Ignition Switch Fatality

[DanH]

I don't typically post accident links, but this sort of component failure could easily be present in many RV's. The pilot was doing a preflight with the ignition key in his pocket...but he still had a hot mag.

I'm not a fan of rotary ignition switches, and always recommend toggles. However, if it's what you want, fine. Just be sure to check it on a regular basis, not just for reliable p-lead grounding, but for key retention in all positions other than off.

http://www.kathrynsreport.com/2019/07/propeller-contact-with-person-cessna.html

BTW, I was once standing at the nose of a Baby Ace with two other pilots and an FAA Master Mechanic, generally shooting the breeze. Somebody moved the prop just a fuzz, the impulse clicked, and it started. There were guys on both sides of the prop disk. It missed all of us by inches.

 

[rocketbob]

There is an old AD on Bendix key switches that I routinely check when doing annuals (76-07-12). Occasionally find one that fails. One has to run the engine per the AD and check the switch but leaving the magneto tester connected to the mags and beeping will also reveal the condition of the switch.

 

[MacCool]

A good reminder for owners of those aircraft with at least one impulse magneto for ignition. My key can't be removed from the switch unless it's off. I'd never been aware that there were ignition switches where the key could be removed....never seen that (AFAIK).

I have dual pmags so in order for my engine to accidentally start, the rotary switch would have to be on for one or both mags as well as the Master and one or both pmag power switches. My preflight glance in the cockpit always includes a quick review of panel switch positions.

 

[Canadian_JOY]

Thank you, DanH for posting this. I'm in the same camp as you for exactly the same reason. Rotary key switches should be avoided by those of us who have the flexibility to remove them from out instrument panel / electrical system designs. Toggles or toggles, those are the two options!

 

[Mel]

A good reminder for owners of those aircraft with at least one impulse magneto for ignition. My key can't be removed from the switch unless it's off. I'd never been aware that there were ignition switches where the key could be removed....never seen that (AFAIK).
I have dual pmags so in order for my engine to accidentally start, the rotary switch would have to be on for one or both mags as well as the Master and one or both pmag power switches. My preflight glance in the cockpit always includes a quick review of panel switch positions.

I HAVE seen old switches that were worn enough that the key WOULD come out when not securely in the OFF position.

 

[Paul 5r4]

confused

I am a little confused. What/how does the key being able to come out lead to a situation where the engine can fire? I am "assuming" the fault is this... the key can come out and if it's not actually in the off position, we are fooled into thinking it's off but actually one or both mags are hot??? If my "assumption" is correct then the fault is just that, (the key coming out in R L Both), rather than within the workings of the switch itself?

Am I understanding correctly?

 

[MacCool]

I HAVE seen old switches that were worn enough that the key WOULD come out when not securely in the OFF position.

I'm sure that could happen, particularly given the age of the GA fleet, I've just never seen it. I did not say that the problem didn't exist

 

[JonJay]

I would venture a guess that the rotary ignition switch represents a huge percentage of ignition switch types, huge compared to other types. So, you will have more instances of failures, way more.
This doesn’t mean the issues presented here aren’t important and don’t need heeding, but let’s not starting yelling fire in the theater. Test them as Dan points out.

 

[PilotjohnS]

Single Point failure

The common rotary ignition switches are a single point fault.

I had a brand new one fail 10 hours into my Phase 1. The engine would cut out at RPMS higher than 1800 or so. Apparently the BOTH position was shorting.

All these rotary ignition switches should have dual wafers, so that a single fault in the BOTH position does short out both mags, like what happened to me.

And I agree we should test the switches to make sure the key only comes out in the off position

 

[mburch]

I am a little confused. What/how does the key being able to come out lead to a situation where the engine can fire? I am "assuming" the fault is this... the key can come out and if it's not actually in the off position, we are fooled into thinking it's off but actually one or both mags are hot??? If my "assumption" is correct then the fault is just that, (the key coming out in R L Both), rather than within the workings of the switch itself?

From the link that Dan posted:

Postaccident examination of the 42-year-old ignition switch revealed that it appeared to be in the off position when observed visually from the pilot's seat, but its actual selected position was more toward the right magneto position. The switch and key were determined to be slightly misaligned with the instrument panel placard markings. On- and off-airframe operational testing of the ignition switch in the as-found position revealed the right magneto was hot, or not grounded.

