Editor’s Note: This is the second article in our new series called “I Can’t Believe I Did That,” where pilots ‘fess up about mistakes they’ve made but lived to tell about. The aim is not to embarrass or demean the author, but to offer lessons that other pilots can learn (and hopefully avoid). If you have a story to tell, email us at: [email protected].
Why a Mag Check?
By Tom Winter
I lost a cylinder last time up. Here’s the story, with all details which I can recall, followed (figuratively, thank goodness) by a postmortem.
Typically flying the Cessna 150 in Cub-emulation mode about 1000 feet AGL, I cruise at 2450 rpm, and I always lean. My co-owner never leans. He’s an actual Cub flyer, and there is no mixture adjustment in his Cub. So I figure it’s on me to keep the plugs from fouling up with lead in our 150. Plugs for magneto ignition have narrow gaps, and there are a couple grams of lead in every gallon of 100LL.
The first abnormal sign was a bad mag check. Three guys, first me, then one of Lincoln’s most experienced pilots, then an older pilot, all thought plug fouling:
- On return to Lincoln from the National KR fly-In at Red Oak, I got a bad mag check–RPM drop over 100 right, less than 25 left–and figured it was fouling and what my engine needed was exercise, at higher than my normal RPM range. So I took off and flew home. Flew fine and smooth. Nice flight.
- Next flight, same bad mag check. Ran at full throttle a bit, then tried mag check at full throttle. Drop difference was 200 RPM! I aborted, came back from run-up, and asked an old hand at the FBO. “Nine out of ten, plug fouling,” he said. “You can’t hurt it on the ground. Lean until it coughs, then richen a bit. Even take off with some lean.” Tried it, and everything seemed OK. Flew fine. 200 miles. Ran strong as ever or stronger; I was happy. This was my first flight involving cloud avoidance. Sometimes under, and sometimes around fair weather cumulus. Joy! Hoped problem was over.
- Next flight, leaned on ground for taxi. Had line guy add a quart of oil (was at 4.5 quarts). It was cold, so we waited in the plane for the lineman to find the oil and return with it.
On the apron, I smelled a mild burnt oil smell. I imagined the line guy had spilled some oil, or that it was cabin heat. (My experienced pilot pax had opened cabin heat. I had never touched it before.) In retrospect, maybe I should have shut down on the chance the burnt smell was an oil leak. But the oil pressure gauge looked normal, and stayed there.
At runup, I got the same bad mag check. “What do you think?” I asked my more experienced pilot pax. “Fouling. It’ll clean out at cruise.”
Took off at full rich. Leaned while settling into cruise (leaned out to max RPM, then in some). More experienced pax showed me how to run the LORAN, which was a main reason for the flight. (Good news, the LORAN works!) Suddenly the arpies dropped from 2500 to 2200, all at once, nothing gradual about it. Carb heat did nothing, so we turned around and made it back to LNK. Full throttle, full rich, still just 2200 rpm. And it was shaking almost like a wet dog. But it got no worse.
Tower was great.
“Lincoln Approach, Cessna 2885S has lost 300 RPM. We’re coming back.”
“Cessna 85S, report airport in sight.”
“Lincoln Approach, Airport in sight”
Tower cleared us on the spot to 35R.
“Affirmative, 35R. But if you need something else, we can deal.”
Wind serene, I had been considering a downwind on 14, which was closest and straight ahead, but no need–steady at 2200 rpm all the way. Flew straight for the numbers, not descending at all. Still several hundred feet over traffic pattern altitude until the numbers were there in the windshield. Used full flaps to “divebomb” to the runway. (Nice landing!)
Pax, an IFR pilot, congratulated me on staying calm throughout. Mechanic at LNK said zero compression in cylinder #2, and non-ferrous flecks of metal were all over the screen. The screen was crushed. He wondered out loud if the wrong ring had been installed. I wondered out loud if I’d been leaning too much. Our mechanic in Crete came up and took the engine off. I’m on the ground waiting…
Fortunately, our larder is already well-stocked with beans and rice. Looks like I’ll be living on beans and rice for a while–again.
