KDX Lighting Stator Rewind How-To

Quick How-To on rewinding your stock lighting stator. The setup I ended up with tested out at 98W, the stock stator at 60W. Now, I don't dual-sport my bike, but I wanted a brighter headlight for those nights when you're caught out just after sunset. 100W is more than enough for a standard headlight / tail light. This procedure was done on an A/H series KDX, but the basic procedure should apply for any bike with a single coil AC generator. Onto the rewind.

You will need:

- A flywheel puller

- A flywheel holder

- Socket set

- Soldering iron

- Side cutters / wire strippers

- 17GA coated magnet wire, 1 small spool should do

Remove the flywheel cover, use a flywheel holder to remove the 17mm nut on the flywheel. use your puller to remove the flywheel. Remember, on the KDX those are LEFT HAND THREADS for the puller.


Unplug the stator wire. The connection is found up on the right side of the bike just by the shock reservoir.

Remove the three phillips screws with an impact hammer if they're stuck, and remove the stator. Pay close attention to the timing reference marks, so you can put the stator back in correctly.

There are two coils on the stator plate, the ignition coil, and the lighting coil. The lighting coil (the one we're concerned with) is the larger, uncoated coil. Go ahead and remove the coil from the stator plate, by removing the two phillips screws. You will need to clip the power wire, do so as close to the coil as you can. This will leave lots to solder back on later. You should now have the coil all by itself.



Now, de-solder the ends of the coil wires, the black ground wire can remain. You want to clean all the old solder off. Remove the old coil of wire. There will be a LOT of it, like 200 yards worth, not kidding.

Pay special attention to how the wire is wrapped around! What direction, and where it starts and where it ends.

Using your coated wire, simply rewind the stator just like it was stock, leaving extra at each end for soldering. Read:

For the best possible result, keep the wire as tight as possible. I had a friend hold the spool taught while I wrapped the stator. This is a time consuming process, be patient and it will pay off. The more wraps of wire you can fit the better, and the tighter they're packed the better.

Be really careful not to scratch the coating off the wire. If there's a short in the stator, you've just dropped your output to nothing. The idea is to fit as much wire as possible on the coil, obviously without hitting the flywheel. The stock wire was wound about to that point, you can use my pics for reference, but that is about all that would fit under the flywheel.

Done wrapping now? Good. Time to solder those ends on. I cut just enough to reach the soldering terminals. This is definitly a measure twice, cut once kinda deal.

Scrape the coating off the last 1/4" or so, and solder the wire to the posts. This was easily done because you paid attention to how it all came off, right? :smirk:

Solder on a new output wire where the old one was attached, leaving about 3" to work with. Your finished stator should look like this:



Cut the output wire to length, solder it back to the output wire you cut in the harness, and heat shrink wrap it to seal it in.

Bolt the coil back to the stator plate:


I like to check the coil for continuity at this point, right from one end of the coil wire to the end of the harness yellow wire. Would be a shame if there was a short and you waited until it was back together to find it, wouldn't it? Any good multimeter is capable of checking continuity.

Reinstall the stator plate, paying attention to the timing marks. This would be a good time to advance or retard your timing, depending on your riding preference. Refer to your service manual for details on that.


Reinstall the flywheel, make sure it doesnt contact your coil! Bolt the flywheel cover back on, and fire it up!

I tested my bike right at the headlight socket, with the stock headlight and tail light on, no other electrical equipment installed.


12.53v AC @ 7.90a AC gives 98.987 watts. Measurements taken just above idle.

Idle gave 12.34v AC @ 7.81a AC, which works out to 96.375 watts.




So, just under 100W for about $18 and a bit of spare time. Good luck!

i love winding my own stuff. great job.

Thanks...a very useful post. What gauge of wire did you use?


What gauge of wire did you use?
You will need:

- A flywheel puller

- A flywheel holder

- Socket set

- Soldering iron

- Side cutters / wire strippers

- 17GA coated magnet wire, 1 small spool should do


As I said over at KDXrider, great write up!


