ok, first thank you to all for this forum ive read hundreds of posts and greatly appreciate all of the shared information.

so 3 weeks ago i purchased a 2005 arcdtic cat 90 which is similar to a kawasaki 90 and is based on the kymco mongoose 90. After the initial test drive eveything appeared to be in order and i brought it home and my son ride it for approx 2 minutes and a loud pop and the quad died, further investigation revealed that a capacitor exploded, blowing a hole inthe cdi.

Basic testing with a dmm revealed the quad used a dc cdi, so throwing the blown cdi up to fate I purchased a new cdi and all was good for 12 to 14 hours of riding. Tonight as my son rode past us I noticed it was running rough and he stopped to tell me it sounded funny, I lifted the back seat and to my dissapointment the cdi box has a bulge in it where it appears another capacitor blew.
I know that there is a reason this is happening, I just need some help pointing me in the right direction.

 

We3lcome to the forums.
Hopefully Lynn Edwards can help.

 

Here is a pic of the current cdi box. The previous cdi box actually blew a hole out the bottom. Unfortunately I threw it away before taking a pic.

 

I don't know anything about Arctic Cats. I went looking for a wiring diagram for your quad and came up empty. So the following is based on generic DC powered CDI operation rather than anything specific to your Arctic Cat:

A CDI has power in (DC CDIs run on 12 volts DC), a low voltage and low power trigger signal to fire the CDI, the ignition coil primary drive output, and ground. That is four wires. Is this what you have? Your picture didn't show the wires. What are the wire colors on your CDI module? If the CDI has short pigtail wires going into the wiring harness through a connector, then report the colors on the "wiring harness" side of any connector if they are different from the pigtail wires.

How sure are you that this is a DC powered CDI?

I've been thinking about possible failure modes on DC CDI. To blow a hole in the CDI takes a fair amount of power. To melt a CDI also takes a fair amount of power. [I'm thinking out loud here as I go along...]

Is the "bulge" in your pic the same place where the "hole" appeared in your last CDI?

The only thing that can provide "power" enough to melt and/or explode things has to come from 12 volts of the battery/battery charging system. This is supposed to be between 12 to 14 volts all the time, but what if it isn't? Suppose the battery charging system went full tilt toward overcharging the battery? In this case with the engine running at speed the battery voltage would climb way high - to as much as 22 volts with the headlights off. I don't what this would do to an arctic cat CDI, but perhaps it could make it overheat and fail. This battery voltage on an engine running at speed would be easy to measure except you have an engine that doesn't run now . So if you regulator is going ***** to the wall and overcharging the battery (raising it's voltage way high) then you would see other problems - like water is disappearing out of your battery cells after every ride? Do you see this? This would also be followed by a ruined battery after a dozen or so rides. Over charging a battery wrecks it over time.

A little less likely but still quite believable scenario is that you had a random failure on the first CDI, and you bought a bad second CDI. The fact that it ran 10 hours or so means that it would have passed any simple test regime at the factory. They plug it in and it works, so it's good right? But there are a zillion things that can be wrong internally that cause it to run too hot and fail over time. Wrong resistor values, backwards tantalum capacitors, poor heat sinking to the base plate inside that conducts heat to the outside world, etc. This kind of infant mortality has been a very common problem on chinese CDIs where CDIs cost 9 bucks and there isn't much margin for any sort of thorough testing (one poster reported five failures in a row from the same vendor). I don't know where Arctic Cat gets their CDIs or what kind of testing they go through. SO there we are...

Even more unlikely, but still remotely possible, is that there is a problem in the ignition coil where it is breaking down at high engine speed, and the high voltage that is supposed to feed the spark plug jumps instead from the ignition coil secondary back to the primary winding (instead of ground which is a lot more common). This arc goes directly into the CDI output which takes it out. But this is very unlikely based on your reported symptoms. You reported that the quad ran rough and then failed a while later. High voltage arc's tend to kill things like semiconductors instantly, silenty, and permanently. It is possible for something inside to be killed instantly, which causes something else to heat up and "bulge out" over time. But your quad would not be running during this transisiton period. It would be dead as a stone.

