912ULS ignition problem

[Dave S]

Good afternoon,

Our Kitfox with a 2003 Vintage 912ULS & 400 hrs recently (last week) suffered an ignition problem. I'd like to check with the group to assure I am not overlooking something with my diagnosis.

A couple notes

A) Some time ago I switched to a narrower spark plug gap on Larry's advice where he has had better luck with with this modification - using 0.024"

B) The main battery was replaced within the last few months, not because it wasn't working, but because of it's age and a slightly lower float voltage

C) Fuel system including starting carb verified OK - it's getting gas. I know a starting carb failing to close completely can cause starting problems.

C) I have not done anything with the stator/pickups other than to assure the wires are OK and nothing is loose.

Here is the sequence of events:

1) On a cold start (70 degrees F ambient) a week ago the engine spun over vigorously without firing at all for much, much longer than what is normal. A second try and it finally fired and ran normally. Went flying

2) Two days later, (About the same ambient) it spun over vigorously without firing, gave it a rest, tried again with no firing; and, did it a third time with the same result. Time to hang it up and figure out what is wrong. Not flying.

3) Got some tools the next day and found that cranking with a timing light on all 8 plug wires, none of them showed any spark.

4) Took apart the panel to get the starting switch out and verified that it was working right, right and left grounds fine, the P leads were fine and the Module end of the P leads were connected properly.

5) The baffling part of this is why would a dual ignition system apparently die on both sides at the same time

6) Today, Based on some anecdotal reports on the web about these engines, put my parlor heater on the modules and warmed them up for about 15-20 minutes Engine fired up instantly and the timing light showed spark. Engine ran normally after that.

7) I am going to let it cool for a day, then go back and double check the manifold grounds for the modules (ULS has a separate lead going to two separate grounds on the manifold. I want to be sure that not only are the grounds at the manifold assured (checked them) but that the manifold has a solid ground to the block.

So question - Can anyone think of anything I have missed or miscalculated on. Modules are kinda expensive. Following this latest experiment with heating up the modules and having good spark, I now wonder if one module was weak all along and the second one decided to join it making it look like a double fail.

Any thoughts and comments welcome.

Thanks for you input.

[avidflyer]

Some random thoughts, can you just heat one module at a time, and see if the other one shows dead right after start? Or if you can borrow a module from someone, try yours without heat, and when/if it doesn't start, change to the good borrowed module and see what happens. I imaging you have the 4 pin modules. Another thing to maybe try if you have the ability to start on just one mag (separate mag switches) heat one module, and try to start on the other one, and if it doesn't start on the cold module, try right away on the heated one and see if it starts. Actually thinking about it some more, even if you have the ignition switch with mags 1,2, both, start, you could jury rig a wire to the starter solenoid and still try to start on just one mag by putting the switch on that mag. As far as ground wires go, I had one coil not producing spark, and looked at the grounds, and they all looked fine. I pulled the bad coil off, and the ground was still connected but I missed it. When I pulled the coil toward me, the plastic insulation on the ground terminal broke, and the coil came free. The wire had broken before, and the plastic was all that was holding it in place. When I checked the coil ground wire before I took it off, I pulled on it and thought it was ok, but if I had pulled a bit harder, the plastic would have broke then. I put a new terminal on and was back in buisness. JImChuk

[Slyfox]

well I just went through the mill on my 2008 rotax 912uls. I went out to the airport a couple weeks ago and it wouldn't start. I left it and came back out later in the day and it started. but fluttered for the first few seconds and than ran just fine. after a couple times in the morning of starting it did, but had the flutter. I figured out what was the flutter. One of the mags was not working. I have individual mag switches and I would right after start up do a mag check. one side was not working at all. after the flutter quit that mag was working. so now I know that one was not working for startup. possible the other one is the same but wasn't doing it. I went back out the next day and went through all my ground from battery to firewall forward to the grounds at the cdi modules. I then went home and did some research on the modules. seems they have troubles maybe from heat, maybe from moisture. they start going bad when they won't fire after sitting for a while, you can heat them and they work. after the engine starts they keep working. Now my modules are the 6 pin so I can buy new ones with soft start which is nice for the uls engine. keep in mind the modules are 900 each. ouch! so what I did was buy one module, I figured out which one was which and put the new one on the left side. I got the wire kit to hook the soft start up. I hooked both sides up, but the right side is a nothing without the new module, if some day I get one I just plug it in. so anyway I put in the new module and put that old module in my seat for if I ever need it. Nice to have the new soft start module for starting on the left side, I only have that one on when starting, after a few seconds it goes to full timing and than I turn on the right side. has improved the starting of this engine. Now my thinking is, if the right mag goes all the way out I'll just grab the other module and put it in, than go buy another new module for the right side.

Now to understand how this system works. the modules get voltage from the stator, nothing else. the mag switches put ground to the modules to turn them off, nothing else. if your mag switches are bad, the engine won't turn off. so forget any majic from the switches to turn on the modules other than if the ground is actually not going to the modules. yes, if your switches are bad that engine if turned fast enough with even your master off will start. so that leaves your stator or coils at fault for not starting the engine. most likely the modules. they have had lots of trouble, mainly the 2006 and 7 ones, but lots of them are having troubles.

One more thing I did at the moment with my modules is give them more spacing between them in hopes to have them run a little cooler, I put some spacers in there. hope this helps.

