Misfire under load

[this old xs]

Hello,

About 4 months ago I got a 1978 XS650 Special that was sitting for 20 years, definitely farm fresh. Got it running and on the road and she's just a treat to ride. I must say this forum has been a really valuable source of information, thanks!

However, ever since I've had it, it has a misfire under load from 4.5krpm but only at full throttle. If you ease up on the throttle it clears up instantly. I'm hoping the brains trust here can help me.

The bike is stock except for mufflers (diffusion megaphones) which I couldn't imagine would cause this.

I have gone through the wiring harness and cleaned connectors, tested stator (multimeter, spanner test), regulator, rectifier, you name it. Coils ohm out perfect. Charging system works. I can only get battery voltage minus 1 or 2 volts at the coils which I know isn't good.

I have put a kit in the carbs (needle and seat, gaskets, choke etc). I've gone from the stock 135 mainjet up to a 155 and lots in between. Back to 135 now. Replaced the butterfly shaft seals (less one but they won't cause this issue), new carb holders.

Ignition timing is good, new advance springs, new points and condensors. Valve clearances and camchain tension all correct. Points gap is right.

New plugs (tried BP7ES and currently iridium). New plug caps non resistor, old ones were resistor but out of spec and iridium plugs are resistor anyway.

Today I tried a brand new single, dual output coil (instructions from here are perfect). No matter what I've done, the full throttle under load misfire persists. It sounds exactly like a rev limiter. No load revs fine.

It has an intermittent big miss at 2.5krpm on light throttle applications such as turning at an intersection but it's intermittent. Probably related though.

I'm pretty well stumped at this point. I ride around it, but sometimes you just want to go full send you know? Could it be the diaphragms in the carbs? They seem ok but they're original.

Any help appreciated!

 

[member43148]

The bike is stock except for mufflers (diffusion megaphones) which I couldn't imagine would cause this.

Betcha that is contributing.
Stock airbox ?
Maybe try one size up on pilots, One size up on mains ?
Diaphragms nice condition?
The needle jet can be raised or lowered by the locating circlips. One notch leaner with larger main jets could help fight that transition stumble from pilots to mains.

 

[this old xs]

Hi XSTime, yes, everything is stock apart from mufflers.
I've got 137.5 mainjets which I tried but no dice. I've tried all clip positions too.

 

[this old xs]

Ok, talking to myself here, but this might help someone in the future as this site is very well indexed by search engines.

I lubed the advance rod and ever so slightly bent the weights. I've always had trouble getting the ignition retarded as much as I'd like but that helped a lot. That brought the misfire to 4.5-5.5kprm. It's very slight above that range but almost gone.

Reading this thread:
https://www.xs650.com/threads/1978-xs650-special-project.65053/page-13
I got to thinking more retardation might be better. Timing marks are |F| and I'd been timing to the first | (manual says 1/2 way between them, should have read it more closely!) Anyway, based on the advice in that thread I timed it to the second |.

I also checked the static full advance by wedging the weights out with bits of vacuum line. Way too much advance. I've previously checked with a borrowed timing light and it was OK, but I never went above 3krpm. Dunno. Anyway I bent the stoppers that the weights rest against at full advance and that took care of that.

Misfire is pretty much the same. I'd love a log of the ignition timing, but sadly points don't offer that feature. The plan now is to buy a wideband O2 sensor and AFR gauge and do some investigation on the mixture situation.

I'll report back when there's something to say.

 

[JOHNDADDY]

" I can only get battery voltage minus 1 or 2 volts at the coils which I know isn't good."
Yup, I'd dig into that first. Especially at higher RPMs.
What's the voltage at the battery at 3,000 RPM?

 

[this old xs]

What's the voltage at the battery at 3,000 RPM?

It's good. I tied a multimeter to my handlebars and went for a ride. Don't remember exactly but it was good voltage, around 14 and no changes during misfire.

I seem to recall jumping battery voltage direct to the coils at one point too, but it mustn't have made any difference or I would have bodged something up permanently.

 

[jaybar6]

It's good. I tied a multimeter to my handlebars and went for a ride. Don't remember exactly but it was good voltage, around 14 and no changes during misfire.

