe. Sometimes, if the valve sticks all the way open, the pushrod can pop out of the socket on the lifter and then get bent.