Saw my disassembled engine Saturday.
The cause of my sudden 300 rpm drop in flight was a hole in the piston. The hole was at the edge, next to the upper plug. I asked to see the plug. The plug looked like any other plug except it had no porcelain around the stem. The A&P didn’t blame it on the plug, but said it could have been a pit in the original piston becoming a hotspot. “Piston’s probably been in there since 1967. Nothing different you could have done to prevent it.” Bless him, he covered me in front of my partner–and to some extent his explanation may even be true!
Other possible causes? Old savvy pilot guys hear the tale and just opine “You leaned too much!” If so, the stock oil temperature gauge is useless. Oil temp says nothing about a hotspot. Another old hand related that a magneto getting loose and advancing would burn a piston.
They tell us in school to check the mags, but they don’t tell us what to do if you don’t like the mag check. OK. Here it is and my experience is what may happen to you if you don’t: Don’t take off if the drop difference is too big. Go back and have the plugs pulled no matter what the old hands tell you. It’s not their engine, it’s not their mortgage, it’s not their life. Even if it’s not the plugs, the plugs will have a story to tell, and through the plughole, they can even look at the piston and the sleeve.
I HAD gone back from runup the flight before, but bought the advice I got, namely to try unfouling the plugs by taxiing lean and flying lean, and then bought the advice of my pax that it would clean up!
So there. Go back, don’t take off until you see the plugs. If it’s just fouling, cheap. If it’s the one time in ten it’s something bigger, you find out on the ground. Still cheap.
Ben Franklin said it. “Experience is a harsh school, but a fool will learn in no other.” Not a fool? Good. Learn from my experience.
Tom Winter retires this year from the University of Nebraska after 45 years of college teaching — mostly Latin. He got his pilot’s license on March 17, 2000, at age 56. In his club plane (he helped start a flying club) and as a co-owner, he averaged 28 hours a year: since becoming sole owner of his 1967 Cessna 150, six years ago, he’s averaging between 50-60 hours a year. Tom’s retirement plan: fly to small-town strips, pull out the folding bike from behind the seat, and tour small town Nebraska. Also spend a week or so at Estrella and get a sailplane rating.
- I Can’t Believe I Did That #2 - February 8, 2013
Good article. There is a lesson for all of us here. How many years/hours since overhaul had it been before the event?
Mike Busch (Savvy Aviator in the EAA Mag) is a strong advocate of periodic checks with a borescope. Most metal failures will pre-announce themselves under such a check.
I couldn’t agree more. It takes only a minute to pull off the ignition lead and remove the spark plug, giving access for the borescope. This is an easy way to tell the condition of your exhaust valve, which is often the culprit of bad cylinder performance. Glad you made it home safe.
Just checked the enging log to answer Mark’s question: The first major overhaul was at 2531 tach hours; the burned out piston was at 2960 — so it happened about 430 hours after the major.
Just my two cents. I’ve had runups espically after long taxi’s where the mag check failed,lean aggressively run engine at power to clean off plugs for about a minute then reduce power go to full rich(unless aiport altitude dictates otherwise)and do the mag check again. Bottom line for me is your very lucky. If any engine check fails YOU DON’T TAKE-OFF!
My Mooney began to run rough at 4500 ft. Called approach and limped home. Performed a leak down and found number one cylinder was 58/80 and one of the plugs from that cylinder(a champion)was missing the porcelain. I have since come to learn that this is not uncommon (See Aviation Consumer). Always check the plugs first, they will tell the story.
Very enjoyable article, as it get’s right to the point. I really appreciate you sharing this experience. Cheers from YMEN.
I have to ask: what does oil temp have to do with leaning? What wrong ring? What screen? What were your leaning procedures? There is a comment that ‘the plugs may have been installed since 1967.’ When did this occur? And a reference to Crete. Where did this occur? Was the engine running mogas?
