Saturday, October 24, 2009


We're getting ready to fly to a strip up by Tungsten, on the NWT-Yukon border. The temperature is about -8, but it was much colder overnight. There is frost on the airplane, and the notable thing about it is the size of the ice crystals. Normally a layer of frost looks pretty much like a layer of white fuzz, but this is, even from several metres away, clearly crystalline. I try to remember back to chemistry class for factors influencing the size of crystals. Purity of the solution comes to mind. Is the pure air and water up here causing larger ice crystals? More likely it's a coincidental ideal combination of humidity and temperature in still air. Critical surfaces, plus the nose so we don't get a shower of ice on the windscreen at rotation, cleared, we start up. We've had the engines tented (that means we wrapped and clipped custom-fitted blankets tightly around them) and block heaters plugged in, so the engines start without protest.

We taxi away from parking for the run-up. My new co-worker is flying and I'm only there because the insurance company wants him to have more time on type before he is on his own. He's perfectly competent, however, so I'm just sitting in the right seat staring out the window. He flew out of here yesterday with my chief pilot, so he knows more about the local area than I do.

Suddenly there's a loud BANG on the other side of the fuselage. I swivel my head around to the passenger cabin to see who is kicking the hell out of my airplane, but the mission specialist says, "What was that?!" Turns out that that was a magneto check. I stupidly tell an experienced pilot "you're not supposed to turn them both off at once!" and he assures me that he didn't. It does it again, the left engine emits a huge backfire with the left magneto selected off. "Okay, don't do that again," I say unnecessarily. It's probably one of my worst traits, telling people unnecessary things. And I hate it when they do it to me.

I really don't think this is caused by fouled spark plugs, but it's something maintenance will ask, anyway, so we attempt to clear the problem by leaning that engine out at moderate power, and then reducing power, this time to 1000 RPM and trying again. "BANG!" Rats.

The radio operator calls us from inside. "Is that you making that noise." Gah, you know your airplane is backfiring badly when it's disturbing the guy in the tower.

The engine runs fine with both magnetos selected on, but we have a no go item. It's remarkable how much redundancy there is in the system. It's quite possible that whatever has happened to the left engine right magneto happened yesterday, while new guy was over top of an endless range of 8000' peaks. Each cylinder has two spark plugs, fired by separate magnetos. And if you get right down to it, the engine will run, albeit badly, with one cylinder not firing at all. Plus there is a whole 'nother engine on the other wing, which is capable of getting the airplane home. But all that is designed for things that go wrong after we leave. Before we leave, they all have to be working. We have a no-go item.

It might be a dead magneto, or it could be a broken connection somewhere, a frayed p-lead making contact with ground, mimicking the condition when the mag is turned off. Just before shutdown with the power pulled right to idle, we do one last mag check. No backfire, but the left engine dies when the left engine left mag is selected off. I'm confident that this is not just a spark plug.

Hangar guy is there, still marshalling float planes, boats and motorhomes into the immense WWII hangar for winter storage. We know he's an AME so I go by and ask if he can help. He says he is too busy, and doesn't have any parts anyway. I understand, and wheedle a bit to get him to agree to look at it long enough to give us a diagnosis. There are magneto-related issues that can be fixed in barely more than the time it takes to open and close a cowling. He agrees to look at it. Meanwhile my coworkers are standing in the bed of a pickup truck waving their cellphones around, trying to get service to report on what's going on.

We taxi the airplane over, the cowling comes off and a multi-tester goes on. He quickly confirms that the magneto in question no longer works. He says it's common in the fall when water gets in. We flew yesterday and the engine hasn't been allowed to become wet or reach freezing temperatures since, so his reason is probably wrong, but we trust his diagnosis. I ask him if we can pay for the use of his hangar so our AME can do the work. He says yes, we can work that out. We drive into town, where our chief pilot, having received our texts, is already working on getting an AME in to help.

I go grocery shopping, and then before heading back go to have another look at the Signpost Forest. My hometown must be here somewhere. My cellphone bleeps and it's a text from my chief pilot. "Where are you right now?"

I text back, "Sign post forest."

Return text: "I have something to tell you. I'll be there in a few minutes."

I head over to the entrance to the forest -- there are really so many signs that you could go quite a while without seeing someone who was in there with you -- and wait. After the time it would take to walk from somewhere close by, like the grocery store, my imagination starts to wander.

When the chief pilot needs to talk to you in person, that's usually bad. Weary of the usual horror stories of getting in trouble or getting fired, my overactive imagination also manages to summon scenarios where something has happened to the owner, or to someone I know. Before I can come up with a scenario whereby my chief pilot has advance notice of the impending destruction of the planet, a car drives up and it turns out everything is fine, it was just a little complicated to explain via text or a poor cellphone connection.

