TE410E/TE610E ignition CDI - Help & Info

[NoMad410]

Introduction:

I'm new to Huskies and off-road, and enjoying very much, but recently one morning, for no immediately obvious reason, my TE410E refused to start. I found there was no spark, and tracing back I realised just how much at the mercy of the electrics we are!

I need to sort out this problem quickly, and reading through various forums I see I am not the only one who's got it. It does not help that the TE410/610 etc workshop manual is pretty useless when it comes to the electrics.

This stuff should be really simple, it's not rocket science, but so far I have found the going a bit tough, and feel hostage to the electrics and its purveyors!

So, I thought that a thread providing some definitive answers to the problems of early Husky 410e/610E ignitions (specially CDIs) is just what is needed. I am no expert in the subject, just learning as I go along. However. I have already found quite a bit of information, and also lots of threads where people are looking for that same information. So I'm collecting it all here. If you are knowledgeable, please do help!

I think the ideal would be to have a good alternative to the current CDI solutions at the end of this, or perhaps even an "open source" CDI design that anyone can use. Perhaps for other models too. For instance, I have seen people claim on other threads (I must start saving those links) that the Lifan 150 CDI can be adapted, though I do not know how yet - here's hoping! A solution like this would be brilliant and probably cost a fifth of the current prices asked on fleabay for CDIs. Even designing a clone CDI shouldn't be impossible. From what I've understood so far, it is not a PIC type and probably just an old thyristor design. Just a case of finding out what wire to plug where...

Any help most gratefully received. I'll be posting this up to ThumperTalk as well, though it would be nice to have a way not to duplicate effort.

 

[NoMad410]

Tracing the problem:

Once I had no spark, I realised I had to check back through the coil, the CDI and the pick-ups. The previous owner thinks it's very unlikely it's the pick-up as he had it changed just before I bought the bike, about three months ago. Hmmm.

The coil was checked for resistance and seems OK. I have not tried to create a spark with it under artificial test conditions yet (not sure what voltages to use) I'll have to assume it's OK for know.

The plug cap was broken, but after replacing it, even though the new one checks OK for resistance, I still have no spark.

Which brings me to the CDI:
After some search I realised that it is a Kokusan Denki model CU 2528 (not immediately obvious). If you cannot see the model number on yours, it's because it's painted on with white paint which fades off very easily. I certainly could not make mine out, although once I knew what it was supposed to be, I could see that the remnants of paint agreed. Great idea this, painting the model number on! Incidentally, it seems that this ignition has also been used for the Cagiva Canyon, remember Husky was under Cagiva ownership at the time. Might work for other bikes, I don't know... I believe the CDI for the non-elelctric start bikes is different, so check - it might not fit, even if it does say TE410/610.

This CDI is the flat module with 8 wires coming out of the bottom. These go to to diffent conectors, four each. The first (outer) four come from the pick-up. I will need some help with these, if I am to avoid removing the alternator case cover. The other (middle) four go out to the bike; as per the wiring diagram, they are Spark Signal (to coil), Earth, Rev Meter driver and Kill Switch. They should be fairly obvious - I might detail them more later.

OK, so far so good, now for the wires that come from the pick-up. These are:
Green, White, Red and Black (on the CDI and the pick-up),
and correspondingly
Yellow/Black, Blue/Black, Yellow and Blue/Orange (on the wiring loom).

What I need to know is which is which. Two will obviously be from some trigger loop for TDC detection, but what are the other two? Since there seems to be no source of power to the CDI, can I assume they are some form of mini generator?

If I want to test them, what signal should I expect? I have access to an oscilloscope, if need be. I can imagine what the trigger would be - I also think it works - since the rev meter jumps on every turn when you try to start the bike, just as it always did. I really don't want to spent a whole lot of money on a new "Husqvarna" CDI only to find it's not needed. Have you seen the prices??!! And how often does this kind of thing happen, anyway?

OTOH, I wouldn't mind spending a small amount on a Lifan 150 CDI, even just to try it out. Does anyone know how I would connect it? I think it has six connectors, so it probably won't have a rev-meter, maybe no kill-switch either? Any hints appreciated.

