What's new
Van's Air Force

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

Engine stumble

crabandy

Well Known Member
O-360 with dual EFII ignition with autowires/plugs, rebuilt 2009, first engine start Sept 2014 and 96 hours without any rough running etc last year. 10 hours since condition inspection, during the condition inspection I sent the Carb to Marvel Schebler and they enriched it slightly and replaced the bowl. First 6 hours after the condition everything ran as it should. I did get fuel at the airport before the issue, but my tanks sumped clean in the pre-flight and I had them double check their truck this morning.
I flew about 2.5 hours yesterday morning without any issues, 3rd leg of the day I slightly leaned in the climb about 120-150 ROP and leveled off at 13.5. I accelerated to cruise speed, CHT's were all 380 or less I left full throttle in and made a big pull to LOP (I only do LOP below 65% power). About minutes after setting 20-80 LOP the engine ran rough like it was ingesting water or had bad carb ice (too instantaneous for carb ice). It felt like a LONG time, data shows just over a minute of rough running. I swapped fuel tanks, turned on the boost pump with no change. When I went full rich with the mixture it bogged the engine more and I leaned it about where I thought it should be it eventually smoothed out.
I pulled the SD cards and looked at the data this morning, about 250 RPM and 10 knts IAS loss in the minute. Fuel flow went from 7.8 to 9.6. EGTS were all 1310-1330 before and dropped to 895-1070 with #2 being the lowest. It was long enough that CHT's even dropped around 30 degrees.

After it smoothed out I tried ROP and LOP and both ran fine, after running LOP for about 30 minutes I started descending. I left the throttle full in and mixture at 20-80 degrees LOP. At about 13.2 the engine ran rough again for about a minute with the same drop in EGT's/CHT's/RPM/Airspeed. Small mixture adjustments as well as large mixture adjustments didn't seem to make a difference. After it smoothed out it ran flawlessly through several
level offs and different mixture/power settings for the last 40 minutes of the flight.

After startup on the ramp for my last leg I had the same stumbling as I advanced the throttle through 1200 RPM to taxi up-hill. It wasn't very noticable, I felt it more than heard it. I was able to hear it after removing my headset. Switching ignitions behaved as I would expect with a slight loss of RPM but still rough running. I aggressively leaned it out and it smoothed out after about 45 seconds. I did an extended runup and everything ran fine, after a cautious takeoff the hour flight back home was uneventful.
The data on the ground showed the #2 EGT during the rough running was 500 degrees while the other 3 cylinders were 1250's.

Looking back over the data for the previous flights of the morning when everything ran normal. During a full power, Full rich climb #2 is my richest cylinder with EGT's 180-200 degrees cooler than the other cylinders. When LOP in cruise it is within 10-15 degrees of the other cylinders.

I'm looking for things to check when I pull the cowl, ideas?
Thanks,
Andy
 
...When I went full rich with the mixture it bogged the engine more....pulled the SD cards and looked at the data this morning, about 250 RPM and 10 knts IAS loss in the minute. Fuel flow went from 7.8 to 9.6. EGTS were all 1310-1330 before and dropped to 895-1070 with #2 being the lowest. It was long enough that CHT's even dropped around 30 degrees....At about 13.2 the engine ran rough again for about a minute with the same drop in EGT's/CHT's/RPM/Airspeed. Small mixture adjustments as well as large mixture adjustments didn't seem to make a difference. .

Carb problem....something that made it go very rich. Stuck float valve, loose bowl, loose jet...? Perhaps an airbox obstruction.
 
Carb problem....something that made it go very rich. Stuck float valve, loose bowl, loose jet...? Perhaps an airbox obstruction.

I did just re-make all of the hard fuel lines, 2 of them after the fuel filter. Perhaps a chunk of something made it to the Carb, I'll probably start there.
 
The data on the ground showed the #2 EGT during the rough running was 500 degrees while the other 3 cylinders were 1250's.

I agree with Mike, this is a dead give away, it's telling you #2 cylinder quit running during the event, I don't think the carb will cause one cylinder to drop off. a sticking exhaust valve is a single point of failure where by changing the mixture, switching tanks or checking both mags/EIs will not clear up the rough running.
 
