One cylinder running warm

azure

azure

Final Approach
Joined
Apr 2, 2005
Messages
8,302
Location
Varmint Country
Display Name
Display name:
azure
A couple of weeks ago I started my engine (IO360-A1B6) and noticed that it acted for a couple of moments as if it were about to quit, then it smoothed out and ran fine. However, during run up when I pulled the throttle back to idle, it DID quit. I had my mechanic check it out thoroughly. The dx: an induction leak on #4 and fuel injectors that were oriented every which way, his mistake at annual. So last week he fixed the induction leak and reoriented the injectors and then said, go fly it.

Well when I flew it this weekend I didn't like what I saw at all. First of all, the engine RPM quickly dropped about 15 seconds after start, just like the previous time. Then I noticed that in cruise the #1 EGT was about 60 degrees above where it had been prior to his adjustment, and was now higher than #3 which has ALWAYS had the highest EGT in cruise. My JPI was reporting DIFFs of 90-110 degrees in cruise, though only 40-50 at 1800 RPM during runup. Before I was consistently seeing 40-60 degrees DIFF both during runup and in cruise. All the other cylinders are running about where they were as far as I can tell: the difference is that #1 is running significantly warmer. The CHTs do not seem to have changed at all, as far as I can tell.

My concern is that the #1 valve or valve guide might be damaged from having stuck at some point and that fuel is now burning where it shouldn't, and that my engine's hesitation after starting could be morning sickness. Although my A&P is willing to do the wobble test to rule that out, I really had to talk him into it. He says that sticking valves are very rare in Lycoming engines. That part conflicts with everything I've read. What say folks here? And what else would you suggest we look at? He plans to disassemble my fuel injection system and check for obstructions and leaks everywhere he can check.

If it matters, the engine is about 200 hours since zero time.
 
Last edited:
Sticking valves in Lycoming parallel valve engines are common thus SB 388. It is rare in angle valve engines like yours but it does happen. I'd do the wobble test and borescope a look at that exhaust valve. Charlie Melot Zephyr ngines
 
Single cylinder high EGT can also be a plug issue. Proably worth looking into before spending big bucks on other issues.
 
At least on my three fuel injected engines(two Continentals and a Lycoming) the injectors thread into the cylinder head- there is no orientation. He could have mixed them up between cylinders, so his re-do might have been random as well. You could switch injectors between the cylinder that peaks first to the one that peaks last and see if that fixes the problem.

A partially plugged injector would also make a cylinder run lean. Hoppes 9 gun cleaner seems to be a common cleaner for injectors.

I think a stuck valve would manifest itself as "morning sickness" rather than a hot running cylinder, at least as an initial symptom.

Finally, your fuel servo might need adjustment or overhaul (uneven fuel flow to all cylinders after checking injectors) or the idle fuel pressure needs adjustment (engine quitting at idle).

If it ran well prior to your last mx, I would be thinking first about switching injectors and being sure they were clean.
 
A bad or fouled plug should be pretty obvious on a mag check.
 
The plugs are apparently okay -- my A&P checked them recently because I was seeing borderline excessive RPM drops on mag check during the warm weather.

The CHTs are normal and no different than they were -- upper 200s and low 300s at any reasonable mixture setting, at least at this time of the year (I've seen 350 or so on #3 during the summer, but never any higher).

I could be misquoting him, but I think my A&P said that my injectors fit into some kind of gasket and there are letters or numbers that are supposed to face upwards. I didn't ask whether the injectors were interchangeable but if they are, that's an interesting idea. Move the #1 injector somewhere else and see if the higher than normal EGT follows the injector. But he says he can directly measure flow rates, and that sounds like a better first step.

Okay, what does morning sickness act like? When it's preheated but is being started for the first time of the day, my engine acts like it's not going to stay running. Most descriptions I've heard of morning sickness are that the engine runs rough, and shakes like crazy or makes weird sounds. Mine doesn't do that, but could it still be morning sickness? It has yet to do it when completely cold, and this weekend when I started it after it had been standing out in the cold for a couple of hours, it started fine without a hint of it.

I don't think the idle check shutdown was due to a misadjusted idle setting. It happened once, and hasn't come close in probably 20 runups since then.
 
azure said:
Okay, what does morning sickness act like? When it's preheated but is being started for the first time of the day, my engine acts like it's not going to stay running. Most descriptions I've heard of morning sickness are that the engine runs rough, and shakes like crazy or makes weird sounds. Mine doesn't do that, but could it still be morning sickness? It has yet to do it when completely cold, and this weekend when I started it after it had been standing out in the cold for a couple of hours, it started fine without a hint of it.
Click to expand...