[...]

The key could be removed from the switch in any of the five positions due to the wear of the switch's internal components, contrary to its intended function that would retain the key in any position except the off position.

 

[gasman]

In my PA28-151, the key switch would go just past off (to the left) and the key would be removed as expected. But the ignition remained HOT!!

So even though you removed the key in the off position, the switch was shorted in the off position. Easy to test, just try to go beyond off (to the left) and see if the motor keeps running.

Very little value in a key switch. In safety or theft.

 

[Bavafa]

There are some good lessons and reminders in here but I don't think those are toggle switch vs. rotary switch. Any device can fail, we just need to be mindful of that and the danger that their failure poses while sticking to the procedure.

Shutting off the engine with the mixture control is one of those important procedures.

 

[Blw2]

This is a topic that makes me cringe...having been renting for all my flying....less familiarity, less control over the maintenance, etc...

A long time ago I got into the habit of checking the mag grounding by momentarily switching the key to "off" before pulling the mixture for shutdown, just to see if the engine stops....but re-engaging it before the prop stops. I don't remember where I picked that idea up...something I read somewhere or maybe an instructor along the way. Regardless, I always felt it wasn't a great test because I'm not actually seeing if the engine dies.

Then one time in a rental, the engine sort of stumbled and came back to life and stumbled and back...but never actually shut down...then I learned somewhere that this sort of test doesn't work for all configurations. I hate to admit now, that I don't recall why that was...something to do with some switch designs...or was it the PMag that MacCool mentioned earlier? Regardless, hard to know all these things about a rental aircraft.... and so I always cringe when I see someone moving a prop on preflight....

 

[MacCool]

Then one time in a rental, the engine sort of stumbled and came back to life and stumbled and back...but never actually shut down...then I learned somewhere that this sort of test doesn't work for all configurations. I hate to admit now, that I don't recall why that was...something to do with some switch designs...or was it the PMag that MacCool mentioned earlier? Regardless, hard to know all these things about a rental aircraft.... and so I always cringe when I see someone moving a prop on preflight....

In my Pmag setup (not chosen by me) the engine can only "accidentally" start from a propeller bump if one or both P-leads are ungrounded at the key switch, AND the batteries have power, AND one or both the Masters are on, AND one or both Pmag switches are on. Obviously DanH was addressing conventional magnetos with an impulse coil, but yeah...I think one needs a to know the specific configuration off the airplane they're flying at the time. A defective key switch could be a major problem in some configurations.

 

[vic syracuse]

Never trust and always verify

This is a good thread that needs to be resurrected from time to time. I've had three engines with supposedly-grounded ignition systems start up on me over the years. I'm amazed I'm still here. Most of the electronic ignition systems are hard to fire just by moving the prop a little bit, some due to needing 12 Volts and others needing a minimum rotational speed. All are different. I still don't trust them.
But magnetos with impulse couplers..... NEVER trust them unless you were the one to check it at shutdown with the key or switch, not just the mixture. There are AD's on some of these switches, as mentioned earlier.

Many times I will take a jumper and short the mag to the engine when I am rotating an engine. That's how little I trust the wiring/keys/switches in magneto-equipped aircraft.

Vic

 

[danny]

AD 76-07-12 Bendex Ignition Switches

If you read the AD it tells you to test the switch a every annual. From a few of the posts I would think it a good idea. I've never seen a bad one but that doesn't stop me from testing as part of every inspection.
danny

 

[Mel]

There are some good lessons and reminders in here but I don't think those are toggle switch vs. rotary switch. Any device can fail, we just need to be mindful of that and the danger that their failure poses while sticking to the procedure.
Shutting off the engine with the mixture control is one of those important procedures.

The main advantage of toggle switches is that you test each one when doing a mag check. Turn the left switch off and the left mag stops. Same with the right switch. Conversely, when you do a mag check with a "key" switch you are not checking the "OFF" position itself. this is where many of these switches fail.

 

[Fred.Stucklen]

Ing Key Switch

I just tested mine. The key can not be removed if either L or R or BOTH is selected, and it Defiantly grounds both PMAGS in the OFF position... AND I added this procedure to my Condition inspection list so it will get done every year!

