Is it possible to catch preignition or detonation prior to cylinder damage?

[Tokirbymd]

If you catch a detonation or pre-ignition event early and take immediate action to lower the temps and it goes back to normal then what? Had what I think is a detonation event on climb out today. Noticed CHT climbing rapidly and pulled the power back to idle immediately as soon as it crossed 400. It continued to rise briefly over 550 then started back down- was only above redline for maybe 10sec? Once everything cooled off was able to bring power back up to low power 20" for descent and landing and everything behaved completely normal after that. Pulled the plug and borescoped the cylinder- no obvious damage to plug or cylinder or piston top. I plan on replacing the plugs just in case. However after speaking to my AP, not sure what else to do other than go fly and watch it carefully. I think I caught it very early and prevented any damage. The only other thing I can think of would be to pull the cylinder and inspect but that seems aggressive if no damage is seen on borescope and goes back to operating normally. Oh, to compound the issue, is the same cylinder that I have had erroneous indications on a couple of flights ago, but this one wasn't behaving like a bad probe, was more of a continuous rise and fall that looked real.

 

[Eric Pauley]

This would probably be a perfect time to pay up for SavvyAnalysis (if you haven't already) to get feedback from someone that looks at these all day.

That being said, the CHT dropoff you're seeing looks pretty non-physical to my eyes. 200F drop in the span of 10s? Here's a column with an example trace of a detonation event: https://resources.savvyaviation.com...es_eaa/EAA_2010-09_destructive-detonation.pdf

As rapid as it is, we're only talking about 100F/min of temperature change. Given the high thermal mass of the cylinger the rate you're seeing doesn't seem possible. You're also not seeing any difference in EGT on that cylinder. My money's on indication error.

 

[Tokirbymd]

This would probably be a perfect time to pay up for SavvyAnalysis (if you haven't already) to get feedback from someone that looks at these all day.

That being said, the CHT dropoff you're seeing looks pretty non-physical to my eyes. 200F drop in the span of 10s? Here's a column with an example trace of a detonation event: https://resources.savvyaviation.com...es_eaa/EAA_2010-09_destructive-detonation.pdf

As rapid as it is, we're only talking about 100F/min of temperature change. Given the high thermal mass of the cylinger the rate you're seeing doesn't seem possible. You're also not seeing any difference in EGT on that cylinder. My money's on indication error.

I hope so but these are 10sec interval readings, real time it seemed a lot more continuous than probe errors I've seen in the past. Also, the precipitous drop in CHT coincided with pulling the power to idle and descending at a rapid rate (145kias) to get back to home base before something bad happened. I've read all of the Savvy columns and have an inquiry in to them as well as a subscriber to the analysis plan. My gut feeling is either preignition from a too widely gapped plug that caused magneto cross fire, vs pure detonation (from what I'm not sure as I was full rich and EGT don't support leaning from injector plugging). Hopefully spending only 10sec or so above redline didn't cause any damage. Problem is, I'm not sure where to go from here if non-destructive analysis doesn't yield any problems- if plugs, cylinder walls, head, and piston all look good, and compressions are ok (ie no broken rings) then what? I guess there still could be stress damage to either a rod or bearing that won't show up until it does?

 

[Eric Pauley]

I hope so but these are 10sec interval readings, real time it seemed a lot more continuous than probe errors I've seen in the past. Also, the precipitous drop in CHT coincided with pulling the power to idle and descending at a rapid rate (145kias) to get back to home base before something bad happened. I've read all of the Savvy columns and have an inquiry in to them as well as a subscriber to the analysis plan. My gut feeling is either preignition from a too widely gapped plug that caused magneto cross fire, vs pure detonation (from what I'm not sure as I was full rich and EGT don't support leaning from injector plugging). Hopefully spending only 10sec or so above redline didn't cause any damage. Problem is, I'm not sure where to go from here if non-destructive analysis doesn't yield any problems- if plugs, cylinder walls, head, and piston all look good, and compressions are ok (ie no broken rings) then what? I guess there still could be stress damage to either a rod or bearing that won't show up until it does?

Definitely curious to hear what they say. I'm no A&P, but I do routinely read the Savvy engine monitor puzzlers.

The two rapid cooling events are the strange part for me. Sure, preignition could cause faster temp spikes than detonation, but the cooling rate should be similar and that rate of cooling (200F from 1 sample to the next, so 10s) just does not seem plausible. With either detonation or preignition you would expect some change in EGT (Again I'm no expert, but I'd predict you'll hear something similar from Savvy).

