CT70 Lighting Coil Exposed

racerx said:

This is the crux of the entire issue, imho. They did intend the system to bias the output toward the HL circuit. (Which begs the question: "what would happen if the yellow & white leads were swapped at the wire harness-to-stator plug?"...I expect that would only "free up" 2-5W max.) From an OEM perspective, it made sense...cheap and effective...for a while.

Click to expand...

Honestly, never give it any thought until today. I received a diagram showing the coil that is used on the Mitsu CDI kit. Lo and behold, there is another tap, which was cut off but yet long enough for me solder a wire to. After making a few checks, I thought it would work. While it is much much better and works (somewhat), it did not solve my problem. Lighting voltage is a little too high and charging is a little low. Can't get but about a 1 amp and that's at 8000RPM. Swapped the yellow and green and saw basically the same. For sure, both coils are identical.

This coil has 3 windings, similar to an OEM Mitsu coil, but the wire size and calculated length of wire per coil is a little different than anything I've seen before. close but no cigar yet. Not sure how much longer I'll tinker with it before I slap in a OEM Mitsu coil and make sure the CDI works as expected on the bike.

By the way, I saw a prototype of a 12v kit with CDI from who I think is the designer of all the kits you see on e-Bay. 3 coils: 1 for CDI and 2 for lighting/charging. Yep, 3 coils crammed on a single 6v stator plate. No further details, but I would assume a custom Rec/Reg or just use 2 legs of a 3 phase Rec/Reg.

Success. The lighting coil now works as it should, and in my opinion, a little better than the OEM Mitsu coil. I was able to change which tap was grounded, which in effect added a few more turns to the charging circuit. This brought the lighting voltage down to about 6.5v, which is about 0.75v higher than an Mitsu coil. Charging is fine now and almost spot on to a standard Mitsu. Will get it soldered up and installed soon.

Just a question to others that have installed the CDI kit. Where did you mount the CDI box? I don't want it just dangling inside the frame and I'm not going to drill a hole either. Thinking about lengthening the wiring harness and amounting it somewhere/somehow near the battery box. Just seeing how others have done it before I overthink it.

I double side taped it in under the gas tank in the frame. Glad to hear it works now.

Thanks Pat. It is going to be a little chilly here tomorrow so maybe I'll find the time to get it installed.

Long story short, learning a lot about the CDI kit.

It's easy enough adding a short bracket to the battery box, along the same lines as Honda did with the turn signal-equipped bikes. Old inner tubes are a handy source of industrial-strength rubber bands. With contact trim adhesive, make `em any length you like. Otherwise, zip-ties are very convenient.

I can now confirm the Mitsu kit on an H with the timing advance intact seems to work just fine. But to clarify, I only ran it for about a minute to verify it would rev up and the lights work. It's a bit too chilly today to take it for a spin and do a real test. I played around with various locations to mount the CDI box. Space is limited on a K0 or HK0 (my case) because of the size of battery box. Under the tank as OLDCT did will work and is probably the easiest. I opted to use my wire stretcher to lengthen the wire and mount it beside of the battery box. I may make a bracket or use a section of inner tube as mentioned by racerx. Not sure yet, but I have plenty of time to sort it out.

Ignore the rats nest. I just wanted to make sure it fits. I have not buttoned up everything yet. Who knows; someone else may suggest a better place anyway!

It's a lot of olives to stuff into a small jar, isn't it?

I can post a pic of the underseat area of my red Dax. There's a full wave reg/rec, lighting control module, and Honda Nice CDI module...which is almost the volume of a pack of cigarettes. The 3.5L tank extends about an inch further into the battery location and the battery is twice the size of a stock K0. It is, literally, the physical limit of what will fit. My point is that you have more room than you probably realize. If your battery will fit in the same space as a stock OE leaker, just swapping in a late K0 -K2 battery box is an easy way to add breathing room. A custom battery carrier that moves the battery closer to the tank would open things up even more.

With tight real estate and small components opening up just a little additional space can make a huge difference. Take a critical look and imagine moving the battery an inch forward. Better still...remove the battery box then hold the battery against the rear of the tank...to visualize what is possible. I think you will be pleasantly surprised.

Well I think I'm going to call it done. I definitely could have drawn it faster than I built it. lol. But I'm glad I decided to go (almost) all in, with it. Hopefully this will be useful enough to be worth the space it takes up in your shop. And don't be afraid to modify it any way you like. I tried to build it in a way that it'll come apart easily for changes...barely any glue at all.

That looks nice. Much better setup than the shaft I lock in the vise and pull & wind by hand! Is that oak tongue and groove flooring?

That is just some nice oak boards that I had squirreled away. It probably used to be a pallet, or cribbing of some sort. I cut the tongue and groove...kinda like a drawer slide setup. Either side can slide, or be screwed in place. The strap is to put your wrist thru so the slide will follow your hand. I'll get it shipped in a day or two. Then you can mess with it when you get back into your experimenting.