I seem to recall jumping battery voltage direct to the coils at one point too, but it mustn't have made any difference or I would have bodged something up permanently.

hi lemmiwinks, I think the electric circuit is fine. sounds like fuel starvation to me. I would check the gas tank cap is not blocked and the fuel petcocks are free from crud.

 

[this old xs]

So on the way back from my usual Sunday cruise, the bike started dying and generally misbehaving intermittently. Got me home as it has always done (fingers crossed I'm yet to be stranded by the side of the road), and died in the shed.

The Brake/Tail light was not on with the ignition on but the bike not running, so that told me there was a wiring problem. Previously the red wire with the white trace had broken off at the fuse box and caused me the same problem. At the time I soldered an extension on as it was getting a bit short.

Well the wire was corroded and I had trouble getting a join. Anyway it broke. I trimmed the oem wire and tried again. Still corroded but I got a half decent join.

Today I had an opportunity to ride, and on the way home, I'm 90% certain the misfire was less pronounced. Pretty sure I didn't imagine it so I took a close look at that wire.

On what I'll call the short side (comes from the starter solenoid) it has 1.6 ohms resistance where it should have none. I think the corrosion goes way up inside that wire and it needs to be replaced.

I tried to bypass it by making a new wire from under the side cover to the fuse box, but it didn't work. That wire must go somewhere else inside the loom. I might have to do some surgery on the loom but it seems like it could be a big undertaking.

 

[this old xs]

hi lemmiwinks, I think the electric circuit is fine. sounds like fuel starvation to me. I would check the gas tank cap is not blocked and the fuel petcocks are free from crud.

Rebuilt the petcock (more times than I care to recall!) Can hear the tank venting through the cap, and I've had that apart but couldn't hurt to check it again.

Wideband O2 and AFR gauge are on their way. When that gets here there will be no hiding mixture issues.

 

[this old xs]

I remembered I found a video on cleaning electrical connectors ages ago. Well here it is:

I reckoned that's a lot easier than tearing into the loom, so I had a go. After cleaning, the wire soldered normally and tested at 0 ohms.

I'll be able to test ride tomorrow. But that red wire with a white trace feeds power to the coils and 1.6 ohms has to cause a voltage drop. Because I'm lazy I didn't take the tank off and measure it tonight. I'll be able to tell if the misfire is there or not without any instruments.

 

[this old xs]

Update #1. Somehow I'm never in enough of a hurry to get TO work to be able to test full throttle. Going home is a different matter entirely, so well have to wait for that test.

I don't want to count my chickens before they're hatched, but the low speed, light throttle 2-2.5krpm misfire is pretty well gone. I felt one and heard one, as in a single misfire each time, where before I could reliably dial it up whenever I chose.

I was getting through it by winding the throttle open, but this morning I could cruise along on tiny throttle openings just riding the torque. What a treat. Fingers crossed the full throttle under load misfire has died a similar death.

 

[this old xs]

The full throttle miss is still there, but only just. It's almost like a flutter now. A fella could almost live with it if he was inclined. If I get time over the weekend I'll pull the tank and check coil voltage and the red wire/white trace resistance.

I might even lift the needles a notch while the tank is off. I got pretty fast at that early on when I was troubleshooting. If all that checks out the AFR gauge will be here soon enough.

 

[jaybar6]

The full throttle miss is still there, but only just. It's almost like a flutter now. A fella could almost live with it if he was inclined. If I get time over the weekend I'll pull the tank and check coil voltage and the red wire/white trace resistance.

I might even lift the needles a notch while the tank is off. I got pretty fast at that early on when I was troubleshooting. If all that checks out the AFR gauge will be here soon enough.

Hi lemmi, did you clean the safety stop/switch with your vinegar cleanse? I have found the switch itself drops half a volt on the red/white wire, all by itself.

 

[Paul Sutton]

That method of cleaning using vinegar/salt is quite commonly mentioned on various forums. However, I do not recommend it even though the connector comes out very clean. When the wire dips below the surface of the vinegar/salt solution some is carried up the copper wire a couple of inches by capillary attraction. No washing with water or bicarb solution is going to remove the salt which is now trapped under the wire insulation. This trapped salt will lead to quite rapid corrosion of the wire. Though we think of copper as being resistant to corrosion this is not so true in the presence of salt.

