81 XJ650 Misfiring After Heavy Rain - Coils Replaced

I got caught in a severe downpour two days ago, and the bike began to misfire at idle. Was able to limp home by keeping my revs up, and I expected that the problem was just wet ignition coils. The misfiring seemed to die out and go away when at high RPM, strangely enough.

However, I decided to test that hypothesis by finally installing the dyna coil kit I bought from chacal a while back. The install went well, but alas, the bike is behaving exactly the same as before the coil replacement. I tried draining the gas out of the tank and carb bowls, then firing it up on an auxiliary tank, and the same thing happened.

Pulled the plugs, and all four appear somewhat sooty. If it is the ignition, it doesn’t seem to be restricted to just one coil. Where could water have gotten that would cause this?

Edit: got a video of how it runs - https://youtube.com/shorts/30D4kGpUWyc?si=tHw1feaz8qB2p6-Q

 

Wiring nearest the road, if all four cylinders the black earth wire from the pick up coils to the TCI in plug. Check all connectors spray with a water repellant follow the wiring testing the voltage at the pigtails going into the coils. Check charging system and electrical connectors from the alternator brushes black and green wires. If you can get it started rev the engine and set a multimeter to 20 volts DC and check if you are getting around 13.5 volts from the alternator. Wiggle the wiring too while pressing the starter might be a loose connection.

 

The alternator does not kick in until revs reach a minimum of 2000 rpm. Do you have the original Yamaha fuse box? If so change it for a blade type. Make sure there is no dampness below the spark plug caps. Apply a little dielectric grease on the rubber to cap joint and the lead to cap joints. My CB750 misfired in the wet so l used the dielectric grease to protect the joints from water ingress.

 

When I replaced the coils, I replaced the plug wires and caps too, so I know there isn’t water trapped in there. Though, I could still benefit from more dielectric grease on the outside of the joints. I used grease when putting the caps on after trimming the wires to fit, but only a slight coating on the wire to help the plug slide on.

I will definitely still check the alternator and that ground pathway. I bought a can of DeoxIt D5 for my vintage computer hobby, and the stuff comes in clutch for electrical connectors.

I’ll say though, I do have some additional data. I used a colortune plug in all four cylinders to see the misfires, and it looked like I was still getting spark when the thing wouldn’t fire. I was able to get it to smooth out somewhat (not perfect) by adjusting the idle mixture screws. Is it possible that water seeped into those screws and inhibited fuel flow?

 

Mixture screws have a small o ring on them l doubt water could get in there. Do you have an inline fuel filter, and original vacuum operated petcock? Open the fuel cap to see how it runs. Mabe a clogged air vent in the tank and the rain was coincidental? If you have good sparks it must be fuel related.

 

You can also check for vacuum leaks around the carburettors with some butane gas or starter fluid. If the bike was running properly before the downpour its unlikely its carburettor related.

 

I actually don’t have an inline fuel filter, which means it’s possible some junk got into one or more of the mini filters inside the carbs. But I don’t think it’s the fuel cap air vent, since I was running the bike off the aux tank and still having issues. I also do have an original petcock, but I rebuilt it and have verified good fuel flow under vacuum.

Starter fluid is probably also a good idea to try. I don’t have any of the classic symptoms of a vacuum leak, but my intake boots do seem to have some surface-level cracking. You can see it in this picture I took when I was taking the carbs off to replace the throttle shaft seals.

 

If necessary you can repair them with inner bicycle tubing and RTV. But it's not going to develop vacuum leaks suddenly especially on more that one cylinder. Even though you can see sparks with your colour tune it can still be missing intermittently. Just to eliminate this l would check the aforementioned wiring and alternator output just to be certain.

 

Mine too. I ended up baking the coils in a low-temp oven overnight, then sealing them with a coat of varnish. I eventually threw those coils out and fitted Honda's finest.
For the HT leads I encased them in electrical sleeving (the corrugated kind) and carefully taped over the ends at the plugs, to make sure nothing got in.