Did the engine have an EGT or more importantly, a CHT gauge? Which engine was this, a Lycoming or a Continental? Which mags, Slicks or Bendix? Which spark plugs, Champion or? Some fine wire Champion spark plugs have recently been blamed for damaging themselves. Damaged spark plugs plus a high CHT can lead to pre-ignition and the damage you describe.
I don’t think sparkplug age is a factor as long as it is gapped properly, isn’t damaged and has less than 5000 ohms resistance. They say dropping a spark plug renders it scrap. I disagree. If the drop damaged it, then it’s scrap. I have seen spark plug insulators crack from improper cleaning and gaping procedures and still pass a ‘bomb’ test. That’s why I inspect every one no matter how new with a magnifier and dye.
A mag “getting loose” usually turns it in the mag’s rotation, retarding timing. Retarding results in lower CHT, higher EGT and lost power.
But if the mags had been worked on recently and timed wrong, a ‘maintenance induced failure’ could definitely cause high CHTs. A high CHT plus something glowing in the cylinder can cause pre-ignition.
Once cracked a spark plug is like a time bomb, it isn’t if, it’s when it causes problems. In operation the spark plug tips have to get hot enough to scavenge the lead deposits, but not so hot that they glow. The purpose of the ceramic is to sink the heat away from the tip. Damage it and you have a pre-ignition source. If those shards of ceramic fall out they can imbed themselves in the piston and act as glow plugs, and the debris isn’t doing the rest of the engine any good.
Again guessing since I don’t know your mag and engine, when fuel quality declined to 100LL there were ADs to set timing on some engines at less than the data plate setting, eg, 22 BTDC rather than the original 25 BTDC. That was done to prevent overheating. An overheat, plus a damaged spark plug can cause pre-ignition. The Tetra-ethyl part of Tetra-ethyl lead requires a certain temp to scavange the lead out of the engine, so lower CHT can lead to lead fouling.
The difference between detonation and pre-ignition is explained here: https://www.savvyanalysis.com/articles/detonation-and-pre-ignition
I probably would have done as you the first few flights, try to get it hot and burn off deposits, and a 100 rpm drop is Ok for most engines. But as the RPM drop increased and the frequency increased there does need to be a point where you insist on some maintenance. You comment that a bad mag check occurred to you and two others, then to you three more times, including the incident flight so yeah, you had established a trend and agree it would have been appropriate to check the plugs, wires and mag timing.
You also comment that the ‘good’ mag only lost 25 rpm. That isn’t good either, indicating that the mag timing may have been advanced more that it should and could have been the one that caused the piston damage. But that’s speculation on my part, since I don’t know the engine.
With very little to go on, it sounds like the engine went into pre-ignition, where a hot glowing thing starts ignition before the spark plug fires. Paradoxically, its often a tang of broken spark plug insulator that does the glowing. Typically when that occurs CHT will skyrocket up well past 550F in a minute or so (Mike Busch has an engine monitor data dump that shows exactly that here http://www.mydigitalpublication.com/publication/?i=24591&p=24. Unfortunately, lacking a multi-probe engine monitor, there’s no way to know when it happens, although the ever increasing RPM drop during the mag check may have been a clue. Of more interest would have been an opinion of how rough the engine ran on the bad mag. A plug that isn’t firing at all is a huge clue.
This story is an argument in favor of installing an engine monitor that alarms when CHTs go ballistic. However, if you don’t react appropriately the damage will continue. That’s where proper training from http://www.advancedpilot.com would have taught you how to save the engine.
Mike busch on leaning http://www.eaavideo.org/video.aspx?v=1678859198001
Mike on mags: http://www.eaavideo.org/video.aspx?v=1367499350001
and more mags: http://blog.savvymx.com/2010/03/mag-check.html
If yo look at the window below the EAA webinar you’ll see a list of webinar presentations. there are 20 pages, and they are all free to members and non-members. Good engine management lessons well worth the cost of admission.