 

[Roadjunkie1]

Hot prop

I just tested mine. The key can not be removed if either L or R or BOTH is selected, and it Defiantly grounds both PMAGS in the OFF position... AND I added this procedure to my Condition inspection list so it will get done every year!

Recently tested mine. Thumbs UP! Key will NOT come out unless it is in the OFF position. Easy to test the switch while timing the mags. The if the switch is working, box doesn't change going through 25 degrees if the key is in the OFF position.

Being a mechanic for multiple decades (!), I am very aware of where I am standing in relation to the arc of any propeller, no matter what is going on. I will duck around the arc of the propeller just by habit. Never had one start or "pop" in all the years I have been working on the beasts! Still don't trust 'em...

 

[Dad's RV-10]

Is it safe/prudent/recommended to fully shutdown the engine periodically (every fifth flight, for example) by moving the key to the "Off" position to verify proper grounding? And then moving the mixture to cut-off after the prop has stopped?

Or is there a viable reason not to do this?

 

[PhatRV]

Is there any harm to the engine if I shut the engine down by turning the ignition key to OFF and then ,pulling the mixture lever to idle?

It seems like a sure way to confirm the engine is safed every time.

 

[Jetmart]

I check that my mags are grounded on every shut down by momentarily go to the off position then back on. Then I shut down using the mixture. (I have a couple of planes that I fly that the prop must be hand pulled through prior to starting.) What I haven't done regularly is to see if I can pull the key out before it gets to the off position.

The two checks are very different.

 

[Webb]

The old key in my Bonanza could be removed. I had a new key that was cut by the numbers instead of a copy can’t be removed except in the off position.

If your key is worn, replace it. A good locksmith can cut a fresh sharp key from the old worn key without copying the key.

 

[Blw2]

Is there any harm to the engine if I shut the engine down by turning the ignition key to OFF and then ,pulling the mixture lever to idle?

It seems like a sure way to confirm the engine is safed every time.

the problem as I understand it there, is that you aren't running the fuel out of the system. There might be enough fuel to run a few revolutions if you move the prop.... pulling the mixture to kill the engine starves the fuel so it theoretically can't run hot mags or no....

As I posted before, the habit I learned a long time ago was to switch the key to off checking it at every shut down...but then letting the engine re-kick...and then do the full shutdown with the mixture. I don't fully understand all the possible configurations though so not sure if this is sometimes or always a good idea....or not

 

[BobTurner]

My cautionary tale

Years ago, I was in a partnership 182. At annual, one cylinder had low compression with air coming out the exhaust. Our A&P recommended flying it hard for 30 minutes, then retesting. I did it. Knowing I was going to rotate the prop, I did a mag check just before shutdown (okay). I shutdown with mixture full aft, key off and removed, fuel valve off, wheels chocked (in hangar) just in case. I hand rotated the prop a number of times, no cylinders firing. I thought everything was fine. I then came up on the suspect cylinder, lean in a bit, pulled the prop thru. I thought I could hear air out the exhaust, but wasn’t sure. I repeated, leaned in a little more, and BANG! The cylinder had fired, the prop swung around 180 deg and hit me in the top of the head. I believe somehow residual gas in the intake manifold had gotten into the cylinder, likely very lean mixture. The next cylinder (120 deg later) did not fire, or I might be dead today. I was staggered but did not lose consciousness. There was blood everywhere. I grabbed a bunch of paper towels, made it to the rest room where I put my head under cold water, more towels, drove to a local clinic where they stitched up the cut in my scalp. Next day we de-cowled the engine, got our A&P out. He couldn’t see what had gone wrong. He said, ‘I can see the P leads are attached to the mags…..’ and as he spoke he touch the P lead, and it moved back. It had failed at the crimp connection, but was stiff enough to lean against the terminal, until vibration or a finger moved it off. I guess the moral is, you cannot be too careful around props.

 

[airguy]

the problem as I understand it there, is that you aren't running the fuel out of the system. There might be enough fuel to run a few revolutions if you move the prop.... pulling the mixture to kill the engine starves the fuel so it theoretically can't run hot mags or no....