 

[Fearless Tower]

What engine? Continental or Lycoming?

If a Conti and it really did get that high, I would expect future problems with that cylinder.

 

[455 Bravo Uniform]

Not sure why you’d be worried about and focusing on detonation. That’s the second time in a week someone has mentioned detonation. Is it trending somewhere on the web?

Ultimately, detonation will hammer the top rod bearing and cause it to grab the crank journal. All you were looking for when looking for broken plug ceramic or tiny melted aluminum balls is evidence that it occurred. You’re not gonna do a tear down just because of that.

Your exhaust valve looked ok?

Your fix is fueling or air intake or head heat dissipation (baffling or poor exhaust valve seating).

Swap temp probes with another cylinder.

It’s what happens to aluminum at 550 deg F that you ought to look into. I don’t know aviation engines all that well, but that’s what I understand.

 

[Tokirbymd]

Yeah am concerned about that. Less so for the cylinder and more so if any bearings or rods were damaged. Not sure how you would identify that though shirt if an IRAN

 

[Hang 4]

Another explanation could be a sensor issue. Notice the "noisy" area after the spike, then an abrupt return to normal. CHT at nearly 600f would likely cause visible damage.

 

[Kenny Phillips]

If you catch a detonation or pre-ignition event early and take immediate action to lower the temps and it goes back to normal then what? Had what I think is a detonation event on climb out today. Noticed CHT climbing rapidly and pulled the power back to idle immediately as soon as it crossed 400. It continued to rise briefly over 550 then started back down- was only above redline for maybe 10sec? Once everything cooled off was able to bring power back up to low power 20" for descent and landing and everything behaved completely normal after that. Pulled the plug and borescoped the cylinder- no obvious damage to plug or cylinder or piston top. I plan on replacing the plugs just in case. However after speaking to my AP, not sure what else to do other than go fly and watch it carefully. I think I caught it very early and prevented any damage. The only other thing I can think of would be to pull the cylinder and inspect but that seems aggressive if no damage is seen on borescope and goes back to operating normally. Oh, to compound the issue, is the same cylinder that I have had erroneous indications on a couple of flights ago, but this one wasn't behaving like a bad probe, was more of a continuous rise and fall that looked real.View attachment 117288

Kind of odd to yank the power at 400° yet have then climb another 150°, I'd still swap probes to make sure everything is OK.

 

[MooneyDriver78]

Did you get a warning or did you just notice the higher temperature?

What temperature do you have the warning set to, I think the default is too high (~500°) IIRC?

 

[vcollazo]

Mike Bush had a discussion of this a few years back. He mentioned a case Where a Cirrus pilot that had this exact situation but no engine monitor wound up with an engine failure and overhaul. Another Cirrus but with an engine monitor reacted just like you and had no damage and a normal landing. I"m not 100% sure, but as I remember the cause was a hot spot on a damaged spark plug causing pre ignition on the latter one.

 

[brcase]

My opinion...

Detonation should be pretty easily detectable and should have time to fix.

Pre-ignition, is likely a mechanical or fuel quality issue, and you are not likely to catch or fix it before having serious issues.
I seem to remember a Savy Aviator article showing the analysis of the Bonanza that experienced it with a minute after takeoff, likely due to a defective Sparkplug Heli coil, acting like a glow plug as it protruded into the combustion cylinder. I seem to recall some serious engine damage, don't recall if it was a Cracked cylinder or burned piston.

Since this appears to have only occurred on one cylinder I think an in depth analysis and troubleshooting is in order. I am thinking it might be a sensor issue, since I also don't see a corresponding EGT temp. But need to look at the data in more detail.

https://www.avweb.com/features/pelicans-perch-43detonation-myths/

Brian
CFIIG/ASEL

 

[Tokirbymd]

Did you get a warning or did you just notice the higher temperature?

What temperature do you have the warning set to, I think the default is too high (~500°) IIRC?

Saw it go from usual 330 to 350 to 400 and knew something was up so I yanked the power immediately. Alarm is set for 400 but I watch them like a hawk as my usual scan

 

[Dan Thomas]

Plugged injector. Leaking intake elbow or coupling, leaning that cylinder. Possibly a failed plug or lead, slowing the burn and allowing time for chemical breakdown and detonation.

