Ignition Problem?

Bob, I borrowed this summary from Annas Dad (post #83 in this thread) and added a few things:

Additionally, I've distilled the various components that have been investigated, and found to not be at fault:

  • Engine Stop Switch – Clean & Inspect
  • ABS Module – Clean & Inspect
  • Tachometer Circuit – Inspect
  • Ignition Coils – Swap Out
  • Battery & Ground Connections – Clean & Inspect
  • Throttle Position Sensor- Replace
  • Engine Control Module – Swap Out
  • Side Stand Switch - Clean & Inspect
  • Spark Plugs - Replace
  • Spark plug wires tested for resistance and connection at cap end renewed (cut off 1/4" and screwed cap back on)
  • Crank position sensor replaced
  • Camshaft position sensor replaced
  • MAP sensor replaced
  • Fuel pressure regulator vacuum port checked for leaks found OK
  • O2 sensors disconnected, computer memory cleared and test run "open loop" no difference, stumble still present.
  • All vacuum lines inspected, 5-way T cleaned and inspected
  • All accessible harnesses and connectors inspected, connectors opened, inspected and cleaned. Nothing suspicious found
  • Clutch switch replaced (but I don't think this is related anyway)
  • Ground wire points on frame inspected, cleaned & tightened.
  • Swapped out fuel cut relay.

I'm probably forgetting a couple of things.
 
Last edited:
wjbertrand said:
Bob, I borrowed this summary from Annas Dad (post #83 in this thread) and added a few things:

Additionally, I've distilled the various components that have been investigated, and found to not be at fault:

  • Engine Stop Switch ? Clean & Inspect
  • ABS Module ? Clean & Inspect
  • Tachometer Circuit ? Inspect
  • Ignition Coils ? Swap Out
  • Battery & Ground Connections ? Clean & Inspect
  • Throttle Position Sensor- Replace
  • Engine Control Module ? Swap Out
  • Side Stand Switch - Clean & Inspect
  • Spark Plugs - Replace
  • Spark plug wires tested for resistance and connection at cap end renewed (cut off 1/4" and screwed cap back on)
  • Crank position sensor replaced
  • Camshaft position sensor replaced
  • MAP sensor replaced
  • Fuel pressure regulator vacuum port checked for leaks found OK
  • O2 sensors disconnected, computer memory cleared and test run "open loop" no difference, stumble still present.
  • All vacuum lines inspected, 5-way T cleaned and inspected
  • All accessible harnesses and connectors inspected, connectors opened, inspected and cleaned. Nothing suspicious found
  • Clutch switch replaced (but I don't think this is related anyway)
  • Ground wire points on frame inspected, cleaned & tightened.
  • Swapped out fuel cut relay.

I'm probably forgetting a couple of things.
Click to expand...

Operator...cleaned and inspected...;)

Good list....can you think of other parts you've wanted to try swapping?
 
Hey Jeff, when you did the coil swap did it include a different set of wires/plug caps, or did you just swap the bare coil and re-use your wires/plug caps ? If you never changed out the plug caps you could have an intermittent failure there.

edit: on second thought it sounds like the symptoms are more severe than a single plug cap failure, and the odds that multiple caps are bad would seem to be pretty low. But maybe if one cap fails it could intermittently take down both cylinders on that coil, don't know.
 
Last edited:
dwalby said:
Hey Jeff, when you did the coil swap did it include a different set of wires/plug caps, or did you just swap the bare coil and re-use your wires/plug caps ? If you never changed out the plug caps you could have an intermittent failure there.

edit: on second thought it sounds like the symptoms are more severe than a single plug cap failure, and the odds that multiple caps are bad would seem to be pretty low. But maybe if one cap fails it could intermittently take down both cylinders on that coil, don't know.
Click to expand...

Doug, I checked each with an ohm meter, they were all close to each other with the long ones having slightly higher resistance as you would expect. I also unscrewed the caps from the wires and cut off about 1/4" of the wire at that end to renew that contact. No change in resistance measurements afterward though. These leads are wire core with a resistor in the cap, so I think the wires themselves are pretty robust. No signs of arcing and no "fireworks" running the bike at the stumbling RPM in my darkened garage. No sensitivity to twisting and pulling was seen with respect to resistance measurements. All the spark plugs that I replaced looked fine, none were different / darker than the others.