As I posted before, the habit I learned a long time ago was to switch the key to off checking it at every shut down...but then letting the engine re-kick...and then do the full shutdown with the mixture. I don't fully understand all the possible configurations though so not sure if this is sometimes or always a good idea....or not

The trouble with a mixture ICO is that with an injected engine, the fuel will boil in the injectors after shutdown from heatsoak, dumping what is left into the intake manifold at the valve. At least one of those cylinders will have an intake valve open to allow fuel into the engine for the majority of the rotation of the prop, and from there it's only a hot-mag away from being able to fire for one or two power strokes before the engine dies again from true fuel starvation. That's more than enough to cause injury or death, as noted from the report in Post #1.

 

[DanH]

He said, ‘I can see the P leads are attached to the mags…..’ and as he spoke he touch the P lead, and it moved back. It had failed at the crimp connection, but was stiff enough to lean against the terminal, until vibration or a finger moved it off.

While on that subject...

This is how you install a p-lead on a Slick. Please, no wires flopping around, supported only by a terminal crimp. Use the ground screw as a hard point for an adel clamp.
.

 

[Paul 5r4]

I have to say, I've never thought of this. I am so glad to learn something so important regarding this life and death level of safety!!! I'll check mine next time at the plane. Also, I love Vic's idea of grounding out the mags with a jumper. I will be making this suggestion standard operating procedure for sure!

 

[BobTurner]

While on that subject...

This is how you install a p-lead on a Slick. Please, no wires flopping around, supported only by a terminal crimp. Use the ground screw as a hard point for an adel clamp.
.

In the case I cited above, the P lead was well supported, and was part of the problem. The support kept the conductor in contact with the mag, even after it was broken - until it moved 1 mm, and then was an open circuit. But it was not apparent in a visual inspection.

 

[Webb]

Thank you Dan H. Very timely for me and I will, as usual, follow your instructions to the letter.

I also will check my present bird out and update if needed.

 

[Flying Canuck]

A couple of weeks ago I had my plane out for display at a local event geared towards girls in aviation. It was open to all and attended by many. My plane was attended at all times by myself and my neighbor that the kids were out there, but that attendance was generally behind the left wing or at the helicopter I was next to.

I had a fair amount of people touch my prop as they walked by, not ideal but not a big problem. I had, however, 3 occasions where some very little kid came up to the prop and reached up and pulled down on the prop enough to rotate it 15 degrees. Positively terrifying, even knowing that my mags were not hot. There is no sense of danger around a prop in the general population. Next time I make something and hang it on both blades of the prop to scream, DON'T TOUCH THE PROP.

 

[newt]

the problem as I understand it there, is that you aren't running the fuel out of the system. There might be enough fuel to run a few revolutions if you move the prop.... pulling the mixture to kill the engine starves the fuel so it theoretically can't run hot mags or no....

On a fuel injected engine, that's plainly ineffective, because you hot-start it with the mixture in the cut-off position and only add fuel after it fires.

So it's self-evident that there's fuel in the system even if you've shut down by pulling mixture. You can even hear it squirting through the nozzles when it boils after shutdown.

As I posted before, the habit I learned a long time ago was to switch the key to off checking it at every shut down...but then letting the engine re-kick...and then do the full shutdown with the mixture. I don't fully understand all the possible configurations though so not sure if this is sometimes or always a good idea....or not

I was trained for this too.

1000 RPM, then look for mag drops on L and R, the quickly "burp" the key to the off position to make sure the engine stumbles.

Then at least you know the key is working and the P-leads are connected when you shut down.

Nobody else in your hangar knows it. But you know it, and that's gotta be worth something.

- mark

 

[rocketbob]

turning prop backwards

One thing I do when having to move the prop is to always turn it backwards. The impulse coupling cannot catch and wind up when rotating the prop in the opposite direction. Some say that on airplanes with gear-driven accessories (Bonanzas) that it can be hard on brushes and vanes but I've never found this to be the case.

 

[Caveman]

unexpected engine start

I had the line guys watch this video as part of their recurrent training.

https://www.youtube.com/watch?v=Ypb6fEgLNpY

Joe

 

[Taltruda]

My story

I had a Baby Lakes biplane with an O-235, no electrical system, and had to hand prop it across the country. I knew the key switch worked as I would shut down with the key routinely at short fuel stops to make starting easier. One morning, I began my routine of pulling the prop through to prime it before starting and it fired! Key was off, bendix mags have the internal safety that if the P-lead come out, it self grounds.. but what happened was the P-lead came loose but didn’t fall out yet, and it fired..

