Ignition Problem?

[woodybelle]

I had a Chevy truck once with a weird miss. Turned out to be a bad crank thrust bearing which caused issues with the ckp sensor due to the reluctor ring moving out of alignment with the ckp sensor. Just a thought to maybe look at mechanical issues.

 

[wjbertrand]

Wow you have a lot of patience. I got nothing for you other than have you tried a new battery?

Have not tried a battery as the one in there seems plenty strong. Fires right up even after sitting for a couple of weeks.

 

[Jim Van]

Hi Jeff :

First off, I'm usually on the ST1100 side of the Forum, but took a look at the ST1300 side tonight and spent about 30 minutes reading the posts on this thread, your responses most closely. So, please accept my condolences !! You have one heck of a very interesting problem. I would hate for you to sell your ST1300 and admit defeat !!

I got a little excited looking at the scope tracings, being a retired EE and electronic tech with about 20 years experience. That narrowed the possibilities, since ALL coils & injectors are affected at the same time. And it was nice to see those steady crank position and camshaft position signals.

My guess is that there may be a bad connection on a connector, pinched wire at a crimp connection that is intermittent at that narrow RPM range ( I have seen this before, only at a certain vibration frequency ), OR an intermittent open IAT sensor.

I have seen connection pins that looked nice & clean, but when the connector halves are pushed together, the contacts don't make a good connection because a pin isn't seated fully into the connector. In fact, I had a problem like this on my ST1100 which caused an intermittent starter button.

Also, a friend of mine had a real interesting intermittent problem with his K75 BMW years ago. It would cut-out and then restart OK. It would happen randomly ( not repeatable, like your problem ) usually weeks apart. Turned out to be a flaky IAT sensor. When the sensor opened ( a thermistor, so an open would simulate a very low temp ) , the ECM was programmed to shut off the ignition ( ??? ). A new sensor cured the problem. I guess it's feasible you could have an intermittent connection inside the IAT only at a certain vibration frequency, or a flaky connection between the IAT and ECM.

Just for S&G's it might be worth unplugging your IAT while the bike is running and see what happens. Same for the coolant temp sensor.

I'm betting on a flaky pinched crimp connector or flaky connection at one pin at a connector that goes to the ECM, if the IAT and CT sensor checks out OK.

Good Luck !! IMHO, You have too much time & effort invested to sell the bike.

 

[wjbertrand]

Thanks for the thoughts, the CT sensor is new as is the connector , as its part of the EFI sub-harness that I just replaced. The IAT has not been replaced but checks out fine per the manual. If you unplug it the ECM returns a code 9 as expected. Plug it back in and cycle the ignition and the code disappears.

The thought of poor connector pin contact or pinched, shorted or open wire is why I replaced the entire main harness. What are the odds two harnesses would have the exact same problem at the exact same RPM?


Sent from my iPhone using Tapatalk Pro

 

[Jim Van]

The chances of that are zero, of course. But are there any connector halves that the connectors on the main harness plug into ?
Of course there is a cause for this problem. You certainly have ruled out the usual suspects.

So, when you unplug the IAT sensor, the engine keeps running but just throws the code ? At what RPM did you try that at , idle or higher ?

Even if the IAT checks OK with a meter doesn't mean it doesn't have an intermittent connection when the bike is running. I know it's a long shot, but you may want to replace it, or swap in a loaner IAT and see what happens.

Did you also try disconnecting the coolant sensor while the bike is running ?

But I would check out the below, before you do the above.

You could have your friend scope all the inputs to the ECM, besides the crank & camshaft input signals, especially the 12 volts that feed a bunch of stuff, including the ECM, coils, fuel injectors, the service connector, the exhaust air inject solenoid valve, and a bunch of relays ( I have a ST1300 service manual ) . You could disconnect that solenoid valve connector and see what happens.

Wouldn't it be interesting if a relay was vibrating at 4,250 RPM and affecting the 12 volts somehow ? I just noticed on the schematic that the 12 volts for the ECM, coils, and fuel injectors is fed by the bank angle sensor relay ......very interesting. Can't remember if you ruled out the bank angle sensor relay . Did you try removing it and replace it with a jumper ?

 

[wjbertrand]

Well I found out about the IAT error because I wanted to start the bike up and test it before putting the air cleaner assembly back on. The FI light started blinking the moment I turned the key on and continued after engine was started. When I plugged it back in, with the engine fast idling (cold) the light continued to blink until I shut it off and restarted it. No more blinking light after a key cycle which is the expected behavior. I ruled out the relays a while back by switching them around, there are 5 or 6 of them under the left side panel that are identical. Besides that the replacement harness came with a whole set of relays as well, so they've all been changed. For the BAS I tried both jumping and then replacing it, with no result. I also noticed it was common to the coils and injectors. Honestly, the only thing I've not replaced is the battery. But if it were that, why would it care if the engine is warm or cold or what RPMs the engine is running at?? The problem does not occur when the engine is cold, nor if the throttle is opened more widely. Only fully warm, only at very light or closed throttle and only at 4000-4250 RPM.