Could it also be related to ignition or spark plug on that one Cyl as another option?
 
So just got done doing some more trouble shooting. I fired the airplane up and it was rough running, it wasn't obvious on the G3X because I had my engine monitor on the checklist page. With the checklist displayed the engine parameters are truncated and only show hottest EGT/CHT. I switched to the main engine page and it was obvious the #2 was not firing correctly with a 500* EGT, exactly as seen on the engine data on the ramp the other day. I aggressively leaned it and #2 started running normal, again about a 1 minute event.
I did another extended run-up and cautious takeoff and short flight but couldn't reproduce the problem ROP/LOP. Landed, fueled up and the re-start/taxi back to the hangar was all normal.
Pulled the cowl and started an early oil change. Carb shows no signs of leakage on the outside or in the throat. I did speak to the Mark at Marvel Schebler (great resource) and besides some sort of leakage/debris carbs generally don't "fix themselves."
At this point I'm fairly certain it is an ignition/spark plug problem. I pulled the plugs on the #2 cylinder and I (not an expert) think they look fine, granted they were firing just fine 30 minutes ago. They are NGK BR8EIX 5044 auto plugs with adapters smeared with anti-seize.
7C0F00C5-8EFE-4E78-A7EA-0BC660ED1125_zpsttjy33gf.jpg


I measured the spark plug resistance at 3800-3900 on both warm. Plug wires measures 800-1250 ohms with the longer wire measuring 1250.

Before I just put new plugs in it and fly it how do I rule out a sticky valve?
 
Just an FYI/observance here, I believe you should install the plug adapters, and leave them in the head, remove only the plugs.

What are the other plugs in the engine looking like, and testing for resistance?

The fact the problem comes and goes would support a sticking valve issue...................
 
Mike S,
I put the socket on the sparkplug and both come out, it's consistently been that way when I remove them. I finger tighten the adapter on the sparkplug and then screw/tighten that assembly to 30 ft/lbs via the sparkplug. All the plugs looked similar when I removed them, but I didn't put the ohm meter on them.
 
Last edited:
I do recall a small burble while switching ignitions during a runup on several occasions within the past 15 hours. With the carb jetted slightly richer now perhaps it's a plug/fouling issue?
 
...About minutes after setting 20-80 LOP the engine ran rough like it was ingesting water or had bad carb ice (too instantaneous for carb ice). It felt like a LONG time, data shows just over a minute of rough running. I swapped fuel tanks, turned on the boost pump with no change. When I went full rich with the mixture it bogged the engine more and I leaned it about where I thought it should be it eventually smoothed out.

After it smoothed out I tried ROP and LOP and both ran fine, after running LOP for about 30 minutes I started descending. I left the throttle full in and mixture at 20-80 degrees LOP. At about 13.2 the engine ran rough again for about a minute with the same drop in EGT's/CHT's/RPM/Airspeed. Small mixture adjustments as well as large mixture adjustments didn't seem to make a difference. After it smoothed out it ran flawlessly through several
level offs and different mixture/power settings for the last 40 minutes of the flight.

...After startup on the ramp for my last leg I had the same stumbling...
...after a cautious takeoff the hour flight back home was uneventful....

...I switched to the main engine page and it was obvious the #2 was not firing correctly with a 500* EGT, exactly as seen on the engine data on the ramp the other day. I aggressively leaned it and #2 started running normal, again about a 1 minute event.

I did another extended run-up and cautious takeoff and short flight but couldn't reproduce the problem ROP/LOP. Landed, fueled up and the re-start/taxi back to the hangar was all normal.

I'm sorry, but I just don't understand this. Why do people insist on continuing to fly their planes when experiencing engine issues? You clearly have an intermittent engine problem that you didn't previously have. Why do you continue to fly it? Find the problem on the ground and get fixed now BEFORE it gets worse in flight or worse, on takeoff!!!

Skylor
RV-8
 
Last edited:
Before I just put new plugs in it and fly it how do I rule out a sticky valve?