My company is flying an AME to Whitehorse, with his toolbox and a magneto. His flight gets to Whitehorse at one a.m., so chief pilot is driving out to pick him up. Company is concerned about people driving around alone on the Alaska highway in the middle of the night, so new guy is going too. Chief pilot will fly home and new guy and will drive AME back, arriving tomorrow morning around seven. AME can nap during five hour drive (yes, the nearest airport with scheduled service is five hours away on the Alaska highway) and then fix the magneto before going to bed. I know that sounds pretty horrible, and it is. Get a call in the morning to get on a plane, fly all day and half the night, then ride in a car for the other half of the night and be expected to work before you get to check into a hotel. AMEs don't have legal duty days and they are horribly abused like this all the time. I never begrudge one a scheduled coffee break.

They'll be back tomorrow, but I miss my co-workers already.

On another topic, but in keeping with the title of this blog entry, readers Sarah and Tyler those who read the blog comments to a fascinating video of a pilot's last flight. One of the remarkable things about the video is that it was made in 1984, on video tape, with a video camera that was melted into unrecognizability in the post-crash fire, and then lay in the elements for a full three years before it was discovered. Deputy sheriff Dale Wood reconstructed the film.

The film is linked from this page. Scroll down to "Cessna L-19 Mountain Crash" to download it. It's a large file--took me an hour on hotel wireless--so if you don't have the bandwidth you can watch the short version linked below it. I watched it knowing only that it had been recovered from a fatal crash, but not the cause, so it was interesting watching and speculating. I'll give you a chance to do that before you read my commentary.

At first I assumed it was going to be a continued VFR into IMC story, because that was the context it was posted in, but the clouds seemed few and above the hilltops. They were flying fairly low, but that was for operational reasons. They were inventorying beetle-damaged timber. As they flew closer to treetops and terrain than I like to I winced a few times and wondered if they were going to clip a wingtip and crash that way. Perhaps transmission wires or a cell tower. But it isn't a built up area at all. The scenery is enough to be distracted by, even conveyed by aged videotape. I wondered if perhaps it would be an engine failure with nowhere suitable to glide to. Throughout the video, the terrain is rising subtly and there are some prominent rocky peaks ahead. The visibility is too good to fly into them unless this is a pilot incapacitation scenario. And then I guess it: they're going to try to outclimb terrain and strike the rocks. But no, I'm wrong; the pilot realizes that he can't outclimb the rocks ahead and starts to turn away. Good. But his bank angle is startling. He ignores the first bleat of the stall horn, and the second. The view snaps down to tree trunks and spins in.

Damnit! Don't you think that if at the first sound of the stall horn he had checked the nose forward, levelled the wings and applied full power he could have maintained altitude over those trees, perhaps with a gentle turn? He might not have had much power in reserve, but surely he wasn't flying along that close to terrain with the throttle wide open already. Or perhaps he had applied full throttle already and made the turn on discovering his negligible climb rate. Seems quite late to have started the climb, though. Even at sea level I wouldn't have been expecting a climb rate that would get me comfortably over those peaks from where he started the turn. But his best angle of climb should have been able to keep him over the tops of the trees which I see ahead of him on the second bleat of the horn. Why didn't he level the wings? Shoulda coulda woulda. In other jobs you make a bad mistake and your boss yells at you.


Chad said...

I watched the video too, butI was surprised to read your description of a "startling bank angle". I thought it was a rather shallow angle. It looked to me like he was probably at a fairly high altitude, which meant his indicated airspeed even at full power was still close to stall. The shallow bank he did try was enough to raise Vs just enough to cause a stall.

But now from what you're saying I'm second guessing myself about the bank angle... I'd have to go watch it again.

Aviatrix said...

The nature of the work I do may bias me. I was startled by the bank angle.

Also, the terrain you're looking at out the window slopes up to the left, so that wings level looks like a left bank.

Matthew said...

I'm with you on not having applied full power until it was too late. It seemed to me like they were going rather slowly, which as you said, was probably for operational reasons. Heavy plane, thin air (I assume ground level was a few thousand above sea level), flying slow. Personally, it seemed he would have been ok if he turned left instead of right.

Sarah said...

Ah, airplane engines. Gotta love 'em, the 1930's tractor technology still works. Usually. And when it fails, you know what's wrong. No gas, no air, no spark. Well, not that there isn't room for some mystery.

And I'm glad your post didn't end with the chief pilot finding you in the signpost forest. Talk about your forbidding foreshadowed "bad news" sopt. Fortunately it was not terrible news, just "complicated".