Morning sickness is a term that is used when any engine runs rough, until it gets warm.

Normally morning sickness is a valve that stays open (doesn't close) until the valve guide gets hot and swells enough to release the valve. then the engine will run normally.

OBTW, morning sickness is not the only thing that will cause an engine to run rough until it gets warm.
 
azure said:
Okay, what does morning sickness act like? When it's preheated but is being started for the first time of the day, my engine acts like it's not going to stay running.
Click to expand...


When engines get pre-heated they often form ice in the induction system. much like carb ice, Warm air from the heater carries a lot of moisture, it drops the water as soon as it hits the cold metal, any of that water that is inside of the induction system must clear before your engine will run smooth. (5-10 seconds at most)

Condensate on the ignition harness or mags will cause the same symptom.
 
If you have a persistently stuck valve, you will know- it is obvious. You can also do a mag check in flight at high power, which should indicate if there is a plug or lead problem.

This still sounds like a fuel problem, and the easiest thing to do would be to swap the injectors and make sure they're clean. My impression of bench fuel flow is that it might not necessarily correlate with actual egt.

If the ignition checks and switching injectors doesn't help you can always pull the valve cover for the cylinder in question and try to move the valve and see if it's stuck when the engine is cold.

Some people swear that Marvel Mystery Oil does wonders to help prevent valve sticking.

Finally, bad idle, engine stops, rough running, etc after checking the above could indicate that your fuel servo might need an overhaul.

Have experienced all of these in various combinations in my previous AEIO 540, IO 470's, and AEIO 360.

Good luck.
 
Let's look at this one issue at a time. It seems to me like it's probable you have some sort of fuel issue. I don't think this is a sticking valve (which, by the way, are common on Lycomings even though I've never had one in my multiple thousands of hours of Lycoming run time).

Unless you have GAMIjectors, I'm not sure what your A&P could have done to switch them around. GAMIs do have particular cylinders they belong in. Otherwise, they should all be identical. To that extent, they might need to be cleaned. The Hoppes #9 for 30 minutes is the way to go here. Otherwise, you might have some adjustments you need to make to the servo. Most likely, you might just need to change the idle screw a bit to leave the throttle plate a bit more open at fully closed. So long as you're still able to hold idle at an appropriate RPM with the throttle closed, that should be fine.

I would also do an in-flight mag check to check your ignition. A lot of times, a bad plug or lead will make itself known on an in-flight mag check, but you won't notice it on the ground.

That said, out of what you've listed, the rough starts are the only one that I'd be particularly concerned with. A 100 degree spread on EGTs isn't so bad.
 
Tom-D said:
When engines get pre-heated they often form ice in the induction system. much like carb ice, Warm air from the heater carries a lot of moisture, it drops the water as soon as it hits the cold metal, any of that water that is inside of the induction system must clear before your engine will run smooth. (5-10 seconds at most)
Click to expand...
This is a Reiff pre-heater, there's no warm air being pumped in. I'm not sure where the extra moisture is coming from, unless it's from the last run of the engine itself, trapped under the cowl.
 
Thanks everyone, it seems like my A&P is probably on the right track then.

Ted, I don't have the GAMIjectors, and he didn't say that he switched the injectors around, as in swapped them. From what I understood, he was talking about the orientation of each injector where it fits into the cylinder. He said something about the letters (or numbers) having to be on top. I haven't seen the system up close so I'm not really sure what he was talking about.

If I get a chance I'll do another in-flight mag check. I did this already in the late summer because of the large RPM drops on the ground, and the drops were much smaller in cruise. The reason the 100 DIFFs concerned me is that it's new, since he reoriented the injectors. I'd never seen #1 run that high before, even before annual when he supposedly randomized the injector orientations.
 
I wasn't aware that the letters needed to be pointing in a particular direction. Even so, you would be surprised how little something like that matters over the quantity of fuel being injected.
 
AcroBoy said:
A bad or fouled plug should be pretty obvious on a mag check.
Click to expand...

I agree, but as ignition is cheap and easy to check it's often at the top of my troubble shooting list if it can be at fault.

Odds are good it's the mix in that cylinder though, so it's fuel or air. If the idle is smooth I wouldn't suspect an induction leak and would look towards fuel. Starting at the injectors and working back.
 
Thanks for all the replies. If the cowl is still on tomorrow I will try to take it up and do a mag check in cruise. Now assuming it IS the mixture in that cylinder, how do I lean the engine? Just treat #1 as the cylinder to peak first (it is, now) and go to 100-125 ROP?
 