Get your "trusted" A&P or buddy (maybe someone from your EAA chapter) who know how and has the tools come and help you do a valve wobble test, It's Imperative you do this before you fly it, my money is on the exhaust valve on #2, for it to be ignition you would need to have two failures for #2 to stop running, if one plug or one wire failed your EGT would rise and #2 would keep running, for the EGT to fall and #2 to stop running you would need both plugs to fail or a plug and a wire or a combination of both.
 
NGK 5044

The NGK 5044 has a removable terminal nut that could come loose with vibration. It is recommended to use the "solid" style, part number 6747 if I'm not mistaken.

Bevan
 
Mike S,
I put the socket on the sparkplug and both come out, it's consistently been that way when I remove them. I finger tighten the adapter on the sparkplug and then screw/tighten that assembly to 30 ft/lbs via the sparkplug. All the plugs looked similar when I removed them, but I didn't put the ohm meter on them.

I use this method from the Lightspeed manual and the adapters remain in the heads?
Install adaptors in cylinder head using the supplied copper washer.
Torque to 35 - 45 ft-lbs using anti-seize compound.
? Install automotive style spark plugs with their washer. Torque to 20 ft-lbs using
anti-seize compound.

Chris
 
Mike S,
I put the socket on the sparkplug and both come out, it's consistently been that way when I remove them. I finger tighten the adapter on the sparkplug and then screw/tighten that assembly to 30 ft/lbs via the sparkplug. All the plugs looked similar when I removed them, but I didn't put the ohm meter on them.

I would recheck your procedure on the plugs and inserts. I don't have access to the books at the moment but I believe lightspeed calls for the inserts to be torqued to 24 ftlbs and the plugs to 20. My inserts always stay in the heads.

George
 
Torque for adapters

Check your installation manual. I think the adapters are set to a higher torque than the plugs so they remain in the engine when plugs are removed as was mentioned before.
 
Get your "trusted" A&P or buddy (maybe someone from your EAA chapter) who know how and has the tools come and help you do a valve wobble test, It's Imperative you do this before you fly it, my money is on the exhaust valve on #2, for it to be ignition you would need to have two failures for #2 to stop running, if one plug or one wire failed your EGT would rise and #2 would keep running, for the EGT to fall and #2 to stop running you would need both plugs to fail or a plug and a wire or a combination of both.

Russ, Thanks for the advice. I'm going to schedule my trusted A&P buddy today.
 
Adapters

Be carefull! All adapters aren't created equal! I had one break off while torquing to 30 foot pounds without the plug being inserted.
 
Mike S,
I put the socket on the sparkplug and both come out, it's consistently been that way when I remove them. I finger tighten the adapter on the sparkplug and then screw/tighten that assembly to 30 ft/lbs via the sparkplug. All the plugs looked similar when I removed them, but I didn't put the ohm meter on them.
Keep doing it this way. What is described below happened to me.
Be carefull! All adapters aren't created equal! I had one break off while torquing to 30 foot pounds without the plug being inserted.
 
Last edited:
I understand the focus on cylinder #2 given a 500 degree EGT during a runup, but returning to your original post, did you or did you not record a drop in all EGT and CHT indications, along with an increase in fuel flow?

I pulled the SD cards and looked at the data this morning, about 250 RPM and 10 knts IAS loss in the minute. Fuel flow went from 7.8 to 9.6. EGTS were all 1310-1330 before and dropped to 895-1070 with #2 being the lowest. It was long enough that CHT's even dropped around 30 degrees.
 
I understand the focus on cylinder #2 given a 500 degree EGT during a runup, but returning to your original post, did you or did you not record a drop in all EGT and CHT indications, along with an increase in fuel flow?

I suspect that may all have happened because of the first sentence quoted below.

"When I went full rich with the mixture it bogged the engine more and I leaned it about where I thought it should be it eventually smoothed out.
I pulled the SD cards and looked at the data this morning, about 250 RPM and 10 knts IAS loss in the minute. Fuel flow went from 7.8 to 9.6. EGTS were all 1310-1330 before and dropped to 895-1070 with #2 being the lowest. It was long enough that CHT's even dropped around 30 degrees".
 