Why can't manufacturers document these things better? We're not dummies, but it's that much harder when you have so little to go on!

Thanks in anticipation, everybody, hope we can get this thing sorted out once and for all. It really shouldn't be such a stumbling block - no damn way.

 

[NoMad410]

I tested the pick-up leads today for resistance.
There was about 400Ω between Green and White, so I guess they must be the trigger. I'll check them out with the scope later.

If anyone could tell me what the Black and Red pair are, that would be great (excitor?) however there is no resistance between them, which surprised me, I'd have thought there would be a winding.

 

[NoMad410]

OK, I wired up the scope to it and this is what I got:

The White and Green seem to be the trigger, as i thought, they produce a very sharp spike, I can't quite make out the frequency but it seems about right - say two three times a second.

The Red and Black were the excitor, I get a wave of about 100V, looks all right, so I'm reasonably satisfied that the pick-up is working OK.

Both of the above were measured at the "input to the CDI" connector, so I'm happy with the wiring too.

Less happy with the output from the CDI, where I seem to have some sort of wave but not too steep, maxing at a couple of volts. I'd have expected at least a couple of hundred volts, with a very sharp profile/drop-off to the wave. So it appears that it may well be the CDI's fault. Can they fail just like that, for no reason? The bike was very gently ridden on the afternoon (just popped down to the shops), no shocks, jolts or jars, and put away as usual. Left for two days in usual conditions. Then on the third day, try to start - no spark.

Before I order a new CDI, Ι'd like to try get a spark from my system anyway, might order a cheapo Chinese one and check if the bike will trigger any spark with it.

Any other ideas?

 

[NoMad410]

I was talking on another forum with a TE410 kickstart owner, and I wonder if the CDI from that machine would do.
They seem to be wired more or less the same, and have the same colour wires coming out of the pickup (even if it is on the wrong side, and dry ). The only thing missing seems to be the rev-counter wire, but what the hell.

 

[NoMad410]

Well, the bike has been standing for almost a month now, and I am not a happy bunny. I think it will all get sorted out eventually, but I think it's a shame that such a nice bike isn't better supported.
And no, I don't think a 14-year old bike is history My road bike is almost 25-years old and is running like new.

The story, briefly so far:

One morning, for no apparent reason, the bike wouldn't start.
Checked various things: First, the NGK spark plug cap was broken - no conductivity - so I ordered a new one. That didn't fix it. I guess it was actualy working previoussly with the broken cap. Sparking coil seems fine.

Checked the pick-up continuity (when I finally found which one it was), and it seemed OK. 400Ω-ish.
Would it really have hurt the manual writers to let us know that the green and white leads are the pick-up and the red and black the excitor? FFS!

Checked the pick-up output on the oscilloscope, nice sharpish spike at, hm, around 5V, that seems fine too.
Checked the output of the exciter on the scope, just over 100V, raggedish-wavey waveform, that seems OK too. [But more on this below].

Checked the CDI output to the sparking coil, not good, no spike just a background noise at about 3V. On the basis of this I decided that the CDI must be gone. Ordered a new one, it behaves exactly the same, no spark to be seen. Getting a bit pissed off now.

Now, re the exciter: There is NO conductivity across these wires. Can that possibly be right? I expected it to be a coil! Of course, what do I know, it might be some obscure open circuit that works off some weird induction principle, but that just don't seem right to me.
What confuses me is that on the scope it shows an acceptable wave when cranked over (> 100V, as I said), but could that just be the high resistance of the scope, picking up some "shadow effect"? I really need some help on this one!

Anyway, I now take the LH (alternator) cover off, to see if I can make any more sense of it. One of the pick-up wires dislodges in the process, so I have to remove and re-solder! Naturally, I now need the very expensive Husky tool to put it back in place.... sigh. At least I'll get to have a good look at the coil windings today. Ho-hum.

I suppose at this stage I could have just cut my losses, and maybe sold the bike for parts, but (a) I like its ride, and (b) I have this weird belief system that involves self-reliance and inventiveness... maybe I should go for throw-away consumerism instead more fun that way.