In my experience, I've seen thermocouples die, or work intermittently, but never give sudden erroneous readings.
 
If you mean the JPI exhaust temperature sensors, according to my A&P they're welded on (makes sense) and moving them so they're all the same distance from the flame is something he didn't really want to tackle right now. I don't think he'd feel any differently about swapping them.
 
azure said:
Thanks for all the replies. If the cowl is still on tomorrow I will try to take it up and do a mag check in cruise. Now assuming it IS the mixture in that cylinder, how do I lean the engine? Just treat #1 as the cylinder to peak first (it is, now) and go to 100-125 ROP?
Click to expand...

If you want to run ROP, then that'll work fine.

If you want to get best economy, just lean until you get your roughness or feel a decrease in power, and then back to smooth.
 
For the JPI, which I have in three engines, the EGT probes are attached as I recall with a hose clamp-like band around the exhaust. Never heard of them being welded either. Also, it's not the absolute temperature that's as important as much as the one that peaks first when leaning, or seeing a change from previous behavior.

Also, never heard of orientation of the injectors, either. All the ones I've seen (GAMI included) are a straight shot and not angled. They are tapped into the cylinder head and are simply tightened to spec. Any of the IA's or A&P's care to comment?
 
Again, thermocouples tend to die outright or work intermittently, but not give false readings. There would be no reason to swap them as long as they are providing a reading.

Easiest thing would be to pull top of cowling and switch injectors. While they're out, clean them in Hoppe's 9 (available from sporting goods stores- it's nitrobenzene), and swap the two that were hottest/coldest, or the first and last to peak as you leaned.

Parts of the story as you relate them from your mechanic are starting to sound suspicious.
 
AcroBoy said:
Again, thermocouples tend to die outright or work intermittently, but not give false readings. There would be no reason to swap them as long as they are providing a reading.

Easiest thing would be to pull top of cowling and switch injectors. While they're out, clean them in Hoppe's 9 (available from sporting goods stores- it's nitrobenzene), and swap the two that were hottest/coldest, or the first and last to peak as you leaned.

Parts of the story as you relate them from your mechanic are starting to sound suspicious.
Click to expand...
I'm pretty sure his exact words were, when I asked a week ago whether he had adjusted the position of the probes, that he hadn't because it would mean "unwelding" them. That's what implied to me that they were welded. Maybe I misheard or misunderstood. But he certainly said it would be a major job -- "a bit extreme" was the phrase he used.

Once again last night, for a few seconds after start, the fuel pressure readings were wildly erratic and the engine acted as if it was not going to stay running. The JPI stays stuck on the battery voltage for a couple of minutes for some reason so I had to manually cycle through the other pages to see what the EGTs and CHTs were. I got to the DIFF page and it was in the 200-260 range. The graphical display showed a smooth decreasing trend from #1 to #4. Then as soon as I advanced to the individual cylinders the display leveled out. I didn't notice what the actual readings were as I raced to get back to the DIFF page. This time it said only 40, 50 degrees. So the episode was over and the engine ran fine from then on. No sign of quitting during idle check, normal RPM drops on mag check, DIFFs in the 40-60 degree range. But even at 1800 RPM, #1 EGT is higher than #3 now, which is new behavior (last 2-3 weeks, since he worked on the injectors).

Mag check in cruise was completely normal. No discernible RPM drop (I assume because of the governor -- you can hear the difference on one mag though, and the engine runs smoothly on either one).
 
If the JPI is stuck on the battery voltage page, is it flashing at you? It might be because you have low voltage.

If your fuel pressure is fluctuating, does turning on the boost pump stop the fluctuations? Your fuel pump may be going out.

High EGT differences at idle aren't uncommon. I really wouldn't put much stock in that.
 
Ted DuPuis said:
If the JPI is stuck on the battery voltage page, is it flashing at you? It might be because you have low voltage.
Click to expand...
Errr.... I think I'd know if the voltage was low. ;) No, it's not flashing, and it's typically 14.0 or 14.1. As I said, I have no idea why it stays on that display for so long.

If your fuel pressure is fluctuating, does turning on the boost pump stop the fluctuations? Your fuel pump may be going out.
Click to expand...
Funny that I didn't think of trying that. Next thing to try.

High EGT differences at idle aren't uncommon. I really wouldn't put much stock in that.
Click to expand...
But then it normalizes by itself. It's a transient thing, the first minute or so after engine start. That's why I keep coming back to morning sickness. But I will try the pump.