On the movie. I was reminded of it by the candid youtubery of the foolish Bonanza pilot flying into IMC. But no, it is an accident that fascinated me by it's seeming inevitability once some poor choices were made. I hope I can learn from it. So sad, the last 6 minutes of their lives... a pretty lake. A cabin. Can a L19 outclimb a mountain with a 8200' take-off altitude? No. The pilot and his helpless passenger's fate was sealed when he tried to keep climbing, probably at full power. At the very end, it seems he realized that's all he can climb, but turns the wrong way, into the slope - or maybe there was no right way at that point. The stall horn bleats, he does lower the nose, but tries to continue the turn - what choice does he have at that point? Another bleat. And spin. And bang. Into eternity.

Michael5000 said...

If your chief pilot DOES bring news of impending doom, you'll share here, right?

And: that was one spooky link. Just in time for Halloween. Be careful up there, hear?

SwL_Wildcat said...

Hola Aviatrix;

I am not a neigh sayer, and I am sure your AME will probably check it, but when the engine backfires like that it could cause a lot of other undetectable damage. I have seen students blow carbs of 172's doing mag checks, and split mufflers, and holes in mufflers. If your AME has not had much sleep, it could be missed. even if you have been flying after the incident, have the AME check it over after he gets a bit of shut eye.

Verification word: sedingin

Hey Buddy; How did you get sedingin the side of yer truck?

A Squared said...

FWIW, WIldcat, that's not a Backfire, a backfire is when flame travels back through the induction and out the intake. An explosion of unburned fuel in the exhaust is correctly called an afterfire, because it is happening after the combustion chamber, there's nothing "back" about it.

Anonymous said...

One thing being a glider pilot teaches you; "Always turn AWAY from the terrain when it is above you" 'nough said.

verification word mityp
He sounded mityp.

SwL_Wildcat said...

Thanks A Squared. I guess I could have gone into a lot of detail, but most people would not know the difference between a backfire and an after fire. An after fire is when the spark plug is not firing, but the engine is rotating, and still drawing fuel from the carb, through the cylinders, and out the exhaust port. The fuel then catches fire, or explodes inside the exhaust system. This causes extreme pressures in the exhaust system, which in turn can do some serious damage. Many a student has done a mag check, turned the ignition to "OFF" for a couple seconds, and then turned it back on. The engine immediately will catch, and burn the fuel vapor from all the cylinders that were previously not firing, and instant "Bang". Then they go "Whoops" and taxi to the runway for takeoff, oblivious to what actually just happened. With a split exhaust and full power on take off you could have a real serious fire hazard. I also witnessed the after effects of a C-172 carburetor damaged in the same manner. The force of the explosion in the exhaust was large enough to push the exhaust valve(s) off their seats. The flame then travels into the cylinder(s). The way the engine is designed 1 intake valve is almost always open. The flame then travels through the induction system. The after fire has now created a backfire as well. This does not always happen, but you should be aware that it can and does happen. That is why I would highly recommend the AME have a quick check. You might have a second engine to get you home, but it’s not going to be a fun ride with the first engine nacelle is on fire.

Sarah said...

At the risk of betraying my ignorance of internal combustion engines, swl_wildcat &/or A squared ..

I've never heard the "bang" you speak of on a mag check. R/L/both, never "off". But I wonder, what do you think of doing the "p lead" magneto check at shutdown by momentarily turning the ignition off? Is there a good, safe way to do this without the risk?

SwL_Wildcat said...

Hello Sarah;

In my opinion it's never a good idea to just shut the mag off, and then return it to any other position while the engine is still turning. If it's for a brief 1/2 second it's probably not going to hurt anything, and if it’s written in your POH to do it that way then you should follow manufactures recommendations. I don't know if it's covered in your POH, but what I would do if I went to off for more than an instant is leave the key in the off position, and pull the mixture to off. Let it sit for 5 minutes then restart. This will give the fuel in the exhaust time to evaporate and dissipate. There are cases where you might want the engine to stop ASAP, IE a dog runs out of a near by hanger or something, and you don't want to wait for the engine to burn off the remaining fuel in the carb before it stops, then I would shut the mags off immediately and leave them off. Please don’t take my advice over anything written in the POH though, as these are just my opinions.

Anonymous said...

How come your airplane has problems all the time?
Is it because of the very strict air law requirements? I wouldn't want that hassle everytime I wanted to drive my car somewhere.

Anonymous said...

Magnetos? Carburators? Leaning? 100LL? *Sigh*

Volkswagen now gets 199kw (270hp)@6000rpm, 350Nm@2500-5000rpm out of a 2l Engine.

Sure, there are various constraints to be considered for aviation grade engines, but still...