 

[Tokirbymd]

Plugged injector. Leaking intake elbow or coupling, leaning that cylinder. Possibly a failed plug or lead, slowing the burn and allowing time for chemical breakdown and detonation.

but no rise in EGT that would suggest lean condition prior to CHT rise.

 

[Eric Pauley]

What kind of CHT probe is it, out of curiosity? FWIW I am sticking with my guess that it's indication error. I look forward to seeing this in next month's puzzler!

 

[Tokirbymd]

What kind of CHT probe is it, out of curiosity? FWIW I am sticking with my guess that it's indication error.

smaller gasket probe under a tanis heater bayonet. I hope its sensor error, but I've had those before and the indication is more erratic. This was smooth up and back down, although as mentioned above, is slightly faster down that I would expect.

 

[Eric Pauley]

My main evidence is how non-physical the temperature drops look. That being said I suppose in theory the probe could have been reading out of spec from the overheating event.

 

[mondtster]

I hope its sensor error, but I've had those before and the indication is more erratic. This was smooth up and back down, although as mentioned above, is slightly faster down that I would expect.

I don't see anything smooth about that data. Of course, at a 0.1 hertz acquisition rate there is an eternity between points.

If you have an open circuit what does the engine monitor read? What does it read right after the circuit is opened? What does it read right after it is closed? This looks like a failing thermocouple or a bad connection to me.

 

[Clip4]

You may have escaped any damage, but I would not fly the plane again until you have accurately determined the cause and corrected it.

 

[Tokirbymd]

You may have escaped any damage, but I would not fly the plane again until you have accurately determined the cause and corrected it.

Current theory is preignition maybe from bad plug or too wide plug gap causing mag cross fire. Changing plugs and doing compression to confirm as well as changing probe. Just finished the Advanced Pilot Seminars course online and they demonstrate an almost identical event. Per them and per Savvy, if no identifiable damage or evidence of detonation on Borescope treat it as a one off event and continue flying, but continue to monitor for repeat events.

 

[Eric Pauley]

Current theory is preignition maybe from bad plug or too wide plug gap causing mag cross fire. Changing plugs and doing compression to confirm as well as changing probe. Just finished the Advanced Pilot Seminars course online and they demonstrate an almost identical event. Per them and per Savvy, if no identifiable damage or evidence of detonation on Borescope treat it as a one off event and continue flying, but continue to monitor for repeat events.

Did your Savvy rep have any thoughts on the 200F temp drop in 10 seconds? That’s far beyond anything I’ve seen in any engine monitor plot without a sensor issue.

 

[Tokirbymd]

Did your Savvy rep have any thoughts on the 200F temp drop in 10 seconds? That’s far beyond anything I’ve seen in any engine monitor plot without a sensor issue.

They felt like it was real, although faulty probe coho not be ruled out. Remember I chopped throttle and started high speed descent when this happened as I was close I home base but high. But not to discount bad probe theory I’m replacing it also.

 

[Eric Pauley]

I think it is basically impossible for that cooling to occur in a working probe, especially that second drop. As you noted the engine was at idle at high speed in the initial drop. Compare the temp drops on the other cylinders there (absolute best cooling scenario) with the drop seen later in the bad cylinder. That drop is several times faster than best case cooling at those temps. Of course bad probe doesn’t preclude other issues but I don’t see how that trace could actually physically occur.

 

[Tokirbymd]

I think it is basically impossible for that cooling to occur in a working probe, especially that second drop. As you noted the engine was at idle at high speed in the initial drop. Compare the temp drops on the other cylinders there (absolute best cooling scenario) with the drop seen later in the bad cylinder. That drop is several times faster than best case cooling at those temps. Of course bad probe doesn’t preclude other issues but I don’t see how that trace could actually physically occur.

Hmm that’s actually a pretty good observation. The rate of drop does go back to match the other curves toward the end of the drop. Makes me feel a bit better but will still proceed with looking over everything carefully.

 

[Magman]

OP says he closed throttle upon reaching 400 and temp continued to over 500.

Very unlikely to get detonation with a closed throttle.

Borescope , comp check and replace plugs was a good call.

I’m presuming injected and would clean injector and pressure test induction

system as well.( Dan #14)

 

[Eric Pauley]

Clearly I'm very bored, because I went ahead and calculated the physics of this. This is all embarassingly approximate, of course.

