Strange Electrical Gremlin (High RPM)

[KDXGarage]

I'll bet one dollar on coil of some sort due to high heat making it worse. From what I have read over the years, a CDI usually works or does not work.

Do I get a Kanuckistani dollar?

[Slick_Nick]

Well, did it work? And the answer is...

Sort of. I obviously have some work to do. But hey, that's learning for you.

It worked in the sense that it makes a spark, and ignites the fuel/air mixture. But the engine doesn't run. It coughs, and wants to rip my leg off via the kickstarter. There is an ignition timing problem. It's almost as though it's firing the plug while the piston is at BDC, if the engine isn't spinning fast enough, it will actually kick back so hard that it will jump up the wrong direction of rotation. I kept the flywheel cover off for this inagural run, so I could see if she lets the smoke out, and could see the flywheel "kick back" against the force of combustion.

In order to trigger the CDI to fire the spark plug, there is a distinct break in the magnets on the flywheel. I'm not an electrical engineer, but I'm guessing this would momentarily allow the magnetic field generated in the stator coil to collapse, and the CDI uses this trigger signal as a means to determine engine position, and fire the spark plug accordingly. If my rewound stator is "too good" it could mean that this field doesn't fully collapse, and the CDI has no reference on engine position as to when to trigger the spark. Remember, my rewind tested to DOUBLE the resistance of the stock coil.

So, at least we learned something. My next option is to grab a stator off the shelf from a vendor, or unwind it, and rewind it with a thicker gauge wire. Perhaps 32AWG? I'm going to dissect the remnants of the stock winding and see if I can better determine the wire gauge used.

[Slick_Nick]

Alright, dug some of the stock wire out of the trash, and got to measuring. Measured a few strands, and got an average measurement of 0.0069" which puts it right around the nominal diameter for 34 gauge round magnet wire, according to the table here:

https://www.coonerwire.com/magnet-wire/

So, back to the eectronics shop I went, grabbed a fresh spool of 34AWG, and I'm going to try another rewind. Not so many wraps this time either. I'm going to really aim for the factory spec of 250-370 ohms, just to see if this works. The thing is, it's still a guessing game. There's no way to cut the wire, measure the resistance, and then keep going. Once the wire is cut, that's that. So maybe I can go too far, check it, and unwind a few wraps if need be.



Looking at a sample RickyStator replacement coil, it looks about half the diameter of the one I just rewound, maybe I did go a bit overboard. Also, I used a much thinner insulation around the core compared to stock, I will beef this up as well.

I am going to weigh the coil before and after unwinding, using that table above I should be able to work out roughly how many feet of wire I wound around the stator using this method.

[Molly's 70]

I was hoping it would work. Dang it!

Remember, If you mess with something long enough, and screw it up, you can always blame the idiot that engineered it.

[KDXGarage]

Well, dang. Keep at it!

[pumpguy]

Have you contacted Jaguar (Michael Forest)? He has much to say on this subject and may be able to offer some insight. He has a web page on this very subject.

[Slick_Nick]

Alright, got it rewound with 34AWG to match the stock winding. Fit on as much as I possibly could.



Tested it out, and got 367 ohms. Right on the top end of factory spec, got every last bit of wire on there!




Installed in the bike, I'll (hopefully) fire it up today!

[Molly's 70]

Well, it looks damn good!

[Slick_Nick]

Alright, welp, that didn't go as planned. I got it all back together, and went to start the bike. It coughed, wheezed, and wouldn't start. I observed a nice, strong spark when the plug was tested outside the engine, it was as though the ignition timing was way off or something. Did I fry my CDI with the super skookum original wind perhaps? Back to the books I went, and suddenly I had a eureka moment while looking at a CDI schematic. Most 5 wire CDI's have an external trigger input, like a separate pickup coil on the crankshaft, etc. All of my KXF's have them, but the KDX doesn't. The KDX uses the ignition source coil itself as a trigger for the CDI to know engine position. This is nothing new, I made reference to this a few posts back. What I didn't fully understand, however, was HOW the CDI interpreted that signal from the stator coil.

The source coil generates an AC waveform, as the N and S poles of the magnets pass the stator. This also generates electricity as it does so to power the CDI, and ignition coil. Looking at a schematic of the CDI, I completely missed that there is a diode on white / red input wire from the stator. When you apply a diode to an AC current, you basically end up with a single phase rectifier. In a nutshell, the "negative" portion of the AC waveform is discarded, and the CDI will "see" pulses of DC positive electricity. Eureka!