 

[NoMad410]

I will re-iterate my main question at this point. Sorry if I'm getting boring, but I really need to understand this!

Is there ANY way that the exciter wires (red and black) are not connected to a coil, but are open??? That is, zero resistance on the ohm-meter?
I am perplexed, because, on the oscilloscope it produces a wave when cranked over, seems about right, over 100V, jaggedy waveform.

On the voltmeter, on the other hand, I get no reading at all, which makes me wonder if the high sensitivity of the scope is fooling me.

Any help with this very much appreciated.

Don't really want to fork out over €350 to get a new stator assembly from husqyparts. I've already spent much more than I wanted to. Plus, at the end of it I'd be no wiser as to what is going on.

 

[NoMad410]

OK, in the absence of any advice to the contrary (or otherwise ) I will assume I have a burnt out exciter coil and I will open it up and fix/rewind it. It doesn't seem TOO difficult, although I am treading very carefully, as I've never done this sort of thing before.

If it's not reparable and needs a full rewind, then I'm probably OK to find the right gauge of wire, although chances are I'll need some help with the adhesives and bindings that are use to hold the whole thing together.

Here's what the stator looks like at the moment.

There was quite a lot of orange crud about, I think that was from the initial resin used to hold it in place, it doesn't look too appetizing, I've cleared out as many bits as I could see. I'd also like to know what that thing which looks like a perished rubber band is - the one that is holding the sleeved solder joins.

The exciter coil is the one on the isolated post, wrapped in some kind of black tape.
Other than that everything seems fine, the alternator (three-phase) windings are all giving resistance values I'd expect. They look like they've been wound by an uncoordinated Neanderthal on too much caffeine, but hey, they work, and I just want my bike going again - not doing a restoration.

Another thing that is in quite a state is the grommet going into the the case, actually it looked pretty bad right from when I got the bike, and is probably leaking ever so slightly (acceptable level). This time I'll probably just bodge a solution with rubber sealant of some sort, but I'd like to find a better and prettier long-term solution. Obviously, this is not the sort of thing you can just order - which, again - I find quite unacceptable. (It's bent in the photo to highlight damage.)

It seems to me that servicing the stator is not beyond the average DIYer, companies shouldn't make it more difficult than need be.

 

[NoMad410]

I am now a little bit past the - "Huh?!! WTF is going on here? " stage and I am starting to get a glimpse of what this problem may require .

Disassembling the stator, the first thing I notice is that it is highly unlikely that this is an OEM winding. Gawdhelpus if it is. So the bike has most probably had stator problems before. The popular view is that stators / alternators are a bit of trouble spot for thumpers, especially off-road ones.
Does this really need to be the case?

As I already stated, the three-phase (battery charging) windings, even though looking extremely crude are all right. It is the single post high-voltage excitor winding that is blown. Apparently even presence of humidity can cause these little buggers to go unpredicatably tits-up.

I tried to separate the windings in the stator so that I could isolate the high-voltage coil, but the leads to them have all been epoxied into place in such a way that if I persisted I'd destroy the low-voltage windings too, which I obviously don't want to do. I ended up cutting away most of the high-voltage wire and leaving the epoxied sections in place, rather than tear up everything and have to rewind the lot. At the end of it I checked all the low-voltage stuff again and it seems fine.

Here's the stator at the moment:

You can see the red and black wires coming out of the connection cable. These went to the black (right) and green (left) braided wire connection of the high-voltage coil (middle on isolated post). This is the one that is burnt out and needs to be replaced/rewound.

At this stage I have several thoughts going through my head -

 

[NoMad410]

The main thought is that there must be an easy way to do this.
I mean besides forking out €300+ for a new stator that may present exactly the same problem in a while.

Searching through the Web I see that surprisingly few people have tackled this problem, although there have been some interesting one-off results. Almost nothing with Huskies, unfortunately, but there are some common principles that should apply everywhere.

I found only one brave chap who rewound his excitor coil. Took him some time but it worked. Not so sure it will be that easy for me, everything seems pretty ridiculously epoxied into place. This is not meant to be a serviceable component, it seems.

