Starter wiring question

[CatatonicBug]

I'm usually pretty good with wiring, but this one has me stumped, as it's an intermittent problem.

On a really cold day (i.e., <25ºF), I walk out to start the bike, turn the key, and the dash lights come on just fine. I flip the e-stop switch to "On" and the dash lights dim ever-so-slightly, just like normal. I press the starter button, and nothing happens! No thunk of the solenoid - nothing. If I ark across the solenoid, it'll crank, and start just fine. But here's the strangest part - if I don't crank it long enough to start, after arching across the solenoid for a second, I can press the starter button, and it works again!

I checked the plug running to the solenoid, and when the button doesn't work, there is no power to the plug when I press the starter button. The same is true at the actual wires in the button housing. It's almost as if there is some other relay in the system that gets stuck in the cold and cuts the power to the button until the bike shakes a bit.

I tried wiggling the wires under the tank, but that didn't help. I even took the 6-wire mini-din plug under the tank apart and cleaned it. Any ideas where else I should be looking for a failure point? I thought maybe it had something to do with the neutral switch, but the way I understand it is that as long as the dash light is on, the neutral switch should be letting the bike start - right?

I have not had this problem on warmer days, so I get a feeling that something mechanical (like the internals of a relay) is getting stuck in the "off" position until it gets rattled loose.

[CaptonZap]

Howdy,

Since I don't have a wiring diagram for an H model, I will couch this in terms suitable for a G model.
When the ignition is turned on, power is supplied to the TCI and starter solenoid through the R/W wire. The circuit then goes through the solenoid coil and out the Blu/W wire to the starter switch, at which point it is grounded, completing the circuit through the solenoid coil. Which then pulls the heavy contacts together to feed the starter.
Since you can jump the solenoid, and I assume that you are talking about the big contacts, we don't know if the coil is flakey. Next time this happens, take the connector apart at the solenoid, and ohm the two wires that go to the solenoid.. You should have continuity. If you don't, hit it, the solenoid, with a screwdriver, and see if it gets continuity. If it doesn't, take it apart, and hope you can see the break, because, if you can't, the break is internal in the coil, and you are SOL.
(You should also have battery voltage on the R/W wire, in the loom side connector, with the switch on.)

If you have continuity through the solenoid coil, put one test lead on the wire in the connector half that is hooked to the loom, the color that is nor R/W (On the 80G, it is Blu/W). This leads to the starter switch, and makes to ground.
Put the other lead on a GOOD ground, and press the start button. You should get continuity. If you don't, take the starter switch apart, being careful not to lose the LOLP's, (lots of little parts) and clean them, and the surface of the bar that they press against.
Reassemble, and at that point, the thing should start.
Another option is to wait till warmer weather.

Good luck, CZ

[crazy steve]

It is probably a relay; the 'safety relay' the LH 'features'. This is half of the interlocks that the XJ has, lacking the sidestand switch.

Of the two wires coming off the solenoid coil, one goes to the start button. The other (red/white) goes to this relay (nope, don't know where it is; you'll have to trace the wire.. ) and if it fails to close, you'll get a no-start.

There should be four wires connected to the relay; two red/whites (and if you tie these together, that will defeat the interlock), a sky blue to the neutral switch/light, and a black/yellow that goes to the clutch lever switch. The other side of the clutch switch goes to ground. There's a diode internally on the relay that connects to the neutral switch. If you get a no-start with the bike in neutral, try pulling the clutch in; if the bike now starts, the diode is failing. If not, then the problem is the relay contacts or connections in the circuit.

None of this is in the Clymer manual, it's only in the FSM supplement that covers the LH....

[CatatonicBug]

Thanks Steve. That makes sense. I hadn't tried the clutch lever test, and I think I might know where the relay is. I've always wondered what the extra relay was for under the seat.

[DAVINCI]

Your ground my be flaky. Remember, the start switch completes the ground circuit through the handlebar, so make sure either the harness ground wire is cleanly connected at the switch housing or that your steering bearings are snug.

[CatatonicBug]

Your ground my be flaky. Remember, the start switch completes the ground circuit through the handlebar, so make sure either the harness ground wire is cleanly connected at the switch housing or that your steering bearings are snug.

Not so on the LH. Because the bars are black, they added a ground wire for the starter button. That's the 6-pin mini-din connector I cleaned.

[motoman]

Not so on the LH. Because the bars are black, they added a ground wire for the starter button. That's the 6-pin mini-din connector I cleaned.

Suspect your issue is much simpler than previously explained, although those points made may come in handy. IMO, the relay contacts have arced enough times the brass contacts have burnt marks. You saying it would start using the start button, ONLY after jumping the solenoid prior. This is certain suspect of the brass contacts, as jumping across solenoid made up for the slight voltage drop giving a slightly higher voltage creating a connection and leaving ANOTHER temporary burnt spot on brass for connection to be made from start button if not left for long......BTDT, replace or fix the start relay.

[CatatonicBug]

Suspect your issue is much simpler than previously explained, although those points made may come in handy. IMO, the relay contacts have arced enough times the brass contacts have burnt marks. You saying it would start using the start button, ONLY after jumping the solenoid prior. This is certain suspect of the brass contacts, as jumping across solenoid made up for the slight voltage drop giving a slightly higher voltage creating a connection and leaving ANOTHER temporary burnt spot on brass for connection to be made from start button if not left for long......BTDT, replace or fix the start relay.

By "start relay", do you mean the solenoid, or the "Starter Circuit Cut-Off Relay (4H7-81950-01)"? I assume you mean the solenoid.

If that was the case, there would still be power going to the solenoid when I hit the starter button. The problem is that there is no power being sent to the solenoid contacts when I hit the button. If the solenoid was actuating when I hit the button, your diagnosis would be correct, but that is not the case here. I already rebuilt the solenoid a year or so ago, so the insides should still be good anyway.

[ViperRon]

I had a similiar problem with mine and drove me nuts. My issue was I could hit the start button once to start and the bike would crank but if it did not start or died I could not get it to turn hitting the start button. When i hit the button i could see the light in the console dim ever so slight. Turn the key off and on and it would turn again untill I let off the button or the engine died. The other thing that cleared the condition was jumpering across the starter relay. The end result was replace the start button. With a meter it made contact but would not carry amps to pull the relay I guess. Was intermittant also mainly cool or damp days.

[crazy steve]

Yep, if you don't have 12v at the starter button before pushing it, the problem is the safety relay or it's connections...