A cylinder is around 35lb, give or take. Given the specific heat of aluminum, the 150F temp rise during that one 10s window, and the energy density of gasoline (similar-ish to avgas), we can compute the minimum theoretical fuel flow to achieve this temp rise. This is assuming 100% thermal efficiency in heating up the cylinder.

Calculation: https://www.wolframalpha.com/input?i=35lb+*+specific+heat+of+aluminum+*+(150+delta+degrees+F/10s)+/+(energy+density+of+gasoline)+/(6lbs/gallon)*6

We get 20gal/hr. So at the fuel flow you had I don't think it is physically possible to heat the cylinder that quickly regardless of what the combustion event is doing. This is even assuming zero aerodynamic cooling. Of course the cylinder also heats irregularly so YMMV.

 

[Magman]

While CHT will increase during detonation my understanding has always been it’s

the localized temps & pressures that that will cause the damage.

“ Volcanic” is the description of the explosion in an ancient vid.

Again; I don’t think it’s an issue in this case.

 

[Dan Thomas]

A cylinder is around 35lb, give or take.

Must one huge engine. A fixed-pitch metal prop is around 25 pounds, and I've never picked up any cylinder that heavy.

 

[Eric Pauley]

Must one huge engine. A fixed-pitch metal prop is around 25 pounds, and I've never picked up any cylinder that heavy.

That's with piston, accessories, packaging per Spruce, so more take than give. In any case, the rate of temperature rise is butting up against the physical energy limits of the avgas.

 

[Tokirbymd]

Clearly I'm very bored, because I went ahead and calculated the physics of this. This is all embarassingly approximate, of course.

A cylinder is around 35lb, give or take. Given the specific heat of aluminum, the 150F temp rise during that one 10s window, and the energy density of gasoline (similar-ish to avgas), we can compute the minimum theoretical fuel flow to achieve this temp rise. This is assuming 100% thermal efficiency in heating up the cylinder.

Calculation: https://www.wolframalpha.com/input?i=35lb+*+specific+heat+of+aluminum+*+(150+delta+degrees+F/10s)+/+(energy+density+of+gasoline)+/(6lbs/gallon)*6

We get 20gal/hr. So at the fuel flow you had I don't think it is physically possible to heat the cylinder that quickly regardless of what the combustion event is doing. This is even assuming zero aerodynamic cooling. Of course the cylinder also heats irregularly so YMMV.

Not sure the calculations are that simple. The CHT reflected in the probe is the temperature of the probe which is butted against a well that is drilled to some fraction of an inch from the inner cylinder wall (I looked but can't find the actual wall thickness right now). So there is heat transfer from the cylinder wall to the probe from direct contact. Now the cylinder wall itself will have a differential temperature gradient from the inner cylinder wall to the depth of the probe, and to the fins. As the inner cylinder temps will vary from 2-3000deg to several hundred degrees during each combustion cycle. There is an insulating barrier so to speak of gas along the inner cylinder wall that keeps the cylinder somewhat cooler than the 2000 deg inner chamber. As the pressure increases (especially during detonation or preignition) this barrier is compressed, allowing the cylinder wall to heat up much more quickly and to a higher temp than normal.

But to your point, after reviewing the data again, borescoping the engine and looking at the plugs, I don't believe the rate of change in temperature was real- either heating or cooling. Have changed both plugs, changing the CHT probe tomorrow. Will go test fly and report back.

 

[Eric Pauley]

Not sure the calculations are that simple. The CHT reflected in the probe is the temperature of the probe which is butted against a well that is drilled to some fraction of an inch from the inner cylinder wall (I looked but can't find the actual wall thickness right now). So there is heat transfer from the cylinder wall to the probe from direct contact. Now the cylinder wall itself will have a differential temperature gradient from the inner cylinder wall to the depth of the probe, and to the fins. As the inner cylinder temps will vary from 2-3000deg to several hundred degrees during each combustion cycle. There is an insulating barrier so to speak of gas along the inner cylinder wall that keeps the cylinder somewhat cooler than the 2000 deg inner chamber. As the pressure increases (especially during detonation or preignition) this barrier is compressed, allowing the cylinder wall to heat up much more quickly and to a higher temp than normal.

But to your point, after reviewing the data again, borescoping the engine and looking at the plugs, I don't believe the rate of change in temperature was real- either heating or cooling. Have changed both plugs, changing the CHT probe tomorrow. Will go test fly and report back.

The irregular heating is definitely a big caveat. Glad to hear the cylinder lives to fly another day!