Ignition Problem?

[woodybelle]

I like your thinking Geoff. Yes the tach signal is supplied to the cluster directly from the ecm, and the fuses B, J and F feed the cluster and none of those supply the coils or injectors. Do you know what part the cam signal plays. Is it only used to sync the injectors with the cam? I would agree that a check of the crank sensor signal is in order, while the symptoms are present.

 

[wjbertrand]

So when they fail to fire do the injectors and coil lose the signal to fire from the ecm or do they lose the voltage from the bank angle sensor relay?

My understanding is that the injectors are directly powered by the ECM, I don't see any other source of power to them in the wiring schematic. The primary side of the coils of course have to be grounded to fire and the tracings show that there's no grounding occurring. Can't be the BAS sensor as the same misfire happens with it out of the circuit.

 

[wjbertrand]

Then I would think it has to be a problem with the 'processor' or the connections from the 'processor', but then you've replaced all of that. It almost sounds like the 'mapping' has a 'blank' place in it. If it were a desktop computer, I'd say it was having bit failures in the memory - sort of like 'electronic' Alzheimer's.

That was almost the very first thing I was ready to purchase until Hal and I swapped ECMs. His bike ran perfect with my ECM installed in it.

 

[wjbertrand]

I was just thinking some more about this. I'm trying to go down a path of diagnostic based on what evidence you have gathered. I was going to say test the supply side of the injectors and coils for dropout when miss occurs, but because you say the tach drops to zero also has me thinking it may be a ecm input problem as I doubt the supply circuit is the same for just the tach and the coils/injectors (although I could be wrong have not looked at diagram). So that leads us to the ECM - nope, you swapped that already. So then the crank sensor or it's wiring - nope you swapped all that too. It would be really nice if you were able to backprobe the crank sensor lines at the ECM and scope that signal for observation when misfire occurs. What I'm left with is a question regarding the mechanical condition of the tone ring on the crank for the crank sensor. Maybe it's loose or has damage and only vibrates at some sort of resonant frequency at certain rpm? If you have already checked into this please forgive me there are too many posts to go back through....

So here is the oscilloscope back probe of one of the crank position sensor, trace #4 on the chart. The misfire is shown by the gaps in traces 1 & 2 (both coils). Trace #3 is the cam position sensor back probe. Those signals seem very steady when the misfire occurs. The second chart shows both coils and one of the injectors. I get the same result from all four injectors, just the timing to the coils moves a little. The dropouts always happen simultaneously.

This data was collected at the sensor/coil/injector end of the harness so left the question whether the signals were reaching the ECM correctly. Changing the harness seems to confirm that they are. I inspected the tone rings for the crank and cam using a flash light and rotating the engine 360 degrees. No missing, corroded, worn or bent teeth are evident and sticking my finger in and pushing on the rings does not detect any looseness.

I even considered the cam that carries the tone ring might be a tooth off on timing but checking it twice shows it's dead on.

I agree that the ECM is probably receiving some kind of faulty input, maybe the alternator?

 

[Uncle Phil]

Jeff - Since my memory ain't what it used to be, is this the bike that crashed and did the problem start after the crash? Just grasping at straws here, but in my software development, I always start at a 'known' spot then proceed to the 'problem' and try to see what has changed.

 

[wjbertrand]

Jeff - Since my memory ain't what it used to be, is this the bike that crashed and did the problem start after the crash? Just grasping at straws here, but in my software development, I always start at a 'known' spot then proceed to the 'problem' and try to see what has changed.

Yes bike was crashed and totaled in 2013 due to the cost of cosmetic repairs and it's age and mileage. The problem with the misfire did not occur until 2 years and about 20,000 miles later however. The ST's harness & sensor locations seem to be pretty well protected from crashing into surfaces (cars/roadways), there would have to be some kind of penetrating damage to reach them. That was not the case in my crash.

 

[Uncle Phil]

Jeff - I was reading back and saw that. Sorry for my mental 'laziness'. If the problem occurs consistently at 4,250 RPMs, seems like it would be electrical and not mechanical. I know you've swapped ECMs, new wiring harness, tested the coils, etc. Another dumb question (a little far out) - have you disconnected the tach and tried it? I know it shouldn't have much to do with it, but maybe a backfeed problem? It sure sounds like something causing electric noise in the circuit.

