Infrequent cutting out and magical recovery

[MFD]

Hello all,

After 18 months of (almost) flawless service, my FZS600 has decided to start impersonating my old Suzuki.

On the way home from work on Wednesday last week, while going up a hill, the engine just cut out with no warning, wasn't even trying to fire. After pulling over onto the pavement and turning it off/on a couple of times, it still wouldn't restart. After a couple of minutes of me taking my gear off I gave it another try, and it immediately fired on the first spin.

Fast forward to tonight, and it packed in not too far from home, same as before. Rode it down probably from 70ish to 20ish in gear, no life from the engine at all. Pushed it on a bit, still wouldn't start, pushed it on a bit more, still wouldn't start... until I was in sight of home.

There was no smell of fuel when cranking either time, and it didn't start with an almighty backfire like other times I've flooded in the past, so I'm inclined to believe its a fuel issue. Both times the fuel pump was silent for all the starting attempts, so I don't think it temporarily ran dry or was struggling to prime or anything like that.

Is there anything near the carbs that can shut off the fuel supply to all 4 cylinders, or am I barking up the wrong tree? Any other likely suspects?

Thanks

[BBROWN1664]

You said the fuel pump was quiet at the time. Try it in the morning and see if the fuel pump is quiet then. If so, start looking at the pump or the associated relay.

[MFD]

The fuel pump (I think) usually makes a few clicking noises when I first turn it on in the morning, has for as long as I've owned it.

There's normally no hesitation under sustained full power either, but I just remembered a few weeks ago the bike momentarily lost all power during an overtake.

[BBROWN1664]

check there is nothing blocking the air filter under the seat

[Fazerider]

Intermittent fuel starvation seems unlikely… running dry is usually preceded by a lot of hesitation. This sounds much more like lack of sparks.
I’d check the wiring harness and connectors for corrosion, shorts where cables can rub on the frame etc and broken wires.

[darrsi]

This all sounds rather familiar to what was happening to my bike intermittently for ages, and it was only when the engine cut out while i was riding it that it led me down another path to investigate.


The fix for me, and it's been months now with zero issues, was to flood the ignition barrel (where you put the key in) with electrical Contact Cleaner Spray. I used the key to open up the top part then just sprayed away, also putting the key in and turning it about at the same time to make sure it got everywhere.


I learned that the barrel is not just a straightforward On/Off mechanism, but there are several different contacts connected to it that become live when turned on. 


Over time, with different weathers abusing the bike, plus from memory i knew i had used WD40 before down the barrel, and also years later had used proper Graphite designed especially to free up sticky key locks, it had eventually become a bit gammy down there so the contacts were simply dirty causing all sorts of starting issues, backfiring, irregular running, then eventually cutting out, as you've mentioned.


Took me ages to get the answer i needed, after spending lots of time and money replacing and renewing other electrical parts, etc, but once i'd used that spray the bike has been spot on ever since.


You need to use a proper electrical contact spray that is specific for this sort of thing that won't leave any unwanted residue, i wouldn't recommend a full on degreaser spray as they can leave a powdery deposit that could still cause problems.


Have a go, it could be a very quick and cheap fix before you start going crazy like i did.


Just a note, make sure the engine is cool before you start soaking the lock with flammable liquid, for obvious reasons. 

[Dynspud]

I had a very similar issue when touring in Lake District a couple of years ago and it turned out to be the sidestand switch.
Any slight bump in the road would cut out the bike.
I'd get off put it on the sidestand and it would, obviously, start straight away.
Try putting the stand down next time it cuts out and see what happens.

[Dynspud]

....and bike in neutral, obviously 

[darrsi]

I had a very similar issue when touring in Lake District a couple of years ago and it turned out to be the sidestand switch.
Any slight bump in the road would cut out the bike.
I'd get off put it on the sidestand and it would, obviously, start straight away.
Try putting the stand down next time it cuts out and see what happens.

That was one of the first things I tried, and ended up just bypassing the stand switch just to rule it out, but the problem still came back later on.
It was at its worst on very cold nights in winter for some reason, but randomly had trouble starting the bike all year round.
When it used to backfire outside my work it used to frighten the life out of the receptionist. 