Bob, I've wondered about the knock sensors thinking they are detecting knock and throwing the timing back but not sure why they would come into play under light / no throttle conditions. Of course the part / no throttle symptom makes no sense to me for the fuel pump either but I'm running out of ideas that do make sense. The other part I haven't tried swapping out is the fuel pressure regulator. The bike is running a Turbo City 58# one at the moment, but I still have the OEM one (50#) with hardly any miles on it. I did check the vacuum port on the Turbo City one with my mightyvac and it holds just fine.
 
Last edited:
Well, if you come up with something to try, let me know and I'll send you what I have (without totally dismantling the '04...of course ;)
 
wjbertrand said:
Doug, I checked each with an ohm meter, they were all close to each other with the long ones having slightly higher resistance as you would expect. I also unscrewed the caps from the wires and cut off about 1/4" of the wire at that end to renew that contact. No change in resistance measurements afterward though. These leads are wire core with a resistor in the cap, so I think the wires themselves are pretty robust. No signs of arcing and no "fireworks" running the bike at the stumbling RPM in my darkened garage. No sensitivity to twisting and pulling was seen with respect to resistance measurements. All the spark plugs that I replaced looked fine, none were different / darker than the others.
Click to expand...

If it were me I'd find a way to swap the entire coil/wire/cap combo together. I don't think taking a static resistance test is sufficient to rule out the possibility that there's an internal breakdown taking place under a very specific load. Also, the symptoms you're experiencing are exactly the same as what I had with my BMW car, perhaps a bit more severe, but otherwise identical, and that turned out to be a bad plug cap.
 
dwalby said:
If it were me I'd find a way to swap the entire coil/wire/cap combo together. I don't think taking a static resistance test is sufficient to rule out the possibility that there's an internal breakdown taking place under a very specific load. Also, the symptoms you're experiencing are exactly the same as what I had with my BMW car, perhaps a bit more severe, but otherwise identical, and that turned out to be a bad plug cap.
Click to expand...

That's the puzzling thing though, the stumble happens at no and almost no load. I can even see the misfire on the tach when coasting through the troublesome RPM with the throttle completely closed against the stop. Of course you don't feel the associated power loss stumble when coasting. If I screw on the throttle it stops immediately and accelerates strongly. None of this makes sense to me from my experience with fuel and ignitions problems.

Did your car misfire under no-load conditions?
 
Last edited:
Being that it is so RPM specific - Perhaps it is the Knock Sensor. I know with Dodge vehicles a bad knock sensor can cause some very strange problems. It could be retarding the timing at certain RPMs.
 
wjbertrand said:
That's the puzzling thing though, the stumble happens at no and almost no load. I can even see the misfire on the tach when coasting through the troublesome RPM with the throttle completely closed against the stop. Of course you don't feel the associated power loss stumble when coasting. If I screw on the throttle it stops immediately and accelerates strongly. None of this makes sense to me from my experience with fuel and ignitions problems.

Did your car misfire under no-load conditions?
Click to expand...

yes.

it misfired at a very narrow RPM band and the tach would visibly glitch when in that range, but it didn't drop down to zero like yours does. Once you got through the misfire band it could be revved to redline and was completely normal. It would accelerate strongly, just like yours does. From idle up to the misfire band it was completely normal also. What I can't remember for sure is if it was better/worse/same while driving under load, but I positively remember revving it in neutral in the garage while trying to figure out the cause.

What saved me is my friend hooked his scope up to the main distributor wire and we watched all 6 ignition pulses in real time while revving the engine. One of the pulses was intermittent and the dropouts were clearly visible on the screen. I think we swapped the bad wire to a different cylinder and watched the pulse dropout move with it. We called the BMW dealer and they told us we'd have to buy the entire 7 wire set ($200+ in mid-90s dollars) or nothing, so I passed on that option. I determined that the wires were solid core, so I went down to the local m/c shop and bought a replacement NGK plug cap for about $2.50. Cut the old cap off, screwed the NGK cap onto the wire, and the problem was solved and never came back again.