About 8 years ago I cleaned a connector on my XS as per that video. Several months later I was thinking about it having previously worked in a lab where copper corrosion was being studied on historical finds. I decided to strip back the insulation on the previously cleaned connector and was very surprised to see how much corrosion had occurred.

 

[this old xs]

I have long wondered (suspected) the kill switch. I'll investigate.

 

[this old xs]

Thanks Paul. I wondered about that myself. I did neutralise it with bicarb afterwards, but I reckon, based on the change in resistance that the corrosion didn't go up that far. Therefore I should be able to chop it out and solder an extension in pretty easy hopefully.

 

[Its been a long time]

I have long wondered (suspected) the kill switch. I'll investigate.

The good news is it's really easy to take it apart to clean the contacts. I found in my switch, the two small coil springs that put a preload on the contacts were rusted and broken. Ordered some Chinesium ones that fitted perfectly and the switch is good as new now. The contacts were good so just needed a polish.

 

[this old xs]

Misfire GONE!!! tl;dr go to last few paragraphs.

I did a bunch of stuff, including cleaning the kill switch contacts which were dirty, thanks for suggesting that. I also cleaned the wires where they attach to the brush holders, and combined I did get a voltage increase but the misfire persisted.

I synched the carbies using my mate's fancy synchro king electronic thing (so awesome) and that really cleaned up the low speed misfire which had reared its ugly head again. My eyeball synch wasn't bad, but this is way better.

Adjusted the mix screw using O2 sensor and AFR gauge. Was pretty rich at 2 1/4 turns out, about 12:1. XS seems to like it a little rich, ended up at high 13, low 14 (just under 2 turns out) which gave a lean pop on the left hand cylinder on deceleration - AFR showing 20:1. Full throttle acceleration was reading 8 or 9:1 which seems fair. Didn't seem to change when misfiring. Test was sketchy only 1st and 2nd gear because I had a pretty janky pipe setup to get the readings. Might revisit in the future with a more secure arrangement and check some top gear cruising readings.

Last night I decided to check the valve clearances because I have a slight tick on start up which goes away after about 20 seconds. I thought too big exhaust clearance closing up when the valve heats up. Nope. Clearances were really, really tight.

Not sure what happened. I remember checking them when I got the bike and they were quite loose. Maybe there was crud on the valves and seats (sitting 20 years remember) and after that burned off it closed up. I dunno, anyway I set them to 10 thou inlet and 15 thou exhaust. I also gapped my iridium plugs down to 12 thou.

No low speed misfire (2-2.5k rpm) and the full throttle 4.5k rpm - 5.5/6k rpm misfire is gone. Fingers crossed it stays that way. I've thought I fixed the low speed one before but it came back. I've never gotten rid of the full throttle one though.

The valve train sounds like an air-cooled VW beetle now so I might have to check again. But I hit 95 before I ran out of road and there was definitely more to come from the engine.

 

[this old xs]

Aaand I see I set the valve clearances completely wrong. Was looking at the clymer and must have mistaken mm for thou because it's 0.1 mm inlet and 0.15 mm exhaust. No wonder it sounds like a VW

 

[this old xs]

I set the valves back to spec and the misfire is back. A while back I read a thread where ggGarry (rumour has it if you say his name 3 times he appears in your thread) advised someone with basically the same problem as me that there is a particular cam chain tensioner which causes premature cam chain wear.

I've got that tensioner and had trouble setting my ignition timing to where it should be, also another sign of a worn cam chain I believe. Just for fun, I set the clearances back to enormous and apart from a little misfire between 4 and 4.5k rpm, it pulls like a train at full throttle.

I presume this is due to altering the valve timing with the excessive clearances.

I was going to mess around and see if I can find a sweet spot between ridiculous clearances and misfiring but that would be a dumb waste of time I reckon. Seems to me like there's no ignoring it, I need to replace the cam chain and probably get a different tensioner too. It's coming into summer here, so what I'll most probably do is put the clearances back to spec and just live with the misfire. For now...