Exhaust smell in cockpit with no CO

[azure]

On a recent long cross-country at 9000, running LOP, I noticed an increased smell of what seemed to be exhaust in the cabin; however, my Aeromedix CO monitor read zero. The smell lingered for a couple of days until I aired the plane out today, then took it up to see if the problem would recur. It did not. I did not, however, run it LOP.

I consulted with the FBO owner (who is also an A&P and IA) as to the possible cause, and his opinion was that I might have had a carbon deposit in the exhaust system that was burning because of the higher EGTs, but no actual exhaust. This does make some sense, as at 9000 I was not concerned about damaging the engine and was running with the #2 cylinder (the richest) only very slightly LOP, with an EGT in the high 1400s to sometimes over 1500.

Can an EGT this high be damaging to the exhaust system?

Does the FBO owner's explanation make sense? Any way to nail down the cause for certain?

 

[Kelvin]

He may be spot on but checking or replacing the CO2 sensor would be on my short list.

 

[Cap'n Jack]

If it were carbon on the inside of the exhaust pipe, you I feel you shouldn't be able the smell it burning off in the cabin- it should leave with the rest of the exhaust. My understanding is that whatever is going through the exhaust pipe shouldn't get into the cabin heater, nor into the cabin. I do note you didn't mention the use of cabin heat.
I'd check the CO sensor (as suggested above), also check for leaks in the exhaust system. Was work recently done on the plane such that something on the outside of the engine/exhaust may have cooked off? That would explain the symptoms too.

 

[Checkout_my_Six]

maybe you were just smelling burning oil...and not exhaust?

 

[Capt. Geoffrey Thorpe]

LOP the CO content of the exhaust should be pretty low and a small exhaust leak might not show much in cabin CO.
Below one of the typical cartoons...
Peak EGT is pretty close to lambda = 1 (stoichiometric) LOP is to the right. Rich operation tends to be around 0.8ish lambda. So you are looking at about a factor of 10 reduction in CO from running LOP. Ballpark.

This doesn't explain why you didn't see/smell when ROP. Could have been something else getting hot and not exhaust in the cabin, or ???

 

[Checkout_my_Six]

Oh boy....

 

[azure]

He may be spot on but checking or replacing the CO2 sensor would be on my short list.

If it were carbon on the inside of the exhaust pipe, you I feel you shouldn't be able the smell it burning off in the cabin- it should leave with the rest of the exhaust. My understanding is that whatever is going through the exhaust pipe shouldn't get into the cabin heater, nor into the cabin. I do note you didn't mention the use of cabin heat.
I'd check the CO sensor (as suggested above), also check for leaks in the exhaust system. Was work recently done on the plane such that something on the outside of the engine/exhaust may have cooked off? That would explain the symptoms too.

Yeah, that is what I was wondering about, something IN the exhaust system shouldn't get into the cabin. But if there were leaks in the exhaust system, it shouldn't matter whether I'm running ROP or LOP... and today I noticed no smell, nor any significant CO (2 to 3 ppm on takeoff, which is normal in my plane).

I probably should order another of their calibration kits to be sure...

 

[azure]

This doesn't explain why you didn't see/smell when ROP. Could have been something else getting hot and not exhaust in the cabin, or ???

Exactly why I do not suspect an exhaust leak at this point... I suppose it could have been a deposit on the engine itself burning. And yes, the plane was recently annualed, along with an oil change.

No one answered my question about whether running at altitude with basically a stoichiometric mixture could be damaging to the exhaust system. How hot does the EGT have to be to be potentially damaging?

 

[Checkout_my_Six]

don't worry about being "stoichiometric" most won't even know if they are there.

exhaust gets hot....any oil near it will burn...and if there is a firewall leak...that's why you smell it.

 

[azure]

don't worry about being "stoichiometric" most won't even know if they are there.

Not so sure about that. Let's say you have the standard single cylinder EGT gauge, with the probe on the leanest cylinder. As Capt. Thorpe said, stoichiometric on that cylinder is basically peak EGT, and the others will be well ROP. So if you lean properly for ROP, you'll hit stoichiometric on that leanest cylinder and then back off.

My plane has an EDM-700, so I have both CHT and EGT readouts on all 4.

exhaust gets hot....any oil near it will burn...and if there is a firewall leak...that's why you smell it.

Yep, that's certainly true, and the likeliest explanation. I'm just wondering if there's a way to be sure.

 

[Cap'n Jack]

don't worry about being "stoichiometric" most won't even know if they are there.

exhaust gets hot....any oil near it will burn...and if there is a firewall leak...that's why you smell it.

LOL. I know what "stoichiometric" is, but I don't know if the materials comprising the exhaust can stand the temperatures @azure mentioned.

Yep, that's certainly true, and the likeliest explanation. I'm just wondering if there's a way to be sure.

Light off a smoke pencil in the cabin and the smoke points away from a draft, maybe?

 

[Checkout_my_Six]

you won't get a purely "stoichiometric" process trust me....there will be bi-products. Even running LOP there is un-burned fuel going out the exhaust. The nature of the delayed flame front puts burned products out the exhaust.