 

[wjbertrand]

Jeff - I was reading back and saw that. Sorry for my mental 'laziness'. If the problem occurs consistently at 4,250 RPMs, seems like it would be electrical and not mechanical. I know you've swapped ECMs, new wiring harness, tested the coils, etc. Another dumb question (a little far out) - have you disconnected the tach and tried it? I know it shouldn't have much to do with it, but maybe a backfeed problem? It sure sounds like something causing electric noise in the circuit.

Others have suggested that too, but I'm struggling to explain why the misfire does not occur until the engine is fully warm and only at light to closed throttle conditions if it were the tach. The engine runs perfect the first 5 minutes or so. and it doesn't matter if the engine is cold or warm, it does not miss at that RPM when accelerating, only when trying to maintain that RPM which unfortunately is right at 70-75 MPH in top gear.

I've disconnected the harness plug at the back of the instrument panel, inspected it and flushed it out with contact cleaner, also cleaned and inspected all the grounds & ground block connector under the nose faring as well. Nothing suspicious there. I guess I could locate the wire within the harness that leads to the tach and cut it, but...

 

[Uncle Phil]

So if it is 'warm', then you would suspect that some electronic component is 'changing' behavior once the bike gets up to operating temps. Just curious - did you 'run it' with the plug loose (I don't know if that is possible on a ST1300)? Don't know if this would be worth fooling with, but maybe using a hairdryer on some of the components to raise the temps to see if they fail when scoped? Does the bike have to be 'rolling' for the problem to occur, or can you reproduce it in the 'laboratory' on the center stand?

 

[wjbertrand]

So if it is 'warm', then you would suspect that some electronic component is 'changing' behavior once the bike gets up to operating temps. Just curious - did you 'run it' with the plug loose (I don't know if that is possible on a ST1300)? Don't know if this would be worth fooling with, but maybe using a hairdryer on some of the components to raise the temps to see if they fail when scoped? Does the bike have to be 'rolling' for the problem to occur, or can you reproduce it in the 'laboratory' on the center stand?

Plug loose? The problem can be reproduced on the Center Stand just revving to 4250 RPM and trying to hold it there, doesn't need to be on the road. As far as the hair drier approach, I'm at a loss of what component to heat up that hasn't already been replaced!

 

[gsforster]

Hmmm. Keeps coming back to bank sensor? The ecm has stopped providing the ground pulse in the scope shots and that can only be caused by either the Ecm momentarily losing its ground or some other input telling it to cut control of those lines line a bank sensor or something. It clearly did not lose ckp or cmp inputs. Sorry for the ramble thinking out loud.

 

[Uncle Phil]

Plug loose? The problem can be reproduced on the Center Stand just revving to 4250 RPM and trying to hold it there, doesn't need to be on the road. As far as the hair drier approach, I'm at a loss of what component to heat up that hasn't already been replaced!

My only thought with the hairdryer was maybe to check the components without the engine running just to see if anything 'changed' when you did your measurements for voltage and such. If it works until the bike 'warms up', you'd think that heat would be messing with one or more components. It sure seems like it must be a combination of things since you've replaced everything except the bike itself. The precision of the RPM where it fails it what seems to point that it is not a ground issue or you would think the RPM would vary. It's like some component is sending a signal to the ECM at that rpm to 'pause' since you've changed out the ECM and the wiring harness and it had no effect.

 

[woodybelle]

Is there any way to disconnect the black and grey wire from the alternator? This wire excites the alternator and also provides power to the bank angle sensor. This would rule out the alternator as a cause I believe. It may be that the problem can only occur in the transition from closed loop to open loop and that is why it is at a specific rpm. It seems that the ecm is seeing something it does not like but there are not that many things one can think of that would cause this.

 

[wjbertrand]

Is there any way to disconnect the black and grey wire from the alternator? This wire excites the alternator and also provides power to the bank angle sensor. This would rule out the alternator as a cause I believe.

So in doing that the bike then runs off of the battery? I'll have to look my schematic and see but I suspect that wire is part of a large harness plug.