Dan and Russ,
The fuel flow from the data cards probably went up because of me enriching the mixture while LOP when I encountered the roughness. My fuel flow tends to bounce around a lot below 10 GPH anyway, the graph for my fuel flow looks like a EKG when zoomed in for a minute. Any interpolation of fuel flow from the data cards is a rough guess. I'm guessing at the lower fuel flows the bowl temporarily fills up and the needle shuts of fuel while higher fuel flows never allows the bowl to fill and keeps a steady stream of fuel coming into the carb.

I'll have to review the data at idle to check the fuel flows.

Is it possible that while flying at full power LOP, #2 exhaust valve sticks open and is unable to suck it's fair share of the air fuel mixture causing the other 3 cylinders to get an overly rich mixture backed up by the lower EGT's?
 
Dan and Russ,
The fuel flow from the data cards probably went up because of me enriching the mixture while LOP when I encountered the roughness. My fuel flow tends to bounce around a lot below 10 GPH anyway, the graph for my fuel flow looks like a EKG when zoomed in for a minute. Any interpolation of fuel flow from the data cards is a rough guess. I'm guessing at the lower fuel flows the bowl temporarily fills up and the needle shuts of fuel while higher fuel flows never allows the bowl to fill and keeps a steady stream of fuel coming into the carb.

I'll have to review the data at idle to check the fuel flows.

Is it possible that while flying at full power LOP, #2 exhaust valve sticks open and is unable to suck it's fair share of the air fuel mixture causing the other 3 cylinders to get an overly rich mixture backed up by the lower EGT's?

I know it is a pain, Andy, but if you download the data file and then do a running average of the data it will smooth out the indications and give a more true reading for the data. It can also be aliased due to how the data is acquired from the red cube. If there is a small averaging window, then the data recorded can be catching the flow at different points as it varies. Maybe the G3X guys can tell us more about that, but averaging will work for now to determine the best idea of fuel flow change.

For grins, what would happen if a valve stuck? First, It would not pump as much air. So, then the A/F is determined by the carb, right? If the airflow goes down, then the fuel usage should drop too. The A/F should be relatively the same for the remaining cylinders. There is some non-linearity of A/F with less CFM (yes, not mass) by the carb, but a 20-25% drop in air flow would definitely drop fuel flow.

So- per DanH's question, the fuel flow increase is very important in the diagnostic journey.

Also, you may have multiple things at work here, but lets kick this item first and get confident in it's behavior.
 
I'm sorry, but I just don't understand this. Why do people insist on continuing to fly their planes when experiencing engine issues? You clearly have an intermittent engine problem that you didn't previously have. Why do you continue to fly it? Find the problem on the ground and get fixed now BEFORE it gets worse in flight or worse, on takeoff!!!

Skylor
RV-8

Skylor,
You have a very valid point.

It reminds me of the time I left a C150 100 miles from home and had my wife (6 month old in tow) pick me up at 11:00 pm. The airplane ran up perfectly but was rough several seconds after going to full power. Mechanics checked it out, couldn't reproduce the problem poured in marvel mystery oil called it a stuck valve now unstuck and flew it home. Airplane flew 10-15 hours over the next couple of weeks without an issue.
I was checking out a student in the same aircraft, we did about an hour of maneuvers and several touch and go's. On one of the touch and go's (student landed a little long on a 2900' strip) about 75 feet AGL the engine started stumbling. Pulling power immediately would have resulted in flaring at the departure end of the runway and hitting the steep ditch and highway 100-150 past the departure end of the runway.
I decided to keep flying (barely) and I was going to pull the power once I cleared the road and land straight ahead or within 30 degrees. My takeoff briefing to myself since learning to fly was an engine failure on takeoff below 1000 feet is straight ahead or within 30 degrees, fly it all the way into the crash.
When I cleared the highway straight ahead was trees, several small creeks and a marshy area. 30 degrees right and a few hundred more feet cleared the trees with a short open area I could touch down slow down as much as possible before hitting a hedge row at 20 knts.
Each time it came to pull the power and do the forced landing a better option presented itself. I went through 5 spots where I was sure I was going to put it down before making it back to a perpendicular runway.
Engine ran-up fine even at full power once on the ground, my student looked at me wild eyed and said "I'm done for the day!" The second time around the mechanics found a loose baffle in the exhaust of 1 side that was blocking the outlet.