As I said over at KDXrider, great write up!

Thanks! Although someone thought I made an underhanded comment, it was not intended as such. Anyway, glad you enjoyed the write up! Try it on your own bike!

Thanks! Although someone thought I made an underhanded comment...

Eh, I chalk it up to text based communication not being the easiest or best way to get across intent. It's all good though!

Try it on your own bike!

You know, I just might. :smirk:

I might have came off a little harsh. It just seemed a wickedly insensitive thing to say considering the current circumstances over there.

Still waiting on actual results on output! I'm still going with you are not getting what you think you are getting, lol. No offense intended or implied, and certainly nothing personal. :smirk:


.....Nicky Stator? lightning.gif


Well, the results are in! Wasn't 100W, but it wasn't far off! :mrgreen:

I went through a very systematic process today, and here are my findings:

First off, I started out by looking at my "stock" headlight, to see if there was any explanation for the extra current draw I was seeing before. Turns out, it was NOT a stock headlight, but a UFO unit. The way it was wired, was such that BOTH "high and low beam" fillaments were being used, at 35W each, for a total headlight draw of 70W. Note the lower fuse type bulb is not used, I'm talking about both fillaments in the main headlamp itself.



Onto the actual testing:

**All photos were taken at idle unless otherwise noted***

Knowing that my previous headlight was capable of pulling 70W, I knew 1 H7 bulb would be no problem @ 55W. I decided to try it anyway.


The test setup is as follows:

An AC multimeter setup to show AC current, wired in-line with the yellow wire coming off the lighting coil. This is the raw output right from the stator.


An AC voltmeter hooked into the headlight socket, in addition to the bulb(s) used for testing.


The bike easily put out 12.7vAC at a draw of only 4.17a.



Next, I decided to add a second H7 bulb in series with the first one, with the stock 10W tail light.


Here's where things get interesting. I don't think both bulbs were pulling a full 55W each, the math doesn't work out. More like 35W each. As I understand it, the bulbs will only take what's available, hence why they're not at full 55W brightness. There may have been 70W available to them, but it needs to be split now between the two bulbs.

Anyway, with both H7's and the stock taillight, the stator put out an average of 12.3v, at 5.50a, at idle. (See youtube video, bottom of this post.)

Revved up very slightly, I got 12.4v and 6.60amps average. The stator is still able to maintain over 12V at idle, so I decided to add more draw.

I changed the tail light bulb to a 20W 1156, and found the breaking point. At idle, the bike could only maintain 11.8-12.0v with a current draw of 7.02a.



2 H7's and a 20W tail light bulb barely hits 12v:


Revved up slightly, the voltage was still struggling to get over 12v.


So, there it is. I had 2 55W H7 bulbs, in addition to the stock 10W tail light, and the stator "kept up." Figure each headlight bulb was getting about 35W apiece.

12.4v and 6.6a = 82W. That sounds right for two 35W headlights, and a 10W taillight to me.

12.3v and 5.5a = 68W at idle. I'm going to use these averages as just that, averages. In the middle. It's possible that the stator could put out a few more watts, or a few less.


Some may wonder why I took a reading with the bike slightly revved, and not solely at idle. The fact is, we ride, we don't idle. I don't care how bright my headlight is at a stop, I care about how bright it is while I rip down the trail. If I do care while I'm stopped, a quick flick of the wrist cures that. Capiche?




Youtube video:

And for the electrical geeks, also took an electrical frequency reading at idle:


is there a reason you decided to run the lights in series?

would you mind running this expierment in the parallel fashion with the 2 55 watt bulbs.. that way we can see the amps top out and the voltage should stay the same.

a good site http://www.furryelephant.com/content/electricity/series-circuits/

Bill is correct. Wiring them in series you are splitting the 12v supply across the bulbs.

lol, seems like it's taking you more work to test the output than it was to wind the dang lighting coil!

looking again I cant really tell but the photo looks like you have the lights wired in parallel already (each bulb with thier own ground and hot) But in the you tube video you say you have it wired in series.