 

Good idea with the electrical sleeving I might try that on my bike.

 

Alright, so I have a bit of an update. Sprayed carb cleaner at the intake boots while running, no change whatsoever. Looks like I don’t have a vacuum leak. The bike still weirdly appears to run better at higher RPM, like it smooths out over 3000. Problem with switching over to the alternator from the battery, perhaps?

Also disconnected the TCI electrical connectors and sprayed contact cleaner in there, no change. I see the connection from the voltage regulator to the alternator, but I’m not sure how to disconnect that one without removing a bunch of stuff that’s in the way. I tried spraying contact cleaner through it from the top, no effect.

Two questions now, since I’m a noob at troubleshooting electrical problems. When I check the alternator voltage, where do I do that, on the battery leads? Second question: where exactly is the pick-up coil? I can follow the wire partway from the TCI, but it sort of disappears under the bike and I have no idea where to look for its connection.

 

Yes, on the battery - look for 14.5 +/-.3VDC at approximately 2000 rpm and up with a fully charged battery.

Pickup coils are under the left crankcase cover and the wiring terminates in the plug that attaches to the TCI, so no connectors along the way to cause trouble.

 

You can wait until you check voltages but this connector is notorious for a poor enough connection to cause overheating to the point of melting the connector

 

Alright so I’ve tested the charging circuit. 14.6 V at the battery when at 2k RPM, and the voltage is rock solid. Time to check more connectors.

 

Good charging system ok. Right handlebar switch grounds through the bars. Check no water got behind pick up coil cover. Unlikely but check anyway.

 

Well, I’ve now disconnected, cleaned, and reconnected every connector on the bike. Nothing has changed. It’s still super slow to take off: the bike will choke and sputter until it gets past 2k RPM, then it will run like normal.

Tried running with the tank open, tried running on prime. No effect. I guess I have to take the carbs apart at this point and see if anything is amiss, but I am well and truly baffled. Hard to imagine it picking back up at high RPM if it’s a fueling issue. Could my TCI be fried?

 

Yes it's baffling. I don't think your TCI is fried it runs. This wouldn't affect the bike suddenly but when was the last time you check the valve clearances?

 

I've had a fooked TCI and it still ran on 2 - Maybe @Rooster53 or somebody familiar with these magic boxes could chime in.

 

Mabe this will help. https://xjbikes.com/forums/threads/solved-bogs-at-low-rpm-revs-over-2k-ignition-help-solved.37614/

 

Yes Rooster knows his electrics. One of your transistors must have been toast or the soldered joints on the board.

 

It would be good if he had a spare TCI to check.

 

That sounds like my exact issue. Unfortunate that his solution was to replace the coils, since I’ve already done that… I guess I will reconnect my plug caps and see if it’s just a bad connection. I have Dyna coils now, so I could also try reconnecting the other end of the wires to the coils. Will probably have to wait until tomorrow though, since I don’t think I’ll have any more troubleshooting time the rest of the evening.

 

Did you do the load test on the bike. Put the multimeter on 20v dc. Put the probes on the battery terminals and press the starter button. Let us know what voltage you get with the starter turning the engine.

 

Indeed I did do this. Voltage dropped from around 13.3 to 11 while the starter was cranking (I didn’t write down the number, this is my best recollection for now). I think the battery is good, it’s a new AGM batt and I keep it trickle charged regularly.

Edit: just found some time to redo the test, since I realize I had the light off last time I tried it. Same result. 13.3 V with no load, 12.5 V with the key switched on, and 11.3 to 11.5 V while cranking the starter.

 

This is so typical of clogged low speed jets where the engine really struggles until the slides raise and the mains kick in. You mentioned you didn't have an inline fuel filter, which is a bit dangerous unless the tank is just pristine.

How much riding had you done before you left for the trip? I was bummed to hear you weren't going to make it to the carb clinic.