Most of us have flown rentals etc that have stumbles/fouled plugs etc that we write and send off to maintenance only to come back as "checked out fine." I'm still learning the whole aircraft ownership I get to do my own maintenance thing, as the situation happened I reacted as best I knew how.
I could've diverted several hundred miles from home at the first available airport at the first stumbling incident. In my opinion myself and a hired A&P would have spent several hours checking out the airplane without reproducing the problem.

Why did I continue to fly? Probably because of the "intermittent" nature of the problem along with the experiences of mechanics not being able to track down and reproduce intermittent problems.
 
I'm sorry, but I just don't understand this. Why do people insist on continuing to fly their planes when experiencing engine issues? You clearly have an intermittent engine problem that you didn't previously have. Why do you continue to fly it? Find the problem on the ground and get fixed now BEFORE it gets worse in flight or worse, on takeoff!!!

Skylor
RV-8

Skylor,
You have a very valid point.


Why did I continue to fly? Probably because of the "intermittent" nature of the problem along with the experiences of mechanics not being able to track down and reproduce intermittent problems.

Perhaps I am hitting a bit below the belt here, but isnt this a good enough reason to fix things before you fly it again???

F4E732E4-3AD2-406A-852F-71722F721E72_zpsk2qd4gpt.jpg
 
Last edited:
The fuel flow from the data cards probably went up because of me enriching the mixture while LOP when I encountered the roughness.

Your first post stated "Small mixture adjustments as well as large mixture adjustments didn't seem to make a difference"...which would lead an observer to conclude the fuel flow increase was not based on pilot input.

..the fuel flow increase is very important in the diagnostic journey.

As is the rest of the question. Did your in-flight EGT and CHT indications go up and down together, or did #2 exhibit significant variance?

Your original post was " EGTS were all 1310-1330 before and dropped to 895-1070 with #2 being the lowest. It was long enough that CHT's even dropped around 30 degrees.

A 175 degree EGT spread (1070-895) is not unusual with a carbed engine, and CHT on a dead cylinder would drop more than 30 degrees in a minute. Is the information accurate?
 
I'm having trouble posting the graphs from cirrusreports, I thought about trying the Saavyanalysis. Any recommendations on websites to download my data to?

DanH, I sent a pm wi a link to my cirrusreports.
BillL, I sent a link via email to my cirrusreports.
 
First try from SaavyAnalysis of my first engine stumble at 13.5 shortly after leveling off and setting the mixture LOP. It didn't save my chart settings, You'll have to reset the graphs and look at the EGT dip at 02:00 and just before 03:00 reference Fuel Flow, RPM and CHT.

At 2:02 a FF spike of 12.2 gph from 7.4 ish ghp, the EGT's dip quite severely.
At 2:56 a FF spike of 9.6 gph from 8 ish gph, the corresponding EGT dip was not as severe as the first event.

https://www.savvyanalysis.com/flight/894237/58649f6a-bdb8-4616-b6cc-c28a4c1a7ba8
 
Last edited:
Do a compression check on #2. That looks like a sticky valve to me. If it tests ok, rotate through a few cycles by hand, and do it again. If it doesn't show then, follow Russ's advise on the wobble test.
 
Last edited:
Harsh Words

Skylor,
You have a very valid point.

Thank you for not taking my comment the wrong way. I just get very concerned every time I read about someone continuing to fly an airplane with an engine that is obviously ill behaved. The NTSB reports are littered with stories of such behavior...

Skylor
 
Overlay fuel flow on the EGT plot and MAP on the CHT plot. Fuel flow is fluctuating a lot. When I look at the data from my flights I see small fluctuations in fuel flow but not swings like that.

My engine is fuel injected so the fuel pressure should be different, but I don't think your carburated engine should vary that much compared to a FI engine.
 