The friend with the scope wasn't all that surprised, in fact I think he predicted it would be ignition related before we started. The voltage at which the spark finally jumps is a function of cylinder pressure, so you get a voltage specific failure mode that's only present in a very narrow operational range. As you open the throttle more, the engine vacuum changes, and the whole cylinder filling dynamic changes, so the spark voltage/pressure curve changes with it. The problem can go away as you rev the engine even harder, which is counter intuitive if you suspect you have a marginal ignition component.
 
dwalby said:
yes.

it misfired at a very narrow RPM band and the tach would visibly glitch when in that range, but it didn't drop down to zero like yours does. Once you got through the misfire band it could be revved to redline and was completely normal. It would accelerate strongly, just like yours does. From idle up to the misfire band it was completely normal also. What I can't remember for sure is if it was better/worse/same while driving under load, but I positively remember revving it in neutral in the garage while trying to figure out the cause.

What saved me is my friend hooked his scope up to the main distributor wire and we watched all 6 ignition pulses in real time while revving the engine. One of the pulses was intermittent and the dropouts were clearly visible on the screen. I think we swapped the bad wire to a different cylinder and watched the pulse dropout move with it. We called the BMW dealer and they told us we'd have to buy the entire 7 wire set ($200+ in mid-90s dollars) or nothing, so I passed on that option. I determined that the wires were solid core, so I went down to the local m/c shop and bought a replacement NGK plug cap for about $2.50. Cut the old cap off, screwed the NGK cap onto the wire, and the problem was solved and never came back again.

The friend with the scope wasn't all that surprised, in fact I think he predicted it would be ignition related before we started. The voltage at which the spark finally jumps is a function of cylinder pressure, so you get a voltage specific failure mode that's only present in a very narrow operational range. As you open the throttle more, the engine vacuum changes, and the whole cylinder filling dynamic changes, so the spark voltage/pressure curve changes with it. The problem can go away as you rev the engine even harder, which is counter intuitive if you suspect you have a marginal ignition component.
Click to expand...

Well, I guess I can always try this. I have the new fuel pump in hand and the filter will be delivered today so let me do that and see if the problem persists. I can't imagine how all four caps / wires could be bad simultaneously as the power loss is dramatic when it happens, as though the kill switch was flipped off for a split second. I'm at the point of trying things that don't make sense to me anymore. Also in my experience bad secondary wires / components will provide a nice fireworks show if you run the engine under misfiring conditions in the dark. Kinda in the eliminate this and that phase just to say they've been addressed.
 
bdalameda said:
Being that it is so RPM specific - Perhaps it is the Knock Sensor. I know with Dodge vehicles a bad knock sensor can cause some very strange problems. It could be retarding the timing at certain RPMs.
Click to expand...

It would be handy if the ST had timing marks somewhere like an old car. I've got an old timing light somewhere too.
 
I have a timing light with adjustable advance/retard on the dial..

You could put a mark on the flywheel then see what happens ..

Or even with a a basic light, you could see if the timing mark moves .
 
Thanks, as the flywheel is inside the crankcase (as are all the rotating bits actually), I'm not sure how I would be able to see anything.
 
I had problems with my side stand switch, the syptoms were similar to what you describe. Engine dies, all lights etc. are still on but of course no tach because the engine had shut off, usually for a brief instant as you describe. I bypassed the switch and rode the bike a few times, problem gone. After replacing the switch the issue never came back, may be something to look at. It happened to me a couple of times at a stop, bike just quit and by putting the stand up and down a couple of times the bike would start afterward.
 
Side stand switch was the second thing I looked at after the kill switch circuit. Also, that switch is new, since the original one was destroyed in the great crash of 2013.

Regarding timing marks, the crank sensor is looking at a different disk or tone wheel that the one you use when checking valve clearance/timing.


-Jeff
 
OK, new fuel pump and new filter are in. Pretty easy job, but you have to watch not to catch the internal pump wiring under the top plate gasket. I was initially baffled why the plate wouldn't seat nice and flat. Lifted it back out and spotted the problem, tucked the wire into the tank and then the top plate plopped right into place. Glad I didn't try to force it. Anyway the bike started up and ran fine, so tomorrow will be the road test. The original FPR is on-deck as the next thing to try.
 
You must log in or register to reply here.
Back
Top Bottom