One more thing to check: You do have a fuse on the battery, right? What is the value of that fuse? In all of the above I'm making the assumption that instantly available power is limited by the main fuse, so any very rapid and catastrophic heating cannot occur. If it did it would simply blow the fuse as long as the fuse circuitry is in place, and the fuse value is kosher. So I'm thinking of a much slower buildup of heat is at the root of your problem.

 

First thank you to Lynn and lt80 for your replies.
although labeled and sold as an Arctic cat, it is manufactured by kymco with an adly gy-6.



I determined it was a DC cdi by checking voltage at the ignition pin location of the cdi harness.
The wiring at the cdi harness is
Pulse blue/yellow
To coil black/yellow
The two grounds black
Kill switch black/white
DC ignition green/white

here is a pic of the wires from the stator tying into the harness

The stator harness its on the left tying into the body harness on the right.
Yes, the battery is fused at the stock location, with a 10 amp fuse.
The old cdi appeared to be the OEM. The new cdi is aftermarket and much smaller, almost the same size as an AC fired cdi, the manufacturer claims advances in circuitry design allow for a more compact design.
So, I cannot compare location of the failure between the two.

I have ordered another cdi from a different vendor.
The battery is new and reads 13.1 @ rest, I will check the voltage during operation when I receive the new cdi and report back. As of right now the battery appears fine and not swollen.





AsIts almost a small as

 

Now that I know that it is an aftermarket CDI, I think that a second bad CDI is the most likely scenario. Adn buying from another sources makes a lot of sense .

Please let us know how it turns out...

 

I received the new cdi boxes Friday.
Voltage at rest was still 13.1.
Voltage on the battery terminals at idle was unchanged and climbed to 14.3 @ higher rpms.

He's got five to six hours of riding this weekend on it so far so hopefully all stays well.

If it happens again I'll post back. Thanks

 

I'm glad it is running . From what you post it sounds like everything is working just fine. Do post back if it fails again...

 

Well unfortunately I'm back.



As of right now the battery is partially drained and on a charger, I have another cdi ready to go in. I'm stumped. DMM in hand. I'm gonna pull the plastics do I can easily follow the harness.

 

So how many hours of run time before it failed again? You reported 5 or 6 hours or run time before your last post. I'm just wondering how much more time ir had before it failed again.

The failure (melted) area does seem to be the same place...

I'm curious as to why a GY6 DC powered aftermarket CDI is orange. I've never seen that on generic GY6 CDIs, yet you report two different vendors with the same weird orange color. You said your engine was an "adly" GY6. I'm not familiar with this at all. What is the difference between an adly GY6 versus a generic chinese clone GY6?

On a DC powered CDI you have 12 volts power going in, a small micro power trigger signal going in, possibly a kill switch signal, and the output to the ignition coil. If the inputs are correct and if the output is correct (i.e. the quad runs), then any failure has to be due to a design or manufacturing error. The melted case indicates heat. That could be due to the 'failure' happening first and then things start overheating, or things are overheating followed by a latent failure. The latter is the most likely in my opinion. Ther are a zillion things that can be done wrong in the manufacturing process that can lead to early failure. Poor heat sinking of the heat producing parts, wrong resistor values that cause transistors to turn off too slowly, tantalum capacitors that are installed backwards, just to name a few.

If your DC input voltage to the CDI is not going way high, then I still think that a bad CDI is to blame. And all your CDIs are orange... I'm still wondering if your CDIs are coming from the same manufacturing source (even if from different vendors), or is there some special requirement for Arctic Cat quads that require a special CDI which they differentiate with an orange box... I don't know.

You also mentioned that it was a "capacitor" that blew. Did you really verify this by digging into the CDI potting, or is it a guess? I'm just fishing for information here since I don't know yet what is going on. I would guess a transistor to be the most likely failure item, followed by a diode after that.... Capacitor would be a distant third...