First try from SaavyAnalysis of my first engine stumble at 13.5 shortly after leveling off and setting the mixture LOP. It didn't save my chart settings, You'll have to reset the graphs and look at the EGT dip at 02:00 and just before 03:00 reference Fuel Flow, RPM and CHT.

At 2:02 a FF spike of 12.2 gph from 7.4 ish ghp, the EGT's dip quite severely.
At 2:56 a FF spike of 9.6 gph from 8 ish gph, the corresponding EGT dip was not as severe as the first event.

https://www.savvyanalysis.com/flight/894237/58649f6a-bdb8-4616-b6cc-c28a4c1a7ba8

Yeah buddy...plotted engine data beats a description every time. Learning how to analyze downloaded EMS data is a skill requiring practice, so I'm going to take a shot. Ya'll check me.

First, only one cylinder was LOP, #4. The rest never peaked. The key is to look for the characteristic hump in the EGT trace for each cylinder, a rise to peak and a descent into either LOP or ROP territory, which is determined from fuel flow.

Here's a snip taken from where Andy settled into cruise following the first power incident. Fuel Flow is blue, green is RPM. If you open the file (link above), expand this area, and work with the cursor to see temperatures, you'll see that #4 peaks and then goes about 30 LOP, while the others never peak. For the next 50 minutes or so they are running in the least desirable ROP zone (between peak and 75 ROP), but there is no way to tell exactly where. RPM shows they were leaned through best power mixture (150~100 ROP), then a bit beyond.



There's a confirmation at the end of the cruise period. Andy pushes over into a descent without changing power settings; TAS rises and altitude falls, meaning MAP rises. As it does, the EGTs rise. If they were LOP, they would move leaner; EGT would fall.



None of this has any bearing on the power interruption. I point it out for several reasons. First, I've noticed some EMS presentations seem to be capable of misleading pilots, who see "LOP" or "ROP" spelled out, and don't track the actual numbers. The system may or may not be calibrated enough to be provide an accurate indication. Second, look at the nice tight temperature grouping during those 50 minutes of cruise. Although it may seem wonderful that all those temperatures are so closely grouped, it does not mean they are all on the same side of peak. Third, it's an exercise in plot analysis.

On to the good stuff; what caused the inflight power interruptions at 02:02:02 and 02:55:51? I'm just going to state my conclusion up front, and then tell you why. It stuck an intake valve.

Let's start by looking at the plot taken at idle. Smoking gun here is the increased manifold pressure when #2 EGT says the cylinder has gone dead. The MAP increase indicates a loss of pumping ability. Think of it this way; at steady throttle, the throttle plate is a fixed leak. Outside is ambient pressure, while inside the manifold, pressure is greatly reduced each time a descending cylinder increases the volume. When one of four cylinders looses the ability to pump, the result is a ballpark 25% increase in MAP, as the size of the leak remained the same.

The MAP increase differentiates this example from an ignition problem. An ignition loss doesn't change the mechanical pump, i.e. piston and valve motion. A double spark plug foul would net a loss of RPM, as the pumping work continues without the power, a dead loss. An RPM loss would result in a slight MP increase, but there was no RPM loss here (note the yellow trace). Now look at the fuel flow trace, which appears to increase.



As Bill said earlier, fuel flow is important to the diagnostic journey. It tells us which valve. Why is perhaps easier to grasp with a picture, so I spent a few minutes drawing one while having my morning coffee. Each row is a progression through 720 degrees of crank rotation, firing order 1-3-2-4. Top row is a stuck #2 intake, bottom row is a stuck #2 exhaust. Here's the key; a stuck intake results in an additional intake stroke, during what would normally be a power stroke. In addition, many carbs will add fuel to the airflow regardless of which direction the flow passes through the venturi; that happens during what would normally be a compression stoke. The end result is a rise in fuel flow. A stuck exhaust, on the other hand, results in a reduction of fuel flow. The cylinder can't pull intake charge very well with an open exhaust port.