That is neat! Nice job on that.

Thanks OLD CT. I had fun building it, and tinkering with it. It's always fun to build something new. After I was pretty much done, I searched for something similar online. Was happy NOT to find anything similar, with my quick look. I was afraid one might pop up for 20 bucks...lol

Ran across something unusual that's worth sharing. The CDI kits out of Malaysia really has my curiosity up, which I'll go into later. Anyway, I decided to use real lamps to test the CDI stator. Since I already had a standard oem Mitsu mounted up and had data to compare back to, I opted to start here. With a lamp used instead of fixed resistors, charging current was way too low and HL voltage was low as well. No charging at all, and HL voltage would not go over ~5volts AC. However, the light looked about right. I first thought I made an error somewhere or something had changed. That turn out to not be case a couple hours later.

Long story short. The generic lamps I purchased are marked "6V 15/15W" but are not 15 watt lamps even though they're marked correctly. They are 25 watt bulbs. At 5 volts AC, it was drawing 3.1 amps vs. Eiko or Stanley which are spec'd at 2.3 amps. @ 6.5v. At rated voltage, the lamps I have consume 3.8 amps @ 6.5v. That is way more than you can get from the AC winding of the stator.

In a nut shell, the bulb was overloading the AC (yellow) winding and not producing enough juice on the green winding to even charge a hearing aid battery. Blew me away for an hour or so. Just another data point to add to the collection about the circuit has to be somewhat balanced or everything gets out of whack.

Another tid bit. It appears that Eiko is no longer making this lamp (A3598H). However, they are still available from Honda and there are some NOS Stanley bulbs on e-Bay. I can be a tight wad (Honda bulb is $12) and this is one time I got bit by going cheap.

allenp42 said:

Ran across something unusual that's worth sharing. The CDI kits out of Malaysia really has my curiosity up, which I'll go into later. Anyway, I decided to use real lamps to test the CDI stator. Since I already had a standard oem Mitsu mounted up and had data to compare back to, I opted to start here. With a lamp used instead of fixed resistors, charging current was way too low and HL voltage was low as well. No charging at all, and HL voltage would not go over ~5volts AC. However, the light looked about right. I first thought I made an error somewhere or something had changed. That turn out to not be case a couple hours later.

Long story short. The generic lamps I purchased are marked "6V 15/15W" but are not 15 watt lamps even though they're marked correctly. They are 25 watt bulbs. At 5 volts AC, it was drawing 3.1 amps vs. Eiko or Stanley which are spec'd at 2.3 amps. @ 6.5v. At rated voltage, the lamps I have consume 3.8 amps @ 6.5v. That is way more than you can get from the AC winding of the stator.

In a nut shell, the bulb was overloading the AC (yellow) winding and not producing enough juice on the green winding to even charge a hearing aid battery. Blew me away for an hour or so. Just another data point to add to the collection about the circuit has to be somewhat balanced or everything gets out of whack.

Another tid bit. It appears that Eiko is no longer making this lamp (A3598H). However, they are still available from Honda and there are some NOS Stanley bulbs on e-Bay. I can be a tight wad (Honda bulb is $12) and this is one time I got bit by going cheap.

Click to expand...

This is an outstanding, real-world, example of what I've been talking about with respect to the increasing difficulties of keeping a "balanced" electrical system balanced...50 years later. As the market for these obsolete items dwindles, there's less and less motivating the remaining manufacturers & suppliers to stay on top of quality. We've already seen ominous signs with ignition condensers; we rarely ever saw one that failed...until recent years. With bulbs, it's a double-whammy...not only is quality down, but look at the sloppy tolerances. This is why I've come to seriously dislike this type of "balanced" electrical system. How does one maintain that balance between electrical load and alternator output when bulb draw varies wildly from spec (i.e. the nominal value stamped onto the bulb base)? This accounts for a lot of problems with turn signal flash rate, where current draw has to be just right, to heat the bimetal switch at the right rate.

racerx said:

As the market for these obsolete items dwindles, there's less and less motivating the remaining manufacturers & suppliers to stay on top of quality.

Click to expand...

Isn't that the truth. Already hit the issue with cheap points and condensers that don't fit very well....but a bulb? For sure, it had me second guessing everything for a couple of hours.

Right after the bulb fiasco, was testing a stator that I repaired for a member here...and no spark. Coil checked good with my Fluke Meter. New OEM Honda points and condenser did not fix it. So I checked the coil with a precision meter I have. Resistance was low by ~ 0.1 ohms. Since I had a file where I had checked several, I knew it was problem. Not a chance to catch that with a good DVM. Popped in a used coil and it worked fine. That is the 2nd one I have encountered. Was not my day yesterday.

allenp42 said:

Popped in a used coil and it worked fine. That is the 2nd one I have encountered.

Click to expand...

Are you talking about the ht/ignition coil, or the primary coil??
I'm thinking it must be the primary coil. I remember you posting about a bad one that appeared to be good.