[His Dudeness]

If it's a sudden cut out with no warning I'd be thinking electrical problem too. If it cuts out again check if there is spark. If all the other electrics work but you have no spark you can focus on the ignition circuit. If you're losing spark on all four cylinders you can focus on points of the ignition circuit that are common to all four cylinders.

The components that are common to all four that are most likely to give trouble due to wear/dirt/corrosion are what the foccers above have suggested. The side stand, the kill switch and the ignition switch.

If the side stand is faulty the ignitor loses a ground so that would cause no spark on all four cylinders. If the kill switch is faulty the ignition coils lose 12V so no spark on all four. Those two circuits go through the ignition switch so if the ignition switch is faulty you lose spark on all four. And as fazerrider said check the wiring at the connectors going to the alarm. An open there would mean the ignitor loses the ground from the side stand so no spark on all four. You should check all those components and their wiring and connectors. If you have no spark on all four cylinders the problem is most likely in one of those areas.

A few other things that are less likely but would cause no spark on all four would be the pick-up coil failing, a problem at the cut-off relay, bad connection at the ignition fuse, failing ignitor.

[MFD]

Thanks for the suggestions. I'm going to try and get time to poke at all the wiring/connectors this weekend.

I've just been staring at the wiring diagram, but after half an hour the mess of diodes where the side stand and clutch switches meet the fuel pump and starter cut off relays still makes no sense. If the side stand or ignition switches went open circuit, taking out the coils, would I still be able to crank the starter?

[His Dudeness]

Does the bike have an alarm fitted? After it cut out did the starter motor turn the engine over? If it did that rules out quite a lot. For the starter to turn over 12V has to get to the cut-off relay. For 12V to get to the cut-off relay it has to go through the 12V circuit in the ignition switch, the ignition fuse and the kill switch so if the starter turned over you could assume all those components must be good.

If the starter turned over the ground to the cut-off relay must have been good too. Were you in neutral when you were cranking the engine? If so the ground from the neutral switch must have been good and it must have got as far as the cut-off relay. The ground from the neutral switch also should have gone to the ignitor and that should have been enough to start the engine but it didn't start so the ground from the neutral switch was good but it didn't make it back to the ignitor. Also when the bike cut out it was in gear so that means it cut out when the ground going back to the ignitor was coming from the side stand. So that means ground couldn't get back to the ignitor from either neutral or the side stand so that would suggest the open must have been at a point that is common to both of them. The most likely common point is the ignition switch or the connectors to the alarm. Or it could be right at the connector to the ignitor.

If the bike cuts out again you can test the theory by checking for ground at the ignitor when the bike is in neutral and the key is on. There are two ground wires at the ignitor, both are black. One is a permanent ground, the other one is the switched ground from the neutral and side stand. You're interested in the switched ground. If you turn the ignition switch on and put the bike in neutral, set your multimeter to DC Volts and put the red lead on the battery positive and the black lead on the black wires in the ignitor connector you should see 12V on the meter. With the leads still attached turn the ignition off. If you're on the permanent ground the meter will stay at 12V. If you're on the switched ground the meter will go to OL or if the switched ground is open all the time because of a fault the meter will read OL all the time and that would prove the fault.

All that is guess work though because we still don't know if it's actually losing spark or not. If it's running now you won't be able to test the ignitor for the switched ground but you can still visually check the wiring and connectors around the ignition, alarm connectors and ignitor and hopefully you'll see something obvious like a frayed wire or a corroded connector. If you can't see anything obvious you'll have to wait until the bike cuts out again and then check for spark. If there's no spark check for the switched ground at the ignitor.

[MFD]

There's a Datatool immobiliser (maybe also an alarm?) fitted. Paperwork says it was fitted by an installer. I don't know whether it just plugs into the "Alarm" connector on the wiring diagram (seems pretty easy to bypass if so...), or splices into the harness in a dozen places... I'll probably find out this weekend.

After the bike cut out, the starter still turned, the engine just had no interest in firing, so either fuel or spark was gone from all 4 cylinders. I tried the starter when in neutral, and probably in gear with the clutch held in too.