Now let's go to the inflight power loss. MAP is not plotted, we won't see a change in MAP, as the throttle plate is fully open. We do, however, see a big jump in fuel flow...exactly concurrent with the start of the #2 cylinder EGT event. Andy told us he went full rich, then lean, and power recovered. It would be very difficult for a pilot to react concurrent with the event, and besides, why go full rich when the initial indication would be roughness while leaning? The normal reaction would be to enrich slightly. I think Andy's full rich application was about 20 seconds after the start of the roughness event; remember, he told us "...I swapped fuel tanks, turned on the boost pump with no change." Actually, power recovery was already on the way, as in that 20 seconds the #2 CHT had dropped over 30 degrees, and the valve head was being bathed in a two-way flow of relatively cool intake charge. It unstuck at 02:02:20.



The power loss at 02:55:55 shows exactly the same pattern; fuel flow jump exactly concurrent with the start of the EGT event:



I'm still puzzling over one tiny detail. Both inflight EGT events begin with a very rapid rise in EGT for a very short time. The first event 85 degrees in 4 seconds. The second was 20~50 degrees (depending on where you start) over 5 seconds. Both events then tank the EGT; cylinder stone dead.

Popcorn ready ;)
 
Last edited:
No thanks on the popcorn Dan, I'm trying to get all this humble pie down...

I really thought I was running LOP, data does show differently. That was the first time I was high enough to have full throttle and be below 65% power and my first go at LOP at full throttle.
I never use the "lean assist" feature, I've peaked (pun intended) at it a time or two but it as I understand it it would take a Very slow leaning up to peak and past in order to work properly.
My previous LOP procedures (what I assume LOP anyway, I'll have to look over previous flights) was to reduce power to less than 65%. Playing with the throttle position around 60-65% power while still ROP evened out my EGT's and CHT's. I would make a big pull on the mixture to roughness or 7-8 gph and fine tune the mixture. I would occasionally enrichen from LOP and verify an EGT rise on all cylinders, #4 always peaked first and I used that value as peak EGT.

I probably transferred my previous LOP habits at partial throttle to full throttle, I saw the normal fuel flow and close EGT's/CHT's and assumed LOP. I normally see CHT's around 320 when LOP (assumed), I attributed the 350's CHT's on the flight in question due to full throttle.

This is from a flight earlier that morning, was I in fact LOP?

https://www.savvyanalysis.com/flight/894240/656989b5-e21e-4913-8439-5ba8c7bd1723

"Now let's go to the inflight power loss. MAP is not plotted, we won't see a change in MAP, as the throttle plate is fully open. We do, however, see a big jump in fuel flow...exactly concurrent with the start of the #2 cylinder EGT event. Andy told us he went full rich, then lean, and power recovered. It would be very difficult for a pilot to react concurrent with the event, and besides, why go full rich when the initial indication would be roughness while leaning? The normal reaction would be to enrich slightly. I think Andy's full rich application was about 20 seconds after the start of the roughness event; remember, he told us "...I swapped fuel tanks, turned on the boost pump with no change."

Affirm, that is how I remember it.

"I'm still puzzling over one tiny detail. Both inflight EGT events begin with a very rapid rise in EGT for a very short time. The first event 85 degrees in 4 seconds. The second was 20~50 degrees (depending on where you start) over 5 seconds. Both events then tank the EGT; cylinder stone dead."

I really hope I wasn't flirting with detonation.......:(
 
Last edited:
This is from a flight earlier that morning, was I in fact LOP?

No.



...EGT events begin with a very rapid rise in EGT for a very short time. The first event 85 degrees in 4 seconds. The second was 20~50 degrees (depending on where you start) over 5 seconds.

I really hope I wasn't flirting with detonation.......:(

Very unlikely at 18 inches and 2500 RPM.
 
Maybe maybe.

Y
I'm still puzzling over one tiny detail. Both inflight EGT events begin with a very rapid rise in EGT for a very short time. The first event 85 degrees in 4 seconds. The second was 20~50 degrees (depending on where you start) over 5 seconds. Both events then tank the EGT; cylinder stone dead.