 

[woodybelle]

Yes the bike would be running off just the battery. It should be ok long enough to test to see if disconnecting the alternator has any effect. The schematic show only 2 wires on the alternator. I don't know if it is 2 plugs or one, the big red wire is usually held by a stud and nut on most alternators.

 

[woodybelle]

My understanding is that the injectors are directly powered by the ECM, I don't see any other source of power to them in the wiring schematic.

The injectors are powered by the bank angle sensor relay and grounded by the ecm, just like the coils.

 

[Blrfl]

Hmmm. Keeps coming back to bank sensor?

Two things make me want to discount that and at the same time give it a closer look:

The bank angle sensor triggers a hard cut of pretty much everything important, and nothing comes back until the ignition switch has been turned off and then on. There are mechanical bits involved, and it's possible that this is a short enough thing for the relays involved to lose contact just long enough to make the ECM reboot but not long enough for everything to get locked down.

That said, the sensors that signal what position the engine is in don't skip a beat, and they all get their power from the ECM. But... It's entirely possible that there's enough capacitance in the ECM to carry them through a couple of revolutions but not enough for the processor.

Whatever this is, it's clearly very brief and would benefit from a few more hours of 'scope time to investigate the ECM power source (going back to what I said in post 18). I'm also becoming curious (again) about what the throttle position sensor is feeding it.

If I had the time and cash to burn, I'd fly out there and help him out with it because I've been staring at this for a year and I'm dying to know what the problem is.

--Mark

 

[wjbertrand]

Two things make me want to discount that and at the same time give it a closer look:

The bank angle sensor triggers a hard cut of pretty much everything important, and nothing comes back until the ignition switch has been turned off and then on. There are mechanical bits involved, and it's possible that this is a short enough thing for the relays involved to lose contact just long enough to make the ECM reboot but not long enough for everything to get locked down.

That said, the sensors that signal what position the engine is in don't skip a beat, and they all get their power from the ECM. But... It's entirely possible that there's enough capacitance in the ECM to carry them through a couple of revolutions but not enough for the processor.

Whatever this is, it's clearly very brief and would benefit from a few more hours of 'scope time to investigate the ECM power source (going back to what I said in post 18). I'm also becoming curious (again) about what the throttle position sensor is feeding it.

If I had the time and cash to burn, I'd fly out there and help him out with it because I've been staring at this for a year and I'm dying to know what the problem is.

--Mark

So I was thinking about the things you said in post 18 when I started shopping for a used harness, i.e. something in the power path to the ECM. That entire path has now been replaced and the rarity of this problem makes it really unlikely that a similar fault exists in the replacement path as might have been in the original harness/path. Thinking of pulling the alternator and replacing the brushes, regulator and rectifier. Turns out these parts are common to Honda cars and so have alternate, much cheaper aftermarket sources.

Regarding the TPS, recall that it has been replaced with a brand new one. If I disconnect the TPS, I get an error code and the ECM resorts to a fall back value, yet the misfire still persists.

 

[Uncle Phil]

The alternator angle makes a lot of sense since it would vary with the RPMS. Maybe it creates enough electrical noise at that particular RPM to cause the problem.

 

[wjbertrand]

The alternator angle makes a lot of sense since it would vary with the RPMS. Maybe it creates enough electrical noise at that particular RPM to cause the problem.

Given the narrow, specific range at which it happens, if it's vibration caused, it makes me think it would have to be something pretty rigid / hard-bolted to the engine. The alternator is really the last thing on the liST. I've kind of put it there because googling "can a faulty alternator cause a misfire?" pretty much comes back with the answer only if it's failing to keep the system voltage high enough to run the various systems. In my case it always puts out more than 14V (14.1-14.3) but I'm measuring this at the battery which is probably a pretty big damper for rapid events. Also, I'm not sure what the sampling rate is for my Datel on-board meter or my hand held multimeter. Maybe it;s not frequent enough to catch a fleeting event? I have noticed that the Datel meter will sometimes gain 0.1V momentarily immediately following a misfire event, but it's not consistent. I wish when I'd had the oscilloscope on hand that we had looked at the character of the alternator output. it's been suggested that if the output looks "hashy" on the scope it could screw with the ECM or be producing interfering RF noise. I dunno...