Bill is correct. Wiring them in series you are splitting the 12v supply across the bulbs.

lol, seems like it's taking you more work to test the output than it was to wind the dang lighting coil!

actually voltage drops after the first bulb in series, but in parallel the bulbs get the same voltage.

I think he is mixing up the terms and had them wired in parallel

Bill is correct. Wiring them in series you are splitting the 12v supply across the bulbs.

lol, seems like it's taking you more work to test the output than it was to wind the dang lighting coil!

I don't mind! I wanna know exactly what this thing is putting out! The bulbs WERE wired in series, paralell test results to follow.

Well, I'm back with what are hopefully the final results!

I wired both headlights up in paralell this time. Please excuse the jimmy-rigged wiring harness, I put it together with a bunch of pieces I had lying around in about 90 seconds. It worked!


Ok, so I have the multimeters set up just like before, the current reading taken tirectly from the yellow wire, and the voltage reading taken directly from the main headlight harness, before it splits to each bulb. My results may surprise you, I know they surprised me!



It's important to note that this time a taillight bulb was NOT installed!

Onto the test!

I started the bike, let it warm up (it's a chilly -10 here right now!) and then took my readings, so everything was as it would be if you were going riding. This time, the bulbs were in fact much brighter. At idle, the stator can not sustain 12V with both bulbs hooked up. I saw 11.3v at 7.33a.



Well below the minimum 12v, but we want to see how many watts we can support at 12v. I revved the engine ever so slightly and voila! 12.0v appeared, a few hundred RPM above idle. A quick check of the amperage being drawn showed 7.69a. Both bulbs were burning bright as hell too!



Then I got greedy, I wanted to see what the stator could really do. I revved it even higher, and (you're not gonna believe this!) I was shown 12.6v at 8.21a! That's 103W! :smirk:



Now, I'll reiterate that the only electrical equipment installed were the two headlights, the tail light was removed. The final reading was taken at an RPM that you probably wouldn't cruise at, but it's nice to see that the stator is at least capable of that much!

So, what does this all mean...

Well, when we look at our three test samples:

Idle - 11.3v at 7.33a = 83W

Just above idle - 12.0v at 7.69a = 92W

Riding RPM - 12.6v at 8.21a = 103W

Obviously, it's a generator, and like any generator, output varies with RPM. So, lets compare these results with my previous tests, and see what we come up with. While the voltage did drop at idle compared to my previous tests, the current also increased, and we all know that voltage x current = watts, so I ended up with wattage outputs that weren't too far off from each other, which makes sense. A coil is a coil, and at a given RPM will only put out a certain number of watts. How those watts are distributed between voltage and current doesn't seem to matter as much, it can only put out X watts at Y RPM. CHkdx was correct in the sense that it doesn't care how the power is distributed after that.

Now, I had this curcuit loaded up with a potential 110W draw. That's huge. Think about that. That's a set of car headlights! I wanted to see at what point the upgraded stator would fail, and I believe in this sense, I exceeded it. I wasn't able to fully pull 110W out of the coil, and even the 103W I did get were at riding RPM, and really, to me, having the brightness just above idle is where it counts. Like the stock stator, there was a noticeable change in brightness as RPM's got higher. Come to think of it, There really would be no way to measure it, but with this coil having to supply over 100W of power, it's going to get HOT! I don't know that you'd want a coil that is going to make much more heat than that, in an enclosed case with no airflow.

So what is it? A 100W stator? An 80W stator? That's up to you. What I can tell through my experimentation, is that the coil is physically capable of putting out over 100W of power. Would I run two 50W bulbs on it everyday? Probably not, but the power is there. I run a 70W headlight, and a 10W tail light. That's good enough for me. It doesn't dim too bad, or flicker at idle, and 70W of headlight is plenty for the amount of time I'm stuck in the bush after dark.

I learned a LOT in doing these tests, I was pleasently surprised at some results. While there has been a lot of trial and error, checking and re-checking, picture taking and note taking, in the end I find it very rewarding. For a minimal investment, I've created a lighting coil for the KDX that's almost twice as powerful as the stock one, and learned a ton about how the bike works. That's good enough for me!