 

[cajunwings]

Hot mags

45 years working on aircraft and have seen numerous versions of hot ignition systems when they were supposed to be off. P lead problems are more common than switch problems. There is no way to be 1000% positive that your ignition system cannot fire a plug if all wires are connected and one pop from one cylinder can kill you. Rocket Bob is right, if you have to move the prop turn it backwards if possible. I always assume a cylinder could fire with any normal direction of rotation prop movement and position myself out of the arc. Good practice with key switches is at idle go to off position, firmly hold it there and wiggle the key while the engine is winding down and see if it fires.

Don Broussard
RV9 Rebuild in Progress
57 Pacer

 

[Christopher Murphy]

I HAVE seen old switches that were worn enough that the key WOULD come out when not securely in the OFF position.

I read this about the ignition switch and the key and I just realized that my key has not been removed from the switch since it was installed in my airplane 25 years ago..

If one decides to use toggle switches please install in a manor that prevents accidently bumping them to the engine off position..

 

[AlexPeterson]

Great thread, Dan - thanks.

I have the key switch, but will now also pull the two breakers for the EI (Lightspeed) whenever needing to move the prop. My previous strategy was to always look for the key on the glareshield as confirmation of dead ignitions. Now it will be key on glareshield + breakers pulled.

 

[DanH]

Great thread, Dan - thanks.

I have the key switch, but will now also pull the two breakers for the EI (Lightspeed) whenever needing to move the prop. My previous strategy was to always look for the key on the glareshield as confirmation of dead ignitions. Now it will be key on glareshield + breakers pulled.

The mention of Lightspeed brings a fine point to mind. Lightspeed, SDS, and a host of others trigger with Hall effect sensors. There is no requirement for a significant rotation rate, so if powered (defective switch or inattentive pilot) they can fire a cylinder with just a fuzz worth of prop movement.

Best I know, it can't happen with reluctor triggering, in particular the skipped tooth type. Electroair and EDIS-based ignitions are examples. It should be possible to turn them slowly while powered.

 

[AlexPeterson]

The mention of Lightspeed brings a fine point to mind. Lightspeed, SDS, and a host of others trigger with Hall effect sensors. There is no requirement for a significant rotation rate, so if powered (defective switch or inattentive pilot) they can fire a cylinder with just a fuzz worth of prop movement.

Best I know, it can't happen with reluctor triggering, in particular the skipped tooth type. Electroair and EDIS-based ignitions are examples. It should be possible to turn them slowly while powered.

I seem to recall the hall effect sensor has to see two magnets within a short period in order to fire. Jiggling it back and forth right at TDC will do it as it is part of the initial test procedure IIRC.

 

[rockwoodrv9]

Thanks for the info

After reading this I feel like I have been real lucky. I have the key switch set up. Just a few hours after I soloed in my plane I forgot to turn the key off. My Master was off, but the next time I went flying, the key was on. I have heard of a Hot Mag, but didnt really understand it. This thread has made me change my shutdown procedure.

I checked my key today and it will not come out in any position other than off. That key comes out and on the dash after each flight from now on. And, I will only turn the prop backwards if I need to adjust for my tow bar.

 

[rv8ch]

If you are like me and are in a shared hangar, you get to move other aircraft from time to time, and you may not have any access to the aircraft except to move it. This unfortunately can preclude even checking to see if the mags are off (cover over the cockpit).

For anyone that's hand-propped an aircraft, you know that sometimes the dog-gone thing won't start no matter what you do, and sometimes you just look at the prop and it starts up.

This thread is an excellent reminder to take care any time you are near a prop.

 

[Mconner7]

Assume the mag is always hot.

There is no such thing as an unloaded gun or a grounded magneto. Use the same caution with both as it’s not always the key that fails to ground a mag. An airplane propellor is a cobra ready to strike.

 

[Roadjunkie1]

Key removal....muscle memory...

This thread has made me change my shutdown procedure. I checked my key today and it will not come out in any position other than off. That key comes out and on the dash after each flight from now on.