As a closing thought, a pre-takeoff mag check only reveals if the mags are still attached to the engine, but an in-flight mag check reveals weaknesses in the whole ignition system. As long as you don’t select ‘off’ it won’t hurt a thing. (If you do select ‘off,’ set mixture to cutoff, turn the mags on and reset mixture to prevent afterfire and possible damage to the exhaust system).
But on the ground, do a mag check at idle, including the off position to confirm both mags ground properly. If all is well, with the mag key in hand, turn the prop to gauge relative resistance of each cylinder. Do it again on pre-flight if you trust that nobody messed with the mag P-leads. It reveals leaky rings, burned valves and cracked heads. Worth knowing before flying the beast.
I’m a long-time aerobatic competitor and have cooked more than one engine. It’s not a badge of honor, but instead a learning opportunity.
Aerobatic engines are exposed to extremes and therefore have shorter life expectancy. I have experienced varying ignition failures, fuel injection issues and plug failures which all masked themselves as other problems (igntion problem which appeared to be a fuel issue, etc.).
A previous poster mentioned a plug ceramic developing a crack that causes pre-igntion. In my current airframe this occurred shortly after I took delivery. This engine is using the massive electrode ‘BY’ style plugs which reduce the potential for lead fouling. On a short cross-country hop I had leaned for best efficiency. Due to the higher compression my EGT numbers run higher than an ‘off-the-shelf’ engine. Suddently the engine stumbled, made a loud ‘POP’ and began shaking like a hardware store paint mixer. I enrichened mixture and the engine smoothed out after a moment or two. This prompted me to turn back and abort the flight. I got over the airport and recreated the failure mode, watching my engine analyzer (six channel EGT/CHT, plus all the other metrics). The analyzer showed fuel pressure and fuel flow fluctuations as the mixture leaned towards peak. Approximately 1gph rich of peak the oscillation would occur.
After leak checking the fuel system, cleaning injectors, fiddling with the fuel pumps, etc.; it was decided to try and fly again. This time the rough operation was allowed to persist and a clue was made – cylinder #3 got very hot in a short amount of time. This prompted a closer inspection of things on cylinder 3 – and an attempt to move the problem. In moving hardware (attempting to move a problem and fearing a cracked cylinder) – the spark plugs were pulled. Found the ceramic was cracked badly allowing the center conductor to become a glow-plug. This causes pre-ignition.
I’ve had a separate instance on a pitts that I ferried for a customer. it just wouldn’t run smoothly on one of the mags. We replaced plugs, replaced the mag harness and even checked timing while we were at it. Put things together and made another runup on the ground. Same problem. Found a ‘tower’ that was cracked; the distributor body on the back of the magneto had a hairline crack which was allowing a shorter path to ground (magneto case).
All airplanes break. If things aren’t right – don’t fly.
Forgot one – I’ve also seen exhaust valves go. One particular engine had been run full-rich it’s entire life (Lycoming O-360).
The valve stem was loaded with lead deposits which abraded away the valve guide. This allowed the valve enough play where it wouldn’t seat properly. Eventually the face of the valve broke off completely, punching through the piston. It split a cylinder wide open, penetrated a piston. Threw a ton of debris into the case. Yikes. 400 hours on the engine and it required a major overhaul.
Take care of your engines, friends!
Thanks for sharing that. I run an O-360 with the extra rich carb to deal with high CHTs in climb, so I worry about exhaust valve health. Lyc has a ‘mandatory’ valve wobble Sservice bulletin we are to do every 400 hrs, looking for just enough but not too much exhaust valve stem freedom. Too little and they can stick. Too much and they can burn. There is another problem, which is a lack of oil to the rockerboxes on most parallel valve Lycs. Both are discussed here: http://egaa.home.mindspring.com/engine1.htm
Lycoming guidance on cylinder health beyond the wobble test is not as helpful as that from Continental motors. TCM SB03-3 intelligently addresses cylinder health and is worth a read by all owners. It has color pictures so pilots might find it useful too. http://www.avweb.com/news/savvyaviator/188758-1.html with a link to SB03-3. An EAA webinar on cylinders is here: http://www.eaavideo.org/video.aspx?v=1204537102001 . Continental requires a borescope inspection before pulling a jug, Lyc is silent on the issue.