[Av8r3400]

Look to your ground leads for the modules. Possible breaks inside the wire insulation or bad connections.

[Slyfox]

in point no. 6 he said he put heat on his modules and it fired right up. sounds like possible modules. grounds yes yes yes, they have real bad ground, wiring of any kind. check that real good.

[Rodney]

Dave. I have an extra ignition module I can send you for testing if you
might need it. PM me with your contact info

Rodney

[Dave S]

Hi Rodney,

Thank you for your generous offer. Based on several folks who have responded with ideas, I finally contacted Lockwood. They indicated the symptoms I was observing were clearly an indication of a "cut in " issue with the modules (which is apparently not that rare with the vintage of the modules I have); and, they confirmed that often one module goes bad and will not trigger spark so a person does not notice it since the second module will start the engine just fine - then the bad one will cut in when the RPM gets up to 2,000 or so a mag check will be normal once the engine is running. Because the first modules starting RPM failure is masked, if the second develops the same problem it seems like a double fail; but it is just the second one going bad when the first one was already bad. Apparently this affects starting only and not the running condition (which is a relief to know).

Another thing that Lockwood said to try was to cool the modules with ice packs - which baffled me because warming seems to make the darn things work normally. Lo and behold - tried that and the engine fired up and ran normally. Let it sit for a day and it produced no spark again - applied heat and and it ran fine. Why either heating or cooling would make the thing develop spark beats me - the voodoo of semiconductors I guess.

One thing Lockwood indicated was that testing all 8 plugs with a timing light takes the mystery out of the diagnosis. I checked just about everything else first - starting carb operation, battery condition and starting RPM as well as the grounds and P leads (which all were normal).

Based on the input from this group and the confirmation from Lockwood, I have decided to just bite the bullet and get two new modules which will also give a person the new "soft start" feature. BTW - Lockwood indicated that some owners replace one module rather than both and let the other one go since it will run OK once the engine is started; however, I hate to think of being out camping somewhere and rely only on one good module for a start. Also not sure the "soft start" would work with one old style and one new module.

Again I want to thank you for your offer; and, I think I will go with new modules & the soft start.

Sincerely

[Slyfox]

as indicated in my first post. yes they work with one new and one old. but you have to use individual mags switches to make it work(soft start). if you have an ignition switch with the mags as part of , it won't work for you need to switch on the mag that has the soft start, start the engine and wait until the timing goes normal before switching the other mag on. I put the offending module in my seat for if I needed it on a trip. that one does work if you get the idle up to 2000. hope this helps

[Jono]

....They indicated the symptoms I was observing were clearly an indication of a "cut in " issue with the modules (which is apparently not that rare with the vintage of the modules I have); and, they confirmed that often one module goes bad and will not trigger spark so a person does not notice it since the second module will start the engine just fine - then the bad one will cut in when the RPM gets up to 2,000 or so a mag check will be normal once the engine is running. Because the first modules starting RPM failure is masked, if the second develops the same problem it seems like a double fail; but it is just the second one going bad when the first one was already bad. ...

This makes me wonder if one should try starting on opposing modules every once in a while just to check that it has not failed..... Does anybody have compelling reasons for or against?

Cheers

Jono

[Dave S]

Jono,

You know, your thoughts are very much in line with another pilot/mechanic who has far more experience than me. He posited, "Why do we only check our ignitions systems at one particular time (runup) and one particular RPM (XXXX RPM)? why not occasionally check at idle, cruise, other RPM settings, other conditions of operation, etc? - a person might uncover something they really want to know about."

I also think, given the unique operation of a rotax ignition, this is another thought that would support the case of having separate mag switches so a person could do this (like Skysteve and others have done).

Obviously my engine was starting on one mag till the other module died for who-knows-how long. Almost all engines, other than aircraft engines, run on a single ignition - no problem with that.

I like your thought on this:)

[Slyfox]

good thought. I found my problem because once it wouldn't start at all. after that happened it started every time. I pushed to find something wrong. Now, I also start my plane with the headsets off and pay close attention to startup and run. I noticed that my plane was starting up but had a different sound after start up. well after not starting at all just cranking I decided to do a mag check right after startup, that's when I found that one module was not firing. after a few second, more than likely after it reached a good rpm like 2000 the offending module started working. At that same time that the module started working of course the engine sounded much better. another good test is to go to one side and than the other when at operating rpm, in other words, do a mag check in flight.

[Dave S]

Update on the ignition problem

Following some extensive diagnostic work and great ideas from many of you on the list, I called up the Lockwood tech folks who confirmed the issue was what they call a "cut in" problem - i.e. normal cranking speed will not result in a spark on a bad module. Likely one of the modules had been bad for a while so when the second one went bad it only appeared to be a double fail with no spark at cranking speed on either side. - With one good and one bad the engine starts on one then the second one starts working when the engine is up to idle speed - so a runup/mag check will never catch this.

Anyway, decided to get two new modules of the "soft start" design and the little harness to make that work. With the new parts the engine starts normally and runs like it should.

30+ knot surface winds today so had to suffice with a ground run til that settles down.

An interesting comment from Lockwood - apparently either warming up a bad module OR cooling it will cause it to work - so I tried icing the modules one morning - sure enough that worked too - I still claim semiconductors are all voodoo:rolleyes:

Thanks to all who submitted ideas on this

[Shadowrider]

Good job fixing this and thank you for posting what you found. I am sure this thread will help someone down the road!