As for the TCI / coils / power - a weak spark typically gets you a lot of missing as soon as the throttle is opened and a rich mixture occurs, particularly under a load. Checking the power to the coils / TCI might be a good idea in case there is a fuse box, kill switch, or connector issue. You won't see a full 12V there with the bike not running, but it should be close.

It would be great if you had a TCI you could throw in there to verify it is functioning OK

 

Hmm, I didn’t know that could happen from lack of an inline filter. I thought there was a risk of clogging the screens on the needle seats, but I didn’t think anything could make it past those screens large enough to clog a jet. My tank is fairly clean, but I can see a small amount of surface rust at the very bottom, on the joint where the metal was assumedly stamped together. Maybe I underestimated it. I’ve rode 3000 miles on the bike since buying it almost exactly a year ago.

If redoing the connections on the plug wires doesn’t help, I’ll pull the carbs and check them out. Also unfortunately, I don’t have a known good TCI to check. I could always pop the cover off mine and see if there’s any visible damage, though I know solid-state components can fail in non-visible ways.

 

Check that 12V to the TCI / coils first just to be sure it is healthy - it might be value added to check it while running since the motor produces lots of vibration that could aggravate a marginal fuse or connection.

It's hard sometimes to tell the difference between a carb and electrical issue on these bikes, but carbs are often the reason for poor performance. However, the fact that you had gone some 3000 miles and then get caught up in a rain storm to expose theses symptoms makes it hard to point to the carbs.

 

Are the choke plungers all working correctly?

 

Definitely check fuses and connections. I've had similar symptoms on other bikes and both times it was the glass fuses. Even when they look good and seated properly and indicate 12V when stopped, a bit of corrosion + vibration makes a poor contact. TCI power dropping out can be indicated by the tach needle bouncing (electric tach), but I think you still have the gear drive tach on the 81?

 

Yes, the 81 has a mechanical tach. It still gets bouncy, but that’s because of misfiring and uneven running lol.

On the bright side, I already upgraded to blade fuses a while ago. And I reseated all fuses as part of troubleshooting.

 

Finally had the time to do this troubleshooting. Hooked up my multimeter to the coils, cranked the engine. It wouldn’t read anything on DC mode, but I was able to measure 5.5V AC. I think that means it’s in-spec.. an electrical signal alternating between +11 and 0 volts would have an amplitude of 5.5V. After that, I disconnected all my plug wires, sanded the contacts, reapplied the caps, and reconnected them to the coils.

When I fired the bike up, it actually seemed smooth for a while. But after a brief road test, it went right back to bogging down at low RPM. At this point, I definitely suspect the fueling. I pulled the carbs, and I’ll inspect the screens in the needle valve seats to see if there’s a problem. Will also try to blast the idle jets with carb cleaner and see if there’s something stuck.

 

I think that means you connected the meter to negative or low side of the coil. That side of the coil is driven by the TCI. The meter needs to go to the positive side of the coil where the R/W wire is, or to the R/W wire on the TCI.

Hopefully, you find an issue with the carbs

 

I took the bowls off, then removed the needle valve seats and pilot jets to check for blockages. There was a tiny bit of debris in the filters on the needle seats, but nowhere near enough to obstruct flow. I found no issues with the pilot jets, and I was able to spray carb cleaner into the emulsion tube to verify that it can flow through.

And of course… no change. At this point, I guess I have to suspect the TCI. Or maybe there’s something more deeply wrong with the valve train/engine. I am completely at a loss, and I have exhausted all troubleshooting I can think of. Feeling defeated here, to be honest.

Could I have screwed up the dyna coil install? Maybe. Seems unlikely though, since it’s behaving exactly the same as it did before. Could bad cam chain tension cause these symptoms? I don’t know, and my bike has an automatic tensioner so it’s even less likely.

 

I doubt you have messed up the new coil installation. Carburettors are clean. Bike was running well before downpour which would suggest tci should be fine. Water in any switches? Spray all switches with water repellent. Trying to but cannot think of anything else at the moment. Mechanical issues unlikely those motors are tough. Any chaffed wires?

 

Did you check the ignition switch?