[MFD]

That escalated quickly...

Being the smart man that I am, I didn't bother putting the wiring diagram or multimeter in my bag today, after all, its only an infrequent fault, it probably won't manifest again for a while.

I made it perhaps 2-3 miles from home before it died, about one bend after overtaking a long line of cars. Did the usual playing around trying to restart, then pushed it up the hill and over the busy road to a garage to ask to borrow their multimeter... naturally the bike resumed working at this point, so I turned it around and headed back home.

Momentary cut out almost right away when turning off the road, then clear for perhaps another mile, until it did another momentary stall, then died completely. Repeat the wait-a-while-and-then-it-works dance, rode it off the curb of the pavement I'd parked on, no issue with the jolt from the drop, but then it packed in again all of about 50 yards later.

Pushed it to a pub, and waited a while again, finally decided to start up and made it the rest of the way home at 30-40 without issues.

While it was dead I jiggled the side stand, ignition and kill switches, with no effect. I couldn't see a spark between an extension bar I held in a plug cap and the cylinder head, also it backfired a couple of times after recovering, so pretty its on the ignition side now.

[MFD]

Right, stripped the tail down, propped the tank up and took side panels off.

I haven't found any (obviously) damaged wiring yet, but there is plenty of it disappearing where I can't see it. The immobiliser is very neatly installed - its been fully spliced and wrapped into the harness, which carries on past the ECU, round the back and down the other side to it. I probably couldn't remove it without having to cut the harness apart at the ECU and remaking some of the wires.

[MFD]

I made a simplified version of the Haynes wiring diagram, only showing things relevant for the ignition/starter/fuel pump/etc (can't seem to upload it right now)... and now I think I get it:

Spark requires ECU ground via alarm SS pins, which go to ignition switch, then to diode box, and to ground via side stand switch (when up) or neutral switch (when in neutral).

Starter requires starter cut-off relay to be grounded via neutral switch (in neutral), or clutch switch, then side stand switch (side stand up AND clutch lever pulled in).

I've confirmed the switched ground at the ECU is only grounded when the ignition is switched on AND the immobiliser is disarmed.

So I think that leaves me with the following possibilities:

- Break between kill switch and ignition coils
- Break in any ECU wires
- Break in wire from SS- to ignition
- Break in wire from ignition to diode box
- Fault in immobiliser/alarm
- Fault in ECU
- Fault in crank sensor
- Fault in diode box?!

The diagram also shows a wire from the neutral switch to the ECU, no idea what function that serves.

[MFD]

Ruled a few of my possibilities out:

Break between kill switch and ignition coils
Starter cut-out relay (and diodes) are further back on frame than coils, break inside the main harness that cuts the coils off without also severing the starter relay further back is highly unlikely.

Break in wire from ignition to diode box
Fault in diode box
I tried starting the engine in gear with the clutch held in this morning, engine span, but didn't fire, which would require the ground returning from the clutch lever to the starter cut-out relay via the same wire used to ground the wire from SS- via the ignition switch, if I've understood the diagram properly.

Fault in crank sensor
Wiring to crank sensor looks good, and resistance is in spec. I'm assuming this is a hardy component and not prone to developing intermittent faults.

Leaving me with:

- Break in any ECU wires (power, grounds and crank sensor really)
- Break in wire from SS- to ignition
- Fault in immobiliser/alarm
- Fault in ECU

[MFD]

You're definitely on the money Dudeness - I went out for a test run, with a bit of wire wedged in the SS ground on the ECU so I could probe it, died within 200yds, no ground. Engine starts up right away if I short the pin to ground, and then dies again when I remove the short.

After a while it decided to work again, and then died for a while after I dropped the seat back down. Wiggling the wiring around the rear of the tail doesn't seem to have any effect. The immobiliser still makes all its usual beeps and things when it won't start - so its power, ground and switched ignition are still good.

Considering it cut out when I placed the seat down, and it wasn't moving, I'm inclined to believe the fault may be in the immobiliser rather than the ignition switch - some kind of transistor-y thing breaking down, as it allows a small amount of current to go to ground when the bike is off. Any better ideas before I make a mess of the immobiliser harness to bypass it... and still have the problem?

