Flywheel Cap Screw Failure

[mike.smith]

I have 80 hours on my AeroVee, which first flew last June. I generally cruise with the secondary ignition turned off (take-off and land with it on). I get more charge to the battery that way. The plane was flying great on the primary alone, or the primary and secondary, with no hiccups and no anomalies. But on a go-around on my last flight the tachometer went haywire, bouncing between 2,000 RPM and 6,000+ RPM. The engine was running normally, so I just flew it like I normally do and landed uneventfully. I figured it was a loose wire.

When I landed I opened the cowl to have a look. The fuse between the alternator and the voltage regulator (30 amp) was blown, so I replaced it, but this obviously meant there was a problem with the alternator/ignition. When I turned the prop by hand I heard a lot of grinding. Not good! But I noticed the grinding was coming from the rear of the engine and not inside the case. I then noticed the trigger cap was not turning on center, and one side of it was pretty well scored. A little more looking around and I saw that one of the triggers had it's plastic casing worn away by the off center trigger cap. I took off the stator assembly and there was what looked like a pile of sand inside. I also found a sheared off cap screw head, and a lot of abrasion and heat damage inside.

So I got the plane in a hangar and proceeded to remove the engine. There is no other way to investigate or fix all this.

In a nutshell, everything attached to the crank hub at the rear of the engine (it's part of the whole flywheel assembly) was loose. Very, very loose. As in screws were backed out or so loose I could turn them with my finger. That allowed everything attached to the crank hub to wobble, and become loose themselves. Everything was wobbling so badly that there was a tremendous amount of friction between all those parts, generating a lot of heat and a lot of rubbing. So much heat and rubbing that some of the aluminum parts look like they became molten on the edges and flung off, which scored other parts as the material skidded across the adjacent spinning parts.

A video and photos are available at: https://www.flickr.com/gp/127253485@N06/e4b1PN

So here is what I observed:
- When I first got to the back of the engine I took off the stator plate. The inside was filled with what looked like sand, and the sheared off head of one of the cap screws. Like the flywheel assembly, the trigger shaft was loose, with almost the same symptoms as the flywheel cap screws: one was sheared off, one was backed out 1/8" or so, the third was loose enough to remove with my finger, and the fourth was firmly and completely set (I later had to spray in WD-40 and then heat it with a torch to get it out). All 4 cap screws had LocTite residue on them from when I installed them.
- The "sand" turned out to be the epoxy that coats the alternator stator copper windings, which had been chewed up by the cap screw head and by the off center magnet ring and trigger shaft.
- On page 31 of the AV2.1 manual it shows the flyweel as a complete assembly. In the picture I see the crank hub already mounted to the flywheel. There is no photo, but the alternator ring with the magnets is on the opposite face, as part of that flywheel assembly. All of that is secured with 4 cap screws.
- What I found was the 4 cap screws that held those 3 pieces (hub, flywheel, magnet ring) together, were extremely loose. One was backed out 1/8" plus. The second was backed out about 1/8". The third was loose enough to turn it out with my finger, and the 4th was seated but turned with just a light push on the allen wrench. Before removing the screws the flywheel could easily be moved fore and aft around the pivot point of the one seated screw (see the Flickr video).
- These were the same 4 screws that were part of the service bulletin issued last year, which I completed before the engine ever ran. I documented it on my Kitlog page, putting on the LocTite primer and LocTite.
- The hub is still firmly mounted on the engine with the gland nut.
- All of the parts have heat damage where the chafing was occurring. You can see the off color of the anodizing in the photos.
- Both ignition triggers had their plastic casings worn away by the off center trigger cap, and the white heat sink paste was oozing out from behind them.
- All of the aluminum parts need to be replaced. Some of the screw holes are slightly elongated, and they have all had heat damage.

So I need to replace the following:
- (2) triggers
- Alternator stator
- Trigger shaft assembly (w/ trigger cap)
- Flywheel assembly (flywheel and alternator ring)
- Rear oil seal
- New cap screws for the trigger shaft

So it appears the flywheel cap screws from the service bulletin backed out, unbalanced everything, allowed contact between spinning parts with resulting friction and high heat, and caused the trigger shaft screws to loosen and back out as well (at least I assume that the unbalanced vibrations and heat are what caused the trigger shaft screws to loosen; it would be quite a coincidence if both sets of cap screws backed out at the same time without some cause/effect).

I have no idea of the "why" of all this, but I did follow the service bulletin to the letter, even using the expensive LocTite primer. Any thoughts about how to prevent this happening again would be helpful. Unfortunately taking apart the back of the engine is the only sure way to check all this. To do that you need to remove the stator plate, remove the 4 cap screws holding the trigger shaft, and then check the 4 cap screws holding the flywheel assembly. You can do all that with the engine on the plane, but it's a real pain in the butt (and the knuckles), and requires re-setting the secondary timing after the trigger shaft is reinstalled.

I think one of my inflight tests will be to turn on both ignitions, then turn off the primary. That will at least tell me if the secondary is working properly, too.

Although writing the check for all the parts hurts (badly), it's really only a day of work to put it all back together, then probably another day to put the engine back on and hook up all the wires and carb. So it could be worse (I'm trying to look on the bright side).