You see, when I unwound the OEM stator, there was no real way to tell which direction the wire wraps were made. The wires are tiny, and the entire thing was encased in a thick, hard epoxy, and had to be destroyed to get it off. I mistakenly assumed that since it's an AC stator winding, the wire wrap direction wouldn't matter. In any other bike, with a separate device to trigger the CDI, this would be true, because the stator coil serves only to generate power for the CDI and coil. In the KDX, however, since it uses the waveform itself as a trigger, the winding direction in fact does matter, and since half the waveform is discarded by that damn diode, the CDI was basically only seeing the pulses that were 180* out of phase from where they should be. From what I can tell, at that little break in the flywheel magnet, there are two N poles set side by side, with that break in the middle. The CDI must see this as some sort of a "double blip" and use that as a reference to determine engine position, much like you would see on a motronic style trigger wheel with a missing tooth, etc. Because I had the coil wound incorrectly, this signal likely wasn't seen by the CDI and it had no idea on when to trigger the spark event, or was triggering it 180* out of phase.

So, off it came, for the 3rd time I got to rewinding. This time, in the opposite direction. Tested again, and while we didn't do as good as 367 Ohms, we are right smack in the middle of stock spec at 323. Epoxy coated, and soldered back onto the stator plate, I kicked it over with the plug out, and observed a nice, strong blue spark. But, the question remained;

Will it chooch??

[Slick_Nick]

Man, it runs like a hot damn! Fired up about halfway through the first kick, and that was an engine that has no doubt been flooded by me sitting there and kicking it for the last 3 days. It's way too slippery to ride, but once the engine was warm I took it up to redline a few times on the stand and It revved smoothly, and consistently. Turns out the CDI was good after all! Hopefully I didn't piss my neighbors off too bad... The rewound lighting stator worked pretty well too, as did my new scamazon ignition coil.

For myself and anyone else out there who may be insane enough to attempt this in the future, here are some photos showing the correct direction to wind the ignition stator coil.

[Slick_Nick]

Hopefully now, we can put this whole issue to bed. I learned a lot, hopefully someone else out there did too. Also got to be pretty proficient at rewinding stator coils! One thing I will say though, is the plastic "ends" of the stators, after all this time and all these rewinds are certainly starting to deteriorate. Probably why all the aftermarket companies do away with them completely and just have two loose wire end pigtails to be soldered in place. The heat from soldering, the removal of epoxy, they all take their toll on those end pieces.

Thanks to everyone who provided insight and help, I feel as though I've now got a pretty solid grasp on how our ignition system works, please ask away if you have any questions. I'm going to remove the tank, drain the gas, change the fluids, clean the carb, air up the tires, and put it under a cover for the next few months (unless I make a trip to warmer climate with it!) Can't wait to try it out in the spring!

[Molly's 70]

Glad it worked out. That is cool!

[KDXGarage]

Congratulations on getting it all sorted out! Thanks for educating us along the way.

[SS109]

My brain doesn't work well any more. You started wrapping on the right in that pic and your wrapping over the top towards you, correct? I want to be absolutely sure as I don't like winding stators as much as you!

Another question if you have the answer handy, what is the output voltage range of the source coil? I decided to get back to my adjustable CDI build and the info would be helpful in making some decisions.

Anyway, as always, great job! When you're ready to leave Hoth and come ride Tatooine let me know.

[Slick_Nick]

My brain doesn't work well any more. You started wrapping on the right in that pic and your wrapping over the top towards you, correct? I want to be absolutely sure as I don't like winding stators as much as you!

Another question if you have the answer handy, what is the output voltage range of the source coil? I decided to get back to my adjustable CDI build and the info would be helpful in making some decisions.

Anyway, as always, great job! When you're ready to leave Hoth and come ride Tatooine let me know.

That’s correct. Everything I found says the “ground” side of the stator winding is the inside of the wrap. The lathe chuck was rotating front to back when looking at it from above like that.

As for the voltage, none of my multimeters do peak voltage, so I wasn’t able to get a reading on that.

Hopefully I can blast off and head over to see you this winter! I’m moving onto a new type at work come January so I’ll be super busy with that but I’ll try to make something work!

[SS109]

That’s correct. Everything I found says the “ground” side of the stator winding is the inside of the wrap. The lathe chuck was rotating front to back when looking at it from above like that.

Thanks for the confirmation.

As for the voltage, none of my multimeters do peak voltage, so I wasn’t able to get a reading on that.

Cool. I'll have to test mine at some point then. Needing low and high voltage range and milliamps as well.

Hopefully I can blast off and head over to see you this winter! I’m moving onto a new type at work come January so I’ll be super busy with that but I’ll try to make something work!

Sweet! Keep me posted and I'll try to make sure to get you out for some good riding.