Great Write Up.........Thanks!

What is the 70 watt headlight, that you are running?

Edited by BlueLightRhyno

I'm not currently. I'm running a retrofitted H7 automotive bulb, 55W. It's bright as hell!

I'm not currently. I'm running a retrofitted H7 automotive bulb, 55W. It's bright as hell!

Is this your best lighting setup, yet? I am assuming that the new bulb, is in the factory KDX housing? I am trying to light up a KX 500, and don't want to relearn everybody else's lessons. It has a lighting coil, but no headlight or switch, and I am unsure of what it puts out. If it needs more, I'm going to do your rewind tricks.....

If you were in my shoes, what would you use?

Are you SURE it has a lighting coil? I didn't think KX500's came with them.

It's a UFO stock replica that I modified to accept an H7 bulb.

Are you SURE it has a lighting coil? I didn't think KX500's came with them.

It's a UFO stock replica that I modified to accept an H7 bulb.

I don't think it came stock, but the previous owner(s) put a lot of time and energy into it.

I haven't pulled the case/cover over the flywheel. It has three wires going up to a possible rectifier with leads going from that to the coil. It also has two extra wires, with bullet connectors, that are taped up, in an attempt to insulate. I haven't put a volt meter on the leads, so I really am unsure of what they are for.

Thanks for the help....

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Similar Content

    • By luke8500
      Hi all,
      I have a 2004 Kawasaki kx250 that died suddenly on me the other day. all i can get it to do is backfire every once in awhile. Tested the resistance and everything of the magneto and stator with a multi-meter and all is good.  Coil is new oem Kawasaki the reeds are brand new vforce and the top end and carb have been completely rebuilt.  key on crank is intact and all the timing marks are lined up. I hooked a timing light to it and spun the motor over really fast and found that it is sparking way to late.  The service manual says that it is supposed to fire 14 degrees before TDC but it is sparking way after TDC.  Im thinking its a fried CDI box since its the only thing i wasnt able to check.  Any help would be appreciated
    • By luke8500
      Hi all,
      I have a 2004 kx 250 2 stroke and it wont start.  it only backfires.  I was out riding it one day and came up to a stop so the bike was just idling and it started to try and die so i revved it up a few times then let it die and now it wont start back up.  It only pops or backfires when you kick it.  Its getting good spark and i just put a brand new plug and coil in it. I took the reeds apart and they look like they're brand new.  It was running great before so i dont know what happened.  Im thinking it could be a fuel problem even tho the plug is a little wet if you take it out after a bunch of kicking.  Any help would be greatly appreciated 
    • By Zach121
      Alright, found three bikes all located where i live in Texas. I know they're all very different, but if you could tell me what you think each bike is worth, and which one would be the best deal. Thanks.
      Notes: Both 2 strokes have new top ends. I think the front brake lever is broken on the kawasaki.
      KX250 2002 $800-$1000?

      YZ125 2000 $1250

      KTM 400SX 2002 $1200

    • By Ryan Turner
      So I'm at a bit of a cross roads, I currently own a 06 rm250 and I love it. I also have a 02 kx250 which I'm selling with intent to buy a newer two stroke. I can't afford a newer ktm and the older ones aren't that great and kawasaki never impressed me. so I'm stuck between 06 yz250 and 05 cr250. I've heard the yz and rm after 06 are almost the same bike just the Suzuki handles better and the Yamaha has better suspension. So I'm not sure if I wanna buy carbon copy of my bike with slight differences. That's why I'm interested in a cr250 I heard the 02+ models RC valves are shit garbage but was mostly fixed in the 05-07 models. I ride at a track and at some bogs down the road from me. So from people who have owned both, what would you recommend. I know the cr is known for a hard hitting power band which I enjoy since im an aggressive rider. Should i just see which deal comes my way first.
    • By dirtbikesrule
      If you own a 2 stroke dirtbike,please reply once. Im just wondering how many people on this site have them.