As part of my shut-down procedure, one of the last things do is pull the key out of the switch with a purpose (not quite jerking it out, but nearly!) and, holding it in my hand, pointing it straight up as I look around the panel to see that everything else is off: a momentary pause to just look. (Oops: ANR switch is still in the "on" position.) It is a routine that, after all this time, has worked itself into the muscle memory of my shut-down procedure. Then I hang it on the altimeter adjustment knob. And don't talk to me until I am done with everything on that shut-down checklist. (ANR switch is actually on that checklist....)

 

[Mconner7]

As part of my shut-down procedure, one of the last things do is pull the key out of the switch with a purpose (not quite jerking it out, but nearly!) and, holding it in my hand, pointing it straight up as I look around the panel to see that everything else is off: a momentary pause to just look. (Oops: ANR switch is still in the "on" position.) It is a routine that, after all this time, has worked itself into the muscle memory of my shut-down procedure. Then I hang it on the altimeter adjustment knob. And don't talk to me until I am done with everything on that shut-down checklist. (ANR switch is actually on that checklist....)

Circa 1981, I was preheating an engine and with a student at the controls, I was pulling the prop through to prime the O-300. With no warning, the engine started and kept running. I looked at the student and he picked up the key from the glare shield where we always placed it when off. If I hadn’t used my training in hand propping and treated it as a hot mag, it would have been a bad scene.

Twice since then, a mag grounding check at shutdown showed that a P lead had failed and the motor ran on with the switch off. Please be careful!

 

[Webb]

I’m not concerned about an accidental start in my Bonanza when the engine is hot. The big continental is a bear to start when hot.

That being said, when the engine is cold, it fires after about a quarter turn. I don’t trust it and if I have to turn the prop, I do so with extra care. As in EXTRA care.

Regardless of hot or cold, I turn the prop with prejudice……just in case.

When refueling, I shut the engine down with the key switch so I get a grounding check. It makes the restart on the hot engine somewhat easier. Some but not a lot.

This is a great thread. It reminds us that grounded props are like unloaded guns. They may not actually be grounded.

Thanks to Dan H, I learned something will use an Adel clamp on the p-lead.

 

[terrye]

While on that subject...

This is how you install a p-lead on a Slick. Please, no wires flopping around, supported only by a terminal crimp. Use the ground screw as a hard point for an adel clamp.
.

I'd like clarification on the attachment of shields on shielded wires. My understanding is that the shield should only be grounded at one end. The picture in Dan's post shows the shield grounded at the magneto. If (using toggle switches) the shield is grounded at the switch end, then is is correct to not ground the shield at the magneto?

 

[rvbuilder2002]

I'd like clarification on the attachment of shields on shielded wires. My understanding is that the shield should only be grounded at one end. The picture in Dan's post shows the shield grounded at the magneto. If (using toggle switches) the shield is grounded at the switch end, then is is correct to not ground the shield at the magneto?

I have never found it to matter which end of the shield was connected to ground, but it is usually most convenient to just do it at the ground screw on the magneto.

A detail that I feel is very important, is to not connect the shield at both ends, and use the shield as the ground source for the ignition switch.

I cannot count the number of times over the years that I have found damaged P leads on airplanes with the damage primarily being a failure of the shield, which opens the ground path for the switch, if it was connected in that manner.
If wired in this manner, a failure of the shield will result in an always hot mag.

All of the Vans Aircraft wiring diagrams have always showed sourcing a separate ground for the ignition switches, and treat the shielding of the P lead, as only that, just a shield for the wire.

 

[Mikeyb]

Using the DanH method the mag switch is only connected to ground at the magneto. In this case the shield serves as the conductor to ground and the EMI shield.

It has three advantages:
1) it grounds the magneto even if the engine is not grounded to the airframe.
2) the magneto current doesn’t travel through the avionics ground when the mag is shorted.
3) It eliminates a separate wire to the airframe ground.

I can’t see a downside.

 

[DanH]

I'd like clarification on the attachment of shields on shielded wires. My understanding is that the shield should only be grounded at one end. The picture in Dan's post shows the shield grounded at the magneto. If (using toggle switches) the shield is grounded at the switch end, then is is correct to not ground the shield at the magneto?

Basic P-lead wiring below, from Nuckolls. Shield is grounded at mag, not the switch end. When the switch is open, the shield is just a shield. When the switch is closed, the shield grounds the primary.
.