Troubleshooting kill switch on a Honda ATV (93 TRX300).
 

For about a year, my ATV would intermittently crank but not start. I discovered that quickly switching the kill switch back and forth repeatedly from 'run' to 'kill' would cause the engine to hit once or twice, and often it would eventually catch and start, after about 30 seconds. Also, whenever it would hit, it would only hit when i moved OFF of the 'run' position into the 'kill' position.

I disconnected the switch assembly, and ended up cutting the two wires free and then just tying them together. Same condition... engine cranks while they are tied together, but wont start... if i remove them and touch them back together while i am holding the start button, the engine will sometimes hit, and it only hits at the moment they are separated.

Disconnected the wiring harness and checked voltages.

Wiring Diagram

In that diagram, at the bottom left corner is a "Starter + Engine Stop switch" box. The kill switch is labelled "IG2" and is a Bl/W wire that runs up to the CDI unit. There appears to be a diode (?) on the line.

With the key in the on position, and the two wires from the kill switch tied together:

Measuring from G (green) to Bl/W (black/white) while the start button is NOT being pressed, i get 12.3 volts.

When i press the start button, the voltage immediately drops below 11 volts.

The service manual only has this to say about inspecting these connections:

Test ignition switch and engine stop switch: (Turn ignition switch to ON and the engine stop switch to RUN) - Bl/W to G - Battery voltage should register.

According to that, i should be seeing the battery voltage when the start button isn't being pressed, which i am.

When the start button is pressed, the voltage shouldn't drop like that should it? Does the fact that it only hits when the wires are separated point to anything? The wiring diagram shows what i believe to be a diode on the wire... could this have something to do with it?

Thanks for any help!!

Have a little more info.

First of all, I think that what i thought was a diode was just a regular connection... so disregard that.

Someone told me that the voltage dropping from 12.8ish to 11ish is normal when the starter is turning (these are the same readings i get at the starter wire when i test that wire on and off the starter, while pressing the start button). If that's true, then i guess the Bl/W wire from the kill switch is functioning properly.

Someone else said that what might have been happening is that when i was flipping the kill switch off and on, it was cutting the voltage off and on to other components, which is probably where the issue lies... and so the problem probably isn't with the kill switch itself.

They suggested that it could be the CDI unit. Those things are $80 used on ebay... anyone got any ideas of other stuff to test before i take the chance on one?

This is a common fault. There are lots of posts on here about it. Most have taken the plug out, re-connected it and laid it on the head. You get one spark when you press the starter and again when you take your finger off, with nothing in between. Some say that if you clean the trigger coil in the alternator housing, the fault will go off but I've had it on a few bikes and never cured it, owners get used to starting by bliping the starter. It is nothing to do with the kill switch and 11v is quite good for a battery while it is cranking the engine over.

merryman, when i unscrew the spark plug and lay it on the head it sparks strong for the entire time i am pressing the start button. would this mean that the issue i am having is a different one than what you are referencing?

See if sparks good with plug screwed into head with the key switch on and kickstart engine an see got spark that way without using the electric start system!!!

Yes, it does seem different, but why should it not start "on the button" if you are getting a good spark all the time? It may be worth checking battery leads and earths as you could be loosing "cranking power" that way.

Yeah, now that i think about it, that seems off... if it's sparking, it should be turning over, as long as gas is getting to it.

I tested the kill switch wires while watching the spark plug, and it is working properly... when the wires are disconnected no spark, when connected strong spark. I'm going to spray some starter fluid in the carb and see if it will turn over.

I wonder if the fuel delivery is controlled electronically on this thing like with a fuel pump or something? If so, maybe the intermittent problem with starting was due to the fuel delivery malfunctioning as i cut the voltage off and on with the kill switch?

Also i think i read somewhere that a spark plug can show a spark unplugged, but still have issues when it is screwed into the block and under compression. I think someone said you can use an in-line spark plug tester to make sure it is working properly... maybe i will get one and test it that way.

Is this an USA model? no electric fuel pump from factory on America Honda Trx300..The 1995 and early 1996 model trx 300 had bad CDI from factory an had recall...only would spark while turning over when you release starter button.As long as you kept the starter button pushed CDI wouldnt spark until you released starter button.. Are you should its an 1993 model?

Okay, i've been messing with it a little more and it seems that when i saw a good spark it was just an intermittent coincidence.

With the spark plug laying on the block, i held the start button again and didn't get any spark. I kept pressing the start button off and on, and fiddled with connections because i knew it was sparking yesterday, and eventually it started doing the thing that merryman and paphogg talked about, where it would spark once when i let go of the spark button, or when i first hit the spark button. So it sounds like the same issue you guys were describing.

At one point it started sparking strong for the entire time i was holding the start button, and i hadnt touched or moved anything... then a few tries later it stopped sparking again. So that tells me it is something intermittent going on.

Also i untied the kill switch wires and touched them together while i was holding the starter button pressed... each time i disconnected the wires the spark plug would fire once. So this is what was happening when i first started having the problem, but still got it to start by flipping the kill switch button... breaking the kill switch connection made it spark once, which made it hit briefly... and i guess every so often whatever was making that action cause it to spark once, would catch and keep it sparking.

I do remember that some times i would hold the start button for a minute or more while i flipped the kill switch back and forth, but i could never get it to hit enough to start. Hours, or days, later i would go back to try it and it would fire up on the first try without doing anything special. This might have been like when i tested it earlier today and the spark plug randomly started firing as normal for a few minutes. Maybe it's an intermittent problem that has gotten worse.

Anyway. So merryman or papa hogg, do you guys recall if replacing the CDI unit has fixed this issue in the past?



What do you mean by 'bliping the starter'? Hooking a jumper cable up to the starter post and starting it that way?

Thanks again for both of your input!

No, bliping is just pressing the button on and off rapidly, this gives you the same result as you got with the kill switch, but I suspect the kill switch will fire the plug when you connect, disconnect, doing this with the starter button fires it at the right time. I tried to cure a Honda 350. Changed plug, coil, CDI, starter and checked all connections and earths, no better. I didn't know about the trigger coil at the time so didn't change that.