Popcorn ready ;)

Good, solid analysis (as usual) Dan. I think the temp rise just might, might, be that the sticking event progressed. In the first few strokes, it almost went to the seat, but not quite. That allowed the combustion force extra hot gasses out the exhaust valve with out the benefit of work being done on the expansion. This superheated the valve very quickly and suspect that it stuck more open and might even have been closed by piston assistance. Then, like your explanation, it cooled due to no combustion and resumed " normal" function.

As example, a diesel engine uses exhaust valve opening right at TDC for an exhaust brake function. The heat of compression is kept in the exhaust and will bootstrap the turbocharger. In fact - it can be so intense that cylinder pressures under braking can be higher than rated power as the intake boost is pushed higher than full load output!!

Just a thought.

finally: This would lead me to look soon at the engine for a possible bent pushrod from the slack. I shiver thinking about what happens there with a gap.
 
Last edited:
Bill, that works for me, if it was an exhaust valve sticking...but the sticky one appears to be an intake.

Does low dynamic compression result in a long combustion time? Looks like coffee with Mr. Taylor in the AM ;)
 
Quite a mystery, #2 seems lean (or less rich) to me compared with the other cylinders - perhaps intermittently.
Look for the obvious first - you mentioned replacing fuel lines - check you are not sucking air somewhere.
Check the carby mounting bolts, check inlet tubes, clamps and flange gaskets. Check the carby's finger strainer and fuel flow.
 
Last edited:
Bill, your supposition about partial sticking at the beginning of the event (yeah, I know, maybe) resulting in increased EGT would work for an intake. There is plenty of data linking decreased compression to increased EGT. Apparently has nothing to do with combustion time, but rather work vs enthalpy.

Example snip borrowed from an engineering school paper:
http://rescomp.stanford.edu/~efroeh/papers/RDH_Engine_Performance.pdf



An intake that started sticking just off its seat would lower dynamic compression, and the EGT indication would shoot up. The valve itself would heat very quickly with limited or no heat transfer to the seat. Stem diameter would increase, stick the valve at a point further open, and EGT would tank with no combustion.

 
Last edited:
Great in depth analysis
I read all of Dan Horton's post with great interest.

Only one thing is bugging me about the whole intake valve sticking.
I cannot refute any of the details or come up with a more likely cause than what Dan has described.
Generally speaking though, exhaust valves are almost exclusively the only valves sticking in aircraft engines.
Rarely is it an intake valve and then only due to improper machining and assembly of cylinder head components.

You'll know soon enough when you remove the rocker valve cover and check you valves.
 
Last edited:
I had the valve seat come out of the head on an intake valve in my Franklin engine----------wonder if the Lyc is subject to the same issue??
 
All my valve troubles were with a Franklin on a Stinson 108
The description of the Op's engine troubles exactly matches my own experience.
But as Dan pointed out a graph analysis of EM data always beats a pilot description.
Mike,
Having an intake valve seat come out is probably more common on a Franklin since there is almost no one left who knows anything about properly assembling a Franklin cylinder. In any case, it is precisely the cause if improper cylinder assembly and could happen in either a Franklin or a Lycoming.
 
Generally speaking though, exhaust valves are almost exclusively the only valves sticking in aircraft engines. Rarely is it an intake valve and then only due to improper machining and assembly of cylinder head components.

Just for fun, I typed "stuck intake valve" into the word string search block on the NTSB site. Only got two hits, both interesting. CEN12LA384 was a 40 hour airplane, and NYC99LA173 was a freshly "repaired" valve stick.

Repeating the search with "stuck exhaust valve" netted 32 returns.

I guess we could say a stuck exhaust is 16 times more likely, but a stuck intake cannot be discounted.

You'll know soon enough when you remove the rocker valve cover and check you valves.

Review NYC99LA173 ;)
 
"Review NYC99LA173 "

So the mechanic unstuck the valve but did not fix why it stuck in the first place, When it stuck again later it stuck closed and the rocker arm ended up being the weak link?
 
I am a little worried that it was improper machining and or assembly, I was the 3rd owner (far as I know) of the engine after it had been rebuilt.

What is the possibility of my operation close to peak EGT causing the sticking valve? I was always less than 65% power and 380 CHT's, but what about 91 octane and 32 degrees of timing in the mix?
 
Back
Top