The other thing I thought of is that I am using the normal start procedure, priming with the fuel pump, even though the Reiff gets the oil temp up to 140 F or so. At that temperature the engine doesn't start as easily as when it's cold, acts almost like a warm start. I wonder if I might be "overpriming" it.
 
azure said:
Errr.... I think I'd know if the voltage was low. ;) No, it's not flashing, and it's typically 14.0 or 14.1. As I said, I have no idea why it stays on that display for so long.
Click to expand...

Start with the obvious. Nothing against you, it's just a common mistake pilots make. :)

Funny that I didn't think of trying that. Next thing to try.
Click to expand...

The next thing to look at, if that does fix it, is what happens to your fuel pressure when taking off if the electric pump is off. If it fluctuates or goes down, then your mechanical pump is going out. That happened on my Aztec.

But then it normalizes by itself. It's a transient thing, the first minute or so after engine start. That's why I keep coming back to morning sickness. But I will try the pump.

The other thing I thought of is that I am using the normal start procedure, priming with the fuel pump, even though the Reiff gets the oil temp up to 140 F or so. At that temperature the engine doesn't start as easily as when it's cold, acts almost like a warm start. I wonder if I might be "overpriming" it.
Click to expand...

Yes, you don't need to prime as much when the engine is warmed up, even if it's cold outside.
 
Ted DuPuis said:
Start with the obvious. Nothing against you, it's just a common mistake pilots make. :)
Click to expand...
Okay, no offense taken. It's just that I've seen the flashing BATT before -- alternator failure last year, then a one-time CB trip earlier this year -- so I'm pretty attuned to that. Any time the JPI flashes at me, I take notice.

The next thing to look at, if that does fix it, is what happens to your fuel pressure when taking off if the electric pump is off. If it fluctuates or goes down, then your mechanical pump is going out. That happened on my Aztec.
Click to expand...
Well, nothing happens, since that is normal takeoff procedure in the Cardinal. The only time I normally use the fuel pump is for priming.

I have noticed, though, that since my A&P re-oriented the injectors, the fuel pressure on takeoff is back over redline, the way it used to be before my first annual in March/April. According to one of the CFO gurus, that's the way it should be -- in fact, he says that if it's under redline on takeoff, something needs adjustment.

But it's very stable and doesn't fluctuate or come back down until I back off on the throttle a bit during climb (also normal procedure in this bird).
 
Rough at start then smooths out could be a cylinder with morning sickness. You should easily see this on the JPI as a cylinder that does not light off and start coming up to temp when first starting.

Secondly, a reversal of cylinders that peak first could be that the injectors were switched- they are each microscopically different from the factory, which is why people pay lots of money to go to GAMI injectors, especially in big bore Continental engines versus Lycomings. It might not matter as much in a smaller displacement engine from either company.

Third, uneven fuel flows and pressures could also indicate your fuel servo needs an overhaul.
 
azure said:
I'm pretty sure his exact words were, when I asked a week ago whether he had adjusted the position of the probes, that he hadn't because it would mean "unwelding" them. That's what implied to me that they were welded. Maybe I misheard or misunderstood. But he certainly said it would be a major job -- "a bit extreme" was the phrase he used.
Click to expand...

It could be whatever the probes are attached to are welded to the exhaust system making them somewhat inconvenient to adjust...

But, I'm a car guy (where we weld taps on the exhaust system so the thermocouples can pass through the pipe and not leak), not an airplane guy. So I don't really know ****.
 
AcroBoy said:
Rough at start then smooths out could be a cylinder with morning sickness. You should easily see this on the JPI as a cylinder that does not light off and start coming up to temp when first starting.
Click to expand...
That's the main thing I'm worried about. So the cylinder with the sticking valve would stay cold while the others came up to temp? I haven't seen that at all -- just wild fluctuations in fuel pressure at first.

I think it only happens after using the Reiff. I flew it to JXN today and tried to let the engine cool off for an hour (it was something like 10C out), but that wasn't long enough, and it just acted like a hot start (but without fuel pressure fluctuations). I left the preheater unplugged now, and will try to start it tomorrow and see if it acts up (should be warm enough that I don't really need preheat).

Secondly, a reversal of cylinders that peak first could be that the injectors were switched- they are each microscopically different from the factory,
Click to expand...
But would he need to remove them all just to rotate them one at a time? It seems like if there was a chance he could have swapped them accidentally, he would know it.

Third, uneven fuel flows and pressures could also indicate your fuel servo needs an overhaul.
Click to expand...
I think that's what he suspects.
 