Am I upset about all this? You'd better believe it! $1,500 in parts and no way to know it won't happen again. I'd like to find a good way to solve the problem, since following the service bulletin was not enough to prevent this. Thoughts are welcome.

[kevinh]

Wow - bummer, but a great write up. I'm really interested in other responses (especially from Aero Conversions). I wonder if there is a better way to safety those bolts than the service bulletin recommendation?

[Gunther]

Sorry to hear of your problems Mike. I would really like to understand how the 30A fuse blew, especially since the alternator is rated at 20A. If the current came from the battery, how did it go backwards through the regulator? It might not have helped in your case, but there are ways to detect an alternator problem early. My favourite method is measuring the AC voltage from each alternator output wire to ground. A healthy alternator will show the same voltage to ground for each conductor, roughly half of 12-14V, so 6-7 volts to ground. If an alternator winding starts to fail, one voltmeter will read higher and one will read lower. I have been thinking about installing an extra voltmeter to experiment with.

Gunther

[mike.smith]

Sorry to hear of your problems Mike. I would really like to understand how the 30A fuse blew, especially since the alternator is rated at 20A. If the current came from the battery, how did it go backwards through the regulator? It might not have helped in your case, but there are ways to detect an alternator problem early. My favourite method is measuring the AC voltage from each alternator output wire to ground. A healthy alternator will show the same voltage to ground for each conductor, roughly half of 12-14V, so 6-7 volts to ground. If an alternator winding starts to fail, one voltmeter will read higher and one will read lower. I have been thinking about installing an extra voltmeter to experiment with.

Gunther

The fuse was between the alternator and the regulator. The voltage spike likely came from the alternator, which was in the final process of destroying itself at that moment.

[kmacht]

$1500 in parts to fix a known issue with the aerovee that the service bulletin obviously did not fix!! That has to hurt. I'm disappointed in Sonex/Aeroconversions that they did not give you a substantial discount on the parts. I guess they don't believe in any sort of warrantee or goodwill for design defects on their engines. The biggest concern to me is that this is a failure mode that there is no way to detect prior to it becoming catastrophic. Hopefully others see this post and let us know if they experience the same thing and that it drives Sonex to come up with a better fix other than just to use some loctite primer. Critical rotating parts on an engine should not be relying on loctite alone to keep them from failing.

Keith
#554

[gammaxy]

The fuse was between the alternator and the regulator. The voltage spike likely came from the alternator, which was in the final process of destroying itself at that moment.

I had the same question as Gunther, especially since I don't have a fuse between my alternator and regulator. Usually, fuses blow due to too much current and it's not clear how a self-destructing alternator would deliver too much current. What I suspect happened is a winding on the alternator touched ground as it was coming apart. The positive terminal on the battery is directly connected to one of the AC tabs on the voltage regulator (which is connected to the windings in the alternator), so it shorted the battery right through the fuse and blew it open. Due to the way the tabs on the voltage regulator are connected, I think this probably did not damage the regulator.

Sorry this happened to you.

[Glenn.hunt]

Wow Mike, that sure bites. I only heave 23 hrs. on my engine since I pulled the alternator off to re-shim the crank. I had around 100 hours on it when I pulled it apart and those cap screws were tight. When I replaced them I used ample Locktite but did not use the primer. Now you've got me concerned about them loosening up. I guess I may need to be proactive and pull it apart again to make sure they are still installed tight.

[vwglenn]

I'd replace that gland nut too. They aren't that expensive and that thing holds the whole thing in place.

One thing that's always kinda bugged me about the Aerovees and the Jabs both is the way they crank up. There's nothing scientific here but it seems to me the starters on these planes may have more torque than they need to get these things fired up. They seem to spool up really fast and I wonder if the torque and/or speed these starters are engaging my be contributing to the premature fatigue of the bolts on the back side of the engines?

Dynamic prop balance would certainly help keep overall vibrations to a minimum and potentially reduce the load on all the spinning parts of the engine.

All of the above statements are opinion. Just thoughts. I'm not in any way an expert.

[mike.smith]

Hey everyone:

It was pointed out to me that I made an error in my description. The service bulletin (which I complied with) was to replace the cap screws for the trigger shaft.

"Sonex Aircraft requires the immediate replacement of all ACV-Z01-36 (3/4” long) Flat Head Socket Cap Screws that hold the Trigger Shaft Assembly (ACV-F01-24) to the Flywheel with ACV-Z01-81 (7/8” long) Flat Head Socket Cap Screws."

Both sets of cap screws failed (trigger shaft and flywheel). There's no way to tell which came first, but the flywheel cap screws are installed by the Sonex factory, since they are part of the whole flywheel assembly.

[petep]

I can't speak with any authority on the AeroVee but the Jab has to turn up 300 plus RPM before the mags generate sifficient spark to run unless you have installed the new cold start kit.

The blown fuse between the battery and alternator could be either a alternator or battery problem but most likely an alternator. I f the regulator ias seeing a need to charge the battery and the alternator go low in voltage I think the current curve woulkd be inverse to the voltage. I am not a EE and did not sleep at Holiday Inn last night so my logic here may be flawed.

Pete