I think time has come for fuel economy and emission (NOx, HC, ...) restrictions to be introduced for aviation engines. Otherwise we'll still see airplanes with this jurassic technology up until 2050.

Jim said...

Anon 15:02 - Airplanes have problems all the time for a combination of factors, including:

1. They are often old. It is not unusual to fly planes that are 30 years old, and even though a lof of the parts are much younger than 30 years old, there are a lot of parts,

2. They are inspected daily. I hop in my car and go driving. Before I go flying I check every light bulb. Constant inspection means you are more likely to find something wrong, unlike driving around with a burned out license plate light for 8 months,

3. The standards are high in flying. When the engine in the car is a bit rough one thinks "I should get that looked at". When it happens in an airplane the pilot will ensure there is a field nearby that can be reached in a glide, and when you make it back to the airport the plane is grounded until it is fixed. OTOH, I'll drive the car to Canadian Tire since the worst that can happen is I'll be at the side of the road and call for a tow.

4. The duty cycle on aircraft engines is totally different. Not unusual to having a commercial aircraft flying 16+ hours a day. I'll discuss engine differences in the next posting, in response to Anon 15:16

Jim said...

Anon 15:02 and Anon 15:16:

I am in total agreement that aircraft engines need to become much more modern, and getting away from leaded fuel is going to continue to come under extreme pressure. AvGas will continue to be different than MoGas, because MoGas formulation changes constantly throughout the year - but the lead will eventually need to be removed. At minimum, new engines will need to be able to run on no-lead, and the fuel systems need to be 10% gasahol-friendly.

Fuel-injected engines are common, however, carbs have lasting presence because they are highly reliable devices (they don't need an electrical supply, for example).

That said, the performance characteristics of aircraft engines are very different. They continually run at 50-75% of power (automobile engines usually run at about 15-20% when floating down the highway). They run for a long time - hours - while most auto engines are used for shorter durations. They are air-cooled and run at much higher temperatures (auto engines run at about 220F all the time).

Most development has been in turbine engines (jet aircraft), because that's where the money is and that's where the huge amounts of fuel are burned.

I would love to see a company that has engines as a core competency (e.g. Honda) build a transformational engine. The challenges to getting it adopted (certification, persuading manufacturers, etc) might not make it a viable business choice.

Verification: sechels. She sells sechels by the sea shore

A Squared said...

Jim, A couple of comments on your post:

and the fuel systems need to be 10% gasahol-friendly.

If we accept your premise that avgas will continue to be a separate commodity (which is reasonable), then there is no reason other than flawed politics to be alcohol compliant.

Fuel-injected engines are common, however, carbs have lasting presence because they are highly reliable devices (they don't need an electrical supply, for example).

Aviation fuel injection systems run fine without electricity. I don't know much about Rotax and similar engines, but the vast majority of certificated fuel injected aircraft engines use a constant flow injection system, which is nothing more than metered fuel under pressure flowing constantly into the intake manifold, through orifices located at the intake port. No electricity required.

I would love to see a company that has engines as a core competency (e.g. Honda) build a transformational engine.

Maybe we could get Porsche interested.

Sarah said...

I hope ethanol fuel has no future in avgas. There are signs that 94UL may work for O-360's and the like .. don't know what higher compression engines are going to need.

A Squared, you being funny? Porche was bitten once already with Mooney and ventures into aviation engines.

Tyler said...

Is this the specific airplane that had the tachometers changed to digital? If so, do you know how the tachometer handled the magneto failure in flight?

I am assuming that the Tachometer monitors both leads, in which case it seems like a reasonable thing for the instrument to light up and say "hey, one lead is doing something different from the other!" when there is a failure. Is this what happens, and if so, would it be grounds for landing ahead of schedule?

A Squared said...


I have a deadpan sense of humor that doesn't translate well to the written word

Sarah said...

I have a deadpan sense of humor that doesn't translate well to the written word

Oh, it did, it did. I thought it was very funny.

Aviatrix said...

The tach change was fleetwide, so yes, this one has the new tachs. They were surprisingly quick to get used to. During the run up we can put the tack in a mode where it displays the drop after one is selected off, just so we don't have to do the math to know that 1500 - 1570 = a 30 rpm drop. There is no special indication that one magneto is off, just that drop of about 30 rpm at runup power. It would be more at cruise power, but we make small power changes throughout the flight, so that could go missed.

I wasn't on the previous flight, so I can't comment on that.

A Squared said...

It (RPM Drop) would be more at cruise power,

Actually, it wouldn't drop at all (except very briefly) in flight. The prop would adjust to maintain the same RPM.

Aviatrix said...

Duh, thanks A^2. Really thought that one through, didn't I?