 

[Tarheelpilot]

Exactly why I do not suspect an exhaust leak at this point... I suppose it could have been a deposit on the engine itself burning. And yes, the plane was recently annualed, along with an oil change.

No one answered my question about whether running at altitude with basically a stoichiometric mixture could be damaging to the exhaust system. How hot does the EGT have to be to be potentially damaging?

One of the worst things that ever happened in engine management was putting numbers on EGT gauges. You can’t get it hot enough to hurt anything...

 

[Capt. Geoffrey Thorpe]

One of the worst things that ever happened in engine management was putting numbers on EGT gauges.

If you want to give up an hour watching an hour and a half video of Mike Bush (must watch at 1.5x speed) that is his whole thing about EGT - numbers mean nothing due to the way they are measured. CHT, on the other hand, can send up the red flag.

He also loves to tell you how LOP will save you money, make your engine last longer and whiten your teeth.

I ain't no A&P, but of all the antilop stuff out there, I have never heard of exhaust system damage as an excuse.

 

[Checkout_my_Six]

the APS guys are worse than Mikey.....

 

[azure]

I ain't no A&P, but of all the antilop stuff out there, I have never heard of exhaust system damage as an excuse.

I haven't either, and it's always been my understanding that you can't get it hot enough to damage anything. OTOH, the FBO owner told me yesterday about a 210 he had seen with a badly deformed collector (not sure where that is in an exhaust system, nor whether this was a turbo engine or NA) that had apparently been exposed to some pretty high temps.

Just wanted to check that understanding against the collective wisdom here.

 

[azure]

First, exactly what type of smell ? The sweet 100LL odor, or a burnt oil odor ? One that lingers several days might be oil getting into heat muff and coming in heat vent ? Are you running any heat at 9000 ?

Yes to both. Definitely not a sweet smell, but a heavy, burnt, possibly burnt oil odor. I was indeed running some minimal heat up there, especially after I started punching through clouds.

 

[wsuffa]

I'd replace the CO sensor first. If for no other reason than peace of mind. I say that based on experience. Here's the story:

My Commander was turbocharged. Started getting some "exhaust" smell in cockpit, no alert on the CO detector. Happened a couple of times, noted that TIT/EGT was higher than normal, as was one cylinder CHT. Other than that, nothing amiss. Was worse on climb (attributed to hot day climb). Didn't see anything with cowling pulled off. <shrug>. Happened again, but on longer flight and I felt more tired than usual. Noticed some white exhaust deposits in unusual place in cowling. More careful check of exhaust system revealed crack on back side of exhaust pipe right above the TC - a place that was very hard to see on that engine. It had torched some of the wiring nearby, and one could see the evidence of the leak on the engine mount. Mechanic had missed the crack, so had I first time around. Tested CO detector - turned out it wasn't working properly. Replaced.

Found out later that CO detectors have a finite life, which is accelerated in an aircraft environment.

 

[azure]

I'd replace the CO sensor first. If for no other reason than peace of mind.

Agreed, in fact I checked this morning how long I've had it, and I bought it in 2011. They're only expected to last 5 years, but they also do a POST that is supposed to tell you when the sensor needs replacing. Mine keeps passing so I've assumed it's still good. Rather than take the chance, I just ordered a new one from Aircraft Spruce, same model.

 

[bobmrg]

Carbon monoxide is odorless, so it is unlikely that you are smelling it.

Bob

 

[azure]

Carbon monoxide is odorless, so it is unlikely that you are smelling it.

Bob

I'm well aware of that - but it's a component in exhaust, so normally I would expect that if it were exhaust I was smelling, a sensitive CO monitor would show a reading greater than zero, even when running LOP... unless the sensor was faulty.

 

[Cap'n Jack]

I'm well aware of that - but it's a component in exhaust, so normally I would expect that if it were exhaust I was smelling, a sensitive CO monitor would show a reading greater than zero, even when running LOP... unless the sensor was faulty.

I wouldn't go so far as to say there is no CO when running LoP, but the concentration may well be under the detection limit for that monitor.
See @Capt. Geoffrey Thorpe 's chart in post #5

 

[azure]

I wouldn't go so far as to say there is no CO when running LoP, but the concentration may well be under the detection limit for that monitor.
See @Capt. Geoffrey Thorpe 's chart in post #5

The #2 cylinder was pretty close to peak, or lambda = 1, the others not too far LOP. That graph shows CO concentration should be about 1% by volume. The question is then what the concentration of exhaust in the cabin air would have to be for me to smell it. Even if it's of order 1%, that should still be 10^-4 or 100 ppm. IIRC, the design sensitivity of the monitor is supposed to be about 1 ppm.

And the other point, which I think is irrefutable, is that if I had an exhaust leak, I should have smelled something yesterday when running ROP.

 

[1RTK1]

First, exactly what type of smell ? The sweet 100LL odor, or a burnt oil odor ? One that lingers several days might be oil getting into heat muff and coming in heat vent ? Are you running any heat at 9000 ?

I always have a trace of CO showing on my detector especially when taxing. If I use cabin heat I get an oil/exhaust smell with no CO rise. The cause for “my” smell is oil soaked scat tubing which hasn’t been replace for decades. I rarely use the heater but that will be fixed at annual.