A worn valve guide allows what you describe. Proper valve cooling relies on even contact with the seat all the way around. If the valve wobbles there is uneven cooling and it can warp. If it warps it burns, leading to hoop stress and loss of a piece, or it breaks at the stem-to-valve head weld. This is why Lyc has a valve wobble check every 400 hrs, but few part 91 operators do it.
A bit of engineering: Almost all Lycs have hollow exhaust valve stems welded to the valve head, and the hollow stem is filled with sodium that melts and conducts heat into the guide. Lyc also uses exhaust valve guides that extend into the port where it bell mouths when hot. Improper leaning builds deposits inside the guide, so when the engine cools off it gets a death grip on the valve stem. At the next cold start the valve sticks. The cam pries it open (Or bends a pushrod and ruptures the pushrod tube so the oil can leak out), the engine runs rough for a minute or two until the head warms up and the valve snaps shut and runs smooth. This is classic Lyc ‘morning sickness.’ problem solved, right?
According to John Schwanner’s diagnostics on http://www.sacskyranch.com/acatalog/More_Info.html the cam and follower are ‘nitrided, which is a very hard but thin coating on the contact surfaces. When the cam pries a stuck valve open and it stays stuck open the valve train is ‘unloaded,’ and the parts hammer against one another. A little wear pad between the rocker arm and valve stem often falls out into the rocker box. That hammering cracks the nitrided cam lobe and follower so they spall with time. Corrosion plays a roll too.
The whole episode usually ‘heals’ in a few minutes as things warm up, the valve snaps shut, the hydraulic units take up the slack for the missing pad and all appears to be well, but the theory is that over time the damaged nitride coatings on the cam lifter or follower spall and destroy each other with an insidious loss of power. If it is an intake cam lobe there is double trouble because opposing Lyc cylinders share intake valve lobes, so both cylinders lose power. Most owners are unaware of it until teardown. A new cam and not cylinders is why overhauled engines get raves for more power (and fuel consumption) vs the ‘tired’ engine.
There is no Lyc or FAA required test for a worn cam. Magnetic bits in the oil filter pleats are a clue, or someone finds that little spacer in the rocker box. A dial indicator on the rockers will tell the story but few check the cam because most blame transient rough running events on the mags, plugs or induction ice.
I had a cylinder crack at the start of what was supposed to be a long cross country (coast-to-coast) and other than an occasional “skip” it didn’t show anything on the mag check or in flight until it nearly died on the spot.
After a similar diversion and landing as described here (to the nearest airport, not where I started), on the ground a run-up seemed like it was a mag problem – so bad that it would die on the bad side in the mag check. It wasn’t the mag, though, it was the cylinder, and the spark plug had lost the ceramic as well – hard to truly say the cause and the effect, but it took a full cylinder replacement.
Once it went bad, if we had been on the ground still the mag check would have caught it, even though it wasn’t a bad mag. Sometimes a good safety check catches more than you think it might.
There is something in your story that jumps out at me: The two hundred rpm drop that led you to stop flying the plane and start poking around was after the fourth take off and after x amount of flight time. I’m wondering if the problem had been addressed after at least the second instance, if not by the third time (which led to an additional hundred rpm drop at run up) that you might have saved that piston? Many of us have had the right mag fail the run up test, done as you and tached it up with a lean mixture to get rid of the lead and then had a normal second run up and flight. Maybe bad/old fuel started the problem or maybe a tired spark plug. At any rate, I am sorry you had to experience this and I thank you for sharing your experience. rog