Another interesting project was a fellow who got so fed up of burning out Yamaha XT CDIs that he designed his own. The design evolved quite a bit, from analog to digital/programmable, but all these designs required some excitor voltage. However, the last design he put out (and here's the good bit) is independent of the excitor, and uses the battery voltage to charge the CDI. If I give up on rewinding, I may well try this route.

One of the more interesting ones was a chap who, realising that the excitor simply gave an AC to the CDI, used a Christmas-tree light transformer, in reverse, to supply the excitor current from the battery charging windings! Brilliant lateral thinking, and it worked! The great thing about this approach was that he was not tied to a battery... This might well work on the Husky, although I am not sure how to approach the transforming of three-phase voltage.

So, as far as I can see, so long as SOMETHING powers the CDI sufficiently to allow the capacitor to charge, we should be able to get a spark.

Even an inverter powered from the battery should do the job, although I am not sure of how this will cope with with higher charging/sparking demands at higher revs. I am also loath to just plug any random AC into the Husky CDI as I do not want to blow it up. I'd feel easier if I had some spec sheet with the CDI's power requirements and safety limits but no such luck...
I may end up gradually feeding some low current AC in through a triac or variac if all else fails.

Anyways, the point is that there seem to be quite a lot of interesting alternatives that could eliminate the "hostage to ignition" problem, see how it goes. It also seems that losing the excitor coil altogether might not be the worst idea in the world.

 

[NoMad410]

I've removed the excitor coil. This one was wound properly, not like the caveman low-voltage windings and it was probably OEM, or professional, anyway.

The process of "unwinding"/removal was neither elegant nor pretty. As I looked on the mass of hellish, fine copper minge, I knew there is no way Jose I am rewinding that. It took me an age just to remove it! Damn.
It's going to have to be Plan B, then, an electronic ignition solution that relies on the batteries. Ho-hum.

 

[NoMad410]

Hang on lads, I've had a great idea...

There's three free posts at the top of the stator, right?
OK, they're not all the same, and the two side ones only have half a covering top, but let's not choke on details.
What if I put some nice thick gauge windings on those? I should have a second low-voltage supply then. at perhaps a bit less than a third the power of the main three-phase one (I'm guessing here).
What if I took this low-voltage and transformed it using the christmas-lights transformer idea, mentioned two posts above?
It would be independent of the main supply, so no messing with the three-phase, or potential damage to rectifier, etc.
It would be reasonably easy to do: Perhaps 100 windings per post, not NN thousand as with the excitor.
The transformer should step it up enough to charge a CDI capacitor.
Best of all, the PITA work of fine winding has already been done by the transformer manufacturer.
Cheap and not much to lose, even if it doesn't work.

Yup, go for that.
And for the first run I'll use a simple inexpensive Public Domain CDI, don't want to risk blowing the Husky one.
Back in a couple of days...

 

[scudster]

Fair dues to ya Nomad!
I bought a 99 TE610e a few months ago as a non-runner. It had no spark too. I too checked all the major components and could find nothing amiss. There was some gunthering done to the pickup coil so I replaced it. Still no spark. The coil came without telling me which wire joins with which CDI wire so I reversed the poles. Would you believe thats what gave me spark!
I'm not saying that this is your solution or anything but it goes to show a small mistake can cause a big headache further down the line.
I whole heartedly agree that the electrics on these machines aren't near as robust as their Japanese counterparts.

 

[NoMad410]

Been a bit busy with other stuff, back on course now.

I thought a little bit about the new winding setup for the excitor (plus transformer): The wire in the original (OEM?) winding was about 0.12 mm thick. Very difficult to tell how many winding loops, I estimate 2000-4000. I decided to use 0.5 mm transformer wire to rewind it. It is quite thick enough that it shouldn't suffer burnout problems except in the most ridiculous situations and easy-ish to work with. It should end up with about 1/10th to 1/20th the number of windings on the original (very ballpark figure). 10x and 20x transformers for the final stepup should be easily available.
Incidentally, this will also be about 4 times the number of windings on the low voltage posts, that wire being 1.2mm, iirc, so there should be some power.