[rayburn600]

Personally I would Ditch the immobiliser/alarm. More than likely, water in the immobiliser. 
They are more trouble than they are worth.


Other thing to check is Ignition relay. For the cost of it replace it.

[His Dudeness]

You're definitely on the money Dudeness - I went out for a test run, with a bit of wire wedged in the SS ground on the ECU so I could probe it, died within 200yds, no ground. Engine starts up right away if I short the pin to ground, and then dies again when I remove the short.

After a while it decided to work again, and then died for a while after I dropped the seat back down. Wiggling the wiring around the rear of the tail doesn't seem to have any effect. The immobiliser still makes all its usual beeps and things when it won't start - so its power, ground and switched ignition are still good.

Considering it cut out when I placed the seat down, and it wasn't moving, I'm inclined to believe the fault may be in the immobiliser rather than the ignition switch - some kind of transistor-y thing breaking down, as it allows a small amount of current to go to ground when the bike is off. Any better ideas before I make a mess of the immobiliser harness to bypass it... and still have the problem?

Nice one That proves that the fault is the switched ground not getting back to the ignitor. Next you have to find where the open is. I'd start at the ignition since it's the easiest place to test. If you follow the wires off the back of the ignition they go to the junction box under the tank. There are two connectors coming off the ignition, a two pin and a four pin. You're interested in the four pin connector. With your meter on DC Volts put the red lead on the battery positive and the black lead on Blue/Yellow in the four pin connector. You should see 12V on the meter when the side stand is up and OL when it's down. Next put the black lead of the meter on the Black wire in the four pin connector and turn the ignition on. You should see 12V on the meter when the ignition is on and the side stand is up and OL when the ignition is off or the side stand is down. That would prove the ground is getting from the side stand through the ignition and that everything is good up to that point so the fault has to be after the ignition.

If you find a fault in the ignition you could jump the Blue/Yellow to the Black with a bit of wire and the bike should start. If the ignition test good and you have ground coming out of it the next place the ground goes is the alarm connector so you would have to identify that Black wire that's coming out of the ignition at the alarm and check that it still has ground when it gets to the alarm. That's easier said than done since all the wires at the alarm connector are black!

[MFD]

I stuffed a bit of wire into the SS- wire at the ignition connector and went out for a couple more test rides.

Bike stubbornly refused to die on the first run, then started acting up on the second run. Confirmed ground connection at the ignition switch connector and not ECU when dead, furthermore, I measured 0.5v between the battery positive and switched ECU ground, as well as the ignition connector with ignition off, confirming the path from that connector to the immobiliser is sound.

Now I'm just debating if its worth the effort of getting some security torx bits to take the immobiliser apart in search of failed solder joints, blown caps, etc... or just ripping the thing out.

EDIT: After it started acting up, I shorted the ECU switched ground to permanent ground, no more drama for the rest of the way home.

[His Dudeness]

If you've confirmed you've got ground coming out of the ignition you'll have to check if you've got it going into the alarm and if you've got continuity between the alarm and the ignitor. I've never had an alarm on a bike so I've no experience of them but I think all they do to immobilise the bike is they stop the switched ground getting to the ignitor so that stops spark and they stop the switched ground between the starter button and the cut-off relay so that stops the starter motor turning over. Your starter is turning over when you press the start button so that part of the alarm is working so that might point more towards the alarm is actually working but there's a break in a wire. But I guess it's possible that something has failed in the alarm that only affects the ground to the ignitor and not the ground between the starter button and the cut-off relay

[MFD]

I cut out the immobiliser and all of its tendrils today, then went out for an hour or so without any issues. So hopefully thats the end of this saga!

[His Dudeness]

Seemed like it was cutting after a short ride before. If you rode for an hour with no problems it's probably good now. Too bad we won't know if it was the immobiliser or the wiring! As far as sagas go that wasn't too bad. You should have a look at Red98's stutter thread for a real saga http://foc-u.co.uk/index.php/topic,2066.msg15672.html#msg15672