That DOES suck about the aftermarket bulb. Sucks that you can't take them for granted. And that I'll be afraid to even buy/use them now. Do you think there is a way to check them...before they're installed...to know if they are actually 15 watt?? Maybe checking resistance?
How could I check the ones that I have. Compare brightness against a Honda bulb?

kirrbby said:

Are you talking about the ht/ignition coil, or the primary coil??

Click to expand...

The primary coil in the stator

kirrbby said:

Do you think there is a way to check them...before they're installed...to know if they are actually 15 watt?? Maybe checking resistance?

Click to expand...

This may sound a bit a odd, but normally I just check for "some resistance" or an open circuit (infinity). When you use a meter to check a lamp filament, you're basically checking the COLD DC resistance of the tungsten filament. AFAIK, most incandescent bulbs have a tungsten filament. When juice first hits the filament it heats up and the resistance increases. Bottom line, you can't use the static values you read, plug it into ohms law and come up with the wattage. Also, don't think I have ever looked closely at the absolute DC resistance of a lamp before, or even given it much thought, other than to see if it was good (not open). It's a story for another day, but impedance (AC resistance) can enter the picture for some devices. I would rather not go down this road for a basic lamp.

For what it's worth, I did connect this lamp to a variable DC Supply. That's how I determined the current draw of 3.8 amps at 6.5 volts and was able to calculate the wattage from the 2 measurements. 3.8 x 6.5 = 24.7 watts.

How to check? Well, off hand the only method that comes to mind is to check the current draw when connected directly to a battery. Put the meter in series with the lamp, set to a current on a scale that can handle up to 3 amps, and see if it is anywhere near rated spec., which can be easily found by running a google search or the bulb number.

I have a box of 10 Stanley bulbs on the way. Doubt I will have to deal with this issue again. I have been skeptical of aftermarket points and condensers for about a year now....ever since the Daiichi versions dried up. Aftermarket bulbs has joined the list.

I just found a spec sheet for the 25w version. I was close on what I measured, and it is still available from Eiko

I scored a lightly used Hitachi CDI kit that came off a K0. Some guy was restoring his bike to bone stock and took it off. Price was reasonable so I rolled the dice.

Now, I did not order it or spec it, but this one has timing advance. It works on a K0 flywheel with 3 speed crank or HK0 flywheel with H style crank (w/o removing the timing advance mechanism). I plan on triple checking just to make sure nothing is close to hitting the coils, but for sure, nothing jumped out. Did not see any crumbs when I removed the flywheel either. Now this is a test 1 of 1, but I see no reason why the KO has to be any different than a HK0 version, other than timing advance in the CDI module.

Like the Mitsu version I tested, HL voltage seems high to me (8.5-9v right at the stator), and I can't reconfigure this coil as I did on the Mitsu. It is what it is. Now this really has my curiosity up. Pat's HL looked good. He's not popping bulbs, read lots of positive comments about the HL being much brighter, so there is something I'm missing. And that something has to be internal resistance of the stator plug, wiring, ignition switch, HL switch, bullet connectors, etc. In other words, the entire loop from stator to bulb to ground. I suspect these guys are taking this into account because they are doing a lot of their testing right on the bike.

Yep, I measured the loop on my HK0. Once I get a known good 15 watt bulb in my hands, going to test this again. And I'll go out on a limb, I doubt I'll see much more than 7 to 7.5v right at the bulb.

Another tip bit as it relates to the Mitsubishi stator only. The primary coil is grounded through a few turns in the lighting coil. If you look at Kirby's post, you'll see there is a short pig tail that comes out of the lighting coil and goes to the primary coil.

See post #9
https://lilhonda.com/index.php?threads/stators.21608/#post-170892

Never understood why it was done this way and still don't. But I do know if the wire going into the lighting coil get's pinched or breaks, and you can just ground the primary coil like is done on a Hitachi Pri coil and both coils seem to work just fine. I can't see any difference from the lighting coil and I have a spark. Maybe one day someone will figure it out. But for now, I am going to move on to the next problem.

kirrbby said:

That DOES suck about the aftermarket bulb. Sucks that you can't take them for granted. And that I'll be afraid to even buy/use them now. Do you think there is a way to check them...before they're installed...to know if they are actually 15 watt?? Maybe checking resistance?
How could I check the ones that I have. Compare brightness against a Honda bulb?

Click to expand...

You're making my point for me. Frankly, I saw this problem on the horizon years ago and it is only going to get worse. That's a substantial reason why I recommend full wave. Once you can power everything from the battery, like a "real" road vehicle, enough options open-up that dodgy-quality bulbs can simply be replaced with something that works. Going to 12v flavor gives you access to the part of the lighting market that's not just growing but exploding. IOW, better to invest the resources up-front in system improvements the allow you to sidestep dodgy quality parts, i.e. bulbs, flasher relays, altogether. IMO, the major stumbling block is, as usual, the up-front cost.