I have also decided only to wind the middle post. Initially I was planning to wind all three, then saw that the gains would be killed by the phase-shifts, considered two, as I think I could get 1.7x the voltage of one if the windings were reversed (this is my back of the envelope calculations, assuming I understand the magnet arrangement correctly, I could be completely wrong ). In the end I decided to keep it simple and only wind the centre post, figuring it would provide enough power to drive everything, if it was enough with the initial arrangement. (Hey, if I'm wrong it's back to plan A.)

I wound the whole thing hoping to do a nice job, not like the Neanderthal work on the low-voltage coils. The initial few layers went really well, very neat, but by the time I reached the last ones, most hope of neatness was gone, and the caveman look was back. I did use transformer tape, which probably helped initially, but I just haven't had enough experience with this kind of work. Never mind, all wound and sound anyway, if not too pretty. A bit of epoxy to hold everything in place, and voila:

Lost count of the winding in the mad flurry, but I reckon about six layers, at around 30 windings per layer, so it should be somewhere near 200 loops in total. I am struggling a bit to guess what voltage this will produce, we'll have to wait and see. My first guesses (don't ask to see my working) were around 20V, but it could be anything, really.

Assembled the whole thing (took the opportunity to photocopy the exact dimensions of the alternator gasket, so easy to cut them out of gasket paper oneself, why rely on aftermarket?) Reconnected the pick-up that had broken away loose. Heat-resistant covering, heat-shrink tube and epoxy were used to keep it all sort of tidy. Sorry, I forgot to photograph just before reassembly, it almost looked neat, you'll have to take my word on it.

I'll be testing it tomorrow, initially I just took a reading on the multimeter to see if it produces anything at all. It showed about 2V AC average when cranked, which I think is OK, since that's over 1/12 of the cycle (i.e. only one of the 12 posts) so I reckon peak should be about ten times that. Will check with oscilloscope later.

Fingers crossed.


====

PS. Here's a pic I took subsequently, when I opened it to check that everything was OK:

 

[NoMad410]

I then cut the excitor wires (red and black) coming from the alternator and attached bullet terminals to all of the ends. This is where the tap for the transformer will be.

The transformer I ordered was a 240<->24+24. This way I thought I can get both a 10x step-up and a 5x step-up, and use whichever works best. I am also concerned not to overload the CDI initially, don't want to burn it out through too much voltage. I know they're pretty robust, but still...

Then it occurred to me that the voltage is probably NOT averaged over 1/12th of the cycle, since there are 4 magnet sets, but only over a third. So my voltage peak is more likely to be in the 6V vicinity rather than the 20V. Will a 10x step-up be enough? Oooops, probably not. For good measure I ordered a 240<->6+6 transformer as well. They're not expensive (£5) and it will give me more options to play with if they're needed.

The first transformer arrived and was put into place for testing using a breadboard for convenience of connection. Here it is ready to test:

Try cranking it over and... no spark. Bleah!!
However, I measured the voltage on a cheapo meter* and it was indeed 20V (indicated) from 2V, 10x step-up, so something is happening right. I am hoping this will mean a peak in the region of 60V, but I can't be arsed to drag out and set up the oscilloscope. Getting a bit jaded. I'll wait for the other transformer now.

-------

* PS. I should add that the cheapo multimeter which I used for measuring the output was of the analog type; i.e. the old ones with an actual needle, you know. For some reason, the digital one just did not register on the Volts-AC scale, maybe a high quality one would, I don't know. There's still a place for good old analog, it seems.

 

[NoMad410]

Second transformer arrived today, and what with the long daylight hours and the pause in the rain, I decided to try it out this evening. First, I tried using 240V <-> 12V, i.e only a 20 times step-up, for which I'm estimating a peak of 120V, that should be enough. I am still not bothering with the oscilloscope.
No spark. Ho-hum.
There is an indicated voltage of about 50, though, come on baby, one more time!

I then tried with the 40x step-up (6V->240V) and blimey, guess what? Nice fat spark! Bit reddish, not too blue, but fucket, that's a nice steady spark when cranked over. OK, I'll admit a slight feeling of relief. What's more, I checked it once more with the 20x step-up, and this too was giving a spark, not very strong, not as noticeable (maybe that's why I missed it first time), but a spark nonethelesss. OK, we may be in business!