I started the engine again today, without preheat, and confirmed what I suspected -- there was no "morning sickness", all the CHTs warmed up evenly at low RPM (except that #3 was noticeably ahead of the other three, which is nothing new), and the fuel pressure fluctuations were much less extreme. I also did a couple of runups, at 1800 and then at 2200, confirming that #1's tendency to run at higher EGT than the others starts around 1800 and gets more pronounced at higher power settings.

Again, #1 runs at higher EGT than the others but at a lower CHT than #3. The CHT part is the same as always; only the EGT has changed since my A&P turned the injectors around.

Does this tend to point to any one of the many possibilities that people have mentioned?
 
AcroBoy said:
Again, thermocouples tend to die outright or work intermittently, but not give false readings. There would be no reason to swap them as long as they are providing a reading.

Easiest thing would be to pull top of cowling and switch injectors. While they're out, clean them in Hoppe's 9 (available from sporting goods stores- it's nitrobenzene), and swap the two that were hottest/coldest, or the first and last to peak as you leaned.

Parts of the story as you relate them from your mechanic are starting to sound suspicious.
Click to expand...

I've replaced ones that were reading high, that were verified incorrect via IR thermometer, and after replacement, were reading correctly.

Never say never. Do whats easiest first.
 
I had a scary flight this evening returning from a fuel run to 57D. On the way there, I almost thought the problem was starting to correct itself -- I was getting DIFFs in the 70-80 degree range, and the EGTs on #1 and #3 were about even. When I started up again at 57D, I noticed during the runup that the #1 EGT was higher than before, and I was getting DIFFs of 70 to 80 degrees, which had never happened at runup settings. Then after takeoff the #1 EGT started rising beyond what I had seen before. It went up to about 1390 during climbout and stayed there. DIFFs hovered at 190-200. I ran the engine practically full rich the whole way back and kept an eye on the CHTs, which fortunately stayed comfortably low, under 300. Scariest night flight I've ever made -- and in hindsight, I probably should have turned back when I noticed that #1 was running higher than it had been recently.

Back at VLL I did a straight-in and hailed my A&P, who happened to be working late. We did a full-power runup and verified what I had seen in the air, then leaned out to find where the different cylinders would peak, under his direction. #2 peaked first at about 1470, but #1 ws still rising and peaked at 1520, the highest EGT I've seen on ANY cylinder since buying the plane (though Tom recorded EGTs like that during the first few hours of break-in). Strangely, #2's EGT stayed about 150 above the others even after bringing the throttle back to idle, and even its CHT stayed up around 340 -- the highest CHT I saw today. We did a second full-power runup without leaning, which caused the #2 CHT to settle into the high 200s. After pulling the power back again, all the EGTs and CHTs came back down to normal idle levels.

Things my A&P is planning to check:

1. Possible problems with the JPI EGT probes - he wants to swap 1 and 2 with 3 and 4 to see if the high readings follow the probes. But does that make sense in view of the high CHT on #2? My gut feeling is no -- something real is/was definitely going on with at least that cylinder.

2. Possible intermittent induction leak(s) due to slipping O-rings, or even a cracked flange.

I'm really concerned that both of those cylinders might have been running in the "red box", especially #1 during flight but also #2 during the runup, and I wonder if the persistently high CHT on #2 might be an indication of detonation.
 
Don't worry about hurting your engine with what you did - it is quite tough. 1520 EGT isn't that bad for it, but it is still not bad at all in the grand scheme (I routinely run 1500-1550 in cruise on my IO-520s, with weaker cylinders). A CHT of 340 is also low. Remember, the limit on Lycoming cylinders is 475-500 (depending on the cylinder). I aim to keep below 380 (that's the goal I recommend), and on the Aztec that is exactly where the hottest cylinders are running at all times. At 340 and below, you are not only below the limit, you're below the goal. I wouldn't worry about it.

As far as detonation goes, to make that engine detonate, you'd need heads right at the limit and hot induction air temperatures (try 100F), neither of which you had.

I'd suspect an induction leak, a partially-fouled plug (this would be my biggest guess), or potentially a clogged nozzle or some other issue with your fuel system.
 
I think by "welded" the sensor locations are fixed (welded) on the exhaust headers - the sensors i would guess are threaded in or otherwise attached to these points.

If your lycoming is running smoothly and everything is consistent and well within limits there is probably no reason for alarm.

Thermocouples are very sensitive and reliable but they are subject to inaccuracies and their mounting locations coupled with things like airflow / cooling can lead to discrepancies.

The difference in readings you are talking about is something like three or four percent. If everything is running smoothly, your A&P is not alarmed, and your readings stay consistent... just fly it.
 
You must log in or register to reply here.
Back
Top