Assemble spark plug into engine, fuel tank on, and let's try.
As Sod's Law would have it there was a blip at this stage, which I won't bore you with.
After cuppa tea to cool steam coming from my ears, I try again.

Three crank-overs on choke, RR..RR..Rumble, and it's going.
Like it had never stopped!
Starts just as it always does. (and that's only the 20x step-up). Rev it a few times, absolutely normal!
You could have knocked me down with a feather.
Switch it off, start it again, no problem. Yee-hah!
OK, put it to bed now, I got other things to do, will assemble everything properly tomorrow.
I need to make some decisions about what step-up settings I'll use, plus rig up a final, encapsulated, waterproof home for the transformer.

All this faffing, and the MOT has expired too. (UK roadworthy certificate, for stateside folk)
But I'm really glad it all seems fine. I was starting to look at new alternators. And what's more, this design is probably way more robust than the original, I don't see the newly-wound coil burning out, and if the transformer does, hey, 5 minute job to replace it!

My thanks and respect to the unknown guy who originally had the transformer idea which I adapted for this.

 

[NoMad410]

Just found out that the fellow who originally posted this idea used the handle JarkkoT, and did it on a 1983 XR500R. That forum would also suggest that he messed around with TL1000 electrics.

If you see this, Jarkko, CHEERS!

 

[NoMad410]

OK, just a couple of minor things more:
Bike is now running fine, and though I have not yet run it for more than 10 minutes without stopping. I feel reasonably confident it will continue to do so in normal use. I will report again a bit later, just to finalise the thread, and also show the final transformer arrangement (strapped to the sub frame just behind the rear-shock reservoir seems nice and convenient. I'll put up a pic when done.

Minor thing 1:
It surprised me that the transformer had a polarity issue! I'd have thought with AC it wouldn't have made much difference either way, but I found that I had to match the phases of the input and output wires. I.e., if I used left-input to black I had to use left-output to black too (IOW, right-input to red, right-output to red, too). I could flip them BOTH and it would still work, but I could not cross them so to speak, it just refused to spark. This caused some real diagnostic headaches in the beginning, and, to be honest I still don't fully understand why, but I accept it. So the black and red leads aren't just decorations, polarity (somehow) matters. I'd be very grateful to anyone who can explain to me why this is so.

Minor thing 2:
The 1.8VA transformer which I was initially using got pretty damn hot, pretty damn quick. Not on idle, but anywhere over 3k revs.
(I am actually quite in wonder of how much power those few silly windings will produce when whirring past the magnets! Isn't the universe marvelous?! Isn't this transformer fυckin' hot?

)
I could see it would not have much future, so I replaced it with a 18VA one, which stays nice and cool.

Minor thing 3: In the end, I went with a 9V to 240V step-up. 6V was working fine but obviously quite so much step-up was not needed. 12V worked OK too, but at very low revs or idling you could hear it miss the occasional spark, so it was probably on the limit and perhaps not quite managing to charge sometimes.

OK that's all for now folks, hope it kept you entertained, I'll put up a final piccy when all is going and on the road...

 

[NoMad410]

Well, I have now ridden the bike for a couple of hundred miles and I am pleased to report that the system works very well and reliably, starts on the button and runs as well as it ever did! I am keeping it, and probably won't even bother replacing it with an OEM one, which α) costs a bit and β) might burn out again (bit of a design weakness, methinks)

Here is where the transformer is located at present, I haven't yet cased it in epoxy, the weather is good LOL, and I needed to test it.

I will move it up to the front, just under the CDI unit, on a specially made little bracket - that way, I won't need to take a tap from the red&black (excitor) cables of the stator, and also it will plug in using proper Husky CDI connectors, which I have managed to find. For anyone wondering, since we're never told such things, these are Sumitomo type connectors B4APT and its match B4AS, which are available from corsa-technic. Hope that helps someone...

 

[lochenjons]

What type of epoxy or coil potting did you use?