I have a recurring headache over this. Note that I have it for sale as a project bike over in the classifieds section. If someone can figure it out it's otherwise a nice bike.

 

[gsforster]

Just thought of something that may be worth a try as I just found a couple vacuum leaks on my 1300 tonight. I was trying to do a starter valve sync but was having trouble with the levels staying put when rpms were slowly raised and then let to idle again - readings would fluctuate. So I thought maybe I had a vacuum leak which should show at idle as vacuum would be strongest then and have the largest effect on mixture. I sprayed some brake cleaner around the intake boots and low and behold cyls 3 & 2 both showed dramatic rpm change and even popping when sucking the brake cleaner through the leaking boots. Pulled throttle bodies (much easier to do when engine is hot still) and pulled a boot off - it is stiff as concrete and badly cracked up. I have had vacuum leaks for a while and I just knew something was off but wasn't sure what. Now I'm pretty sure I know, and I'm ordering all 4 new intake boots.

I just wonder if this may be what is going on with yours even at 4250 rpms = maybe a vacuum leak causing lean misfire? Grab the can of brake cleaner and give each boot a little shot see what gives....
Geoff

 

[wjbertrand]

Thanks for thinking about it. My boots look good and no change spraying stuff around them. Also, we've previously documented this as an electrical problem. Further, dynamometer testing shows correct fueling.


Sent from my iPhone using Tapatalk Pro

 

[dwalby]

Geoff,

somewhere in the 317 posts there is a multi-channel scope trace that shows the exact symptom. The ECU appears to be getting some bad input and is purposely cutting out the ignition for a very specific amount of time each time the misfire occurs. The troubleshooting problem is without documentation from Honda to explain what all the inputs to the ECU are for, and on which wires, there's nothing left to do but guess. And access to the wires is limited because they go into a connector and don't have a direct probe attachment point (unless you push needles through the insulation). Honda makes a very expensive (~$1500) debug connector to get access to the wires, but its not available anywhere at any price apparently, Jeff tried some dealerships and Honda directly IIRC and nobody had one.

 

[Uncle Phil]

Just a thought - could you get the same connectors - reverse gender maybe- and make an adapter with leads from each wire to pigtail off that would insert into the circuit for testing?

 

[wjbertrand]

That's basically what the diagnostic harness appears to be. Can't explain the exorbitant price though. I'm sure it could be done, but I am done. Bike is listed in the classified section here. Make me an offer.


Sent from my iPhone using Tapatalk Pro

 

[peterswa]

My suggestion would be to check sync between the left and right banks throttle plates of the throttle body (TB), When I took the nut off the front of the TB to install the pulley for my Audiovox CSS-100, the linkage to the right side of the TB slipped, causing the right throttle plates to be out of sync with the left. Then I would perform the Starter Valve Sync procedure. Poor TB sync can cause surging of a warm engine in the 4000-4500. range.

 

[dwalby]

there are 300+ previous posts that went over all of these possible causes, which were systematically eliminated one by one, and the final diagnosis was determined using an expensive multi-channel storage scope.

If you can troubleshoot real-time digital logic problems with a multi-channel storage scope, or logic analyzer, and happen to have one at your disposal, buy the bike from Jeff and have at it. Don't expect any documentation or help from Honda on theory of operation, they had no clue how to approach the debugging themselves.

 

[wjbertrand]

My suggestion would be to check sync between the left and right banks throttle plates of the throttle body (TB), When I took the nut off the front of the TB to install the pulley for my Audiovox CSS-100, the linkage to the right side of the TB slipped, causing the right throttle plates to be out of sync with the left. Then I would perform the Starter Valve Sync procedure. Poor TB sync can cause surging of a warm engine in the 4000-4500. range.

Never touched the right to left linkages, they were tight and secure as found. And besides the problem started all by itself and not as a result of anything I'd done. The starter valves are closed when the engine is warm, yet the problem occurs only when the engine is fully warmed. Runs perfect when cold and through the warm up phase. And of course the starter valves have been synced and re-synched. Lastly the problem is electrical in nature based on the oscilloscope testing. I'm open to further ideas, not sure I have it in me to follow up anything though.


Sent from my iPhone using Tapatalk Pro

 

[Igofar]

Could you make a list of all the things that you've done, replaced, tested, and ignored because you didn't think that could be the cause.
This may help us trouble shoot better.

 

[wjbertrand]

I could attach an updated list, to include the recent stuff I've done, but there's already a word document attached to one of my messages in this thread with 99% of the things tried listed.


Sent from my iPhone using Tapatalk Pro

 

[peterswa]

I find it interesting that considering all you've done there hasn't been any change in the failure. The only thing not touched that I can see is the Combination Meter and I'd think you've pulled the fuse for the cruise control. Very interesting and very perplexing.

An interesting point is that failure seems to be linked to a warm engine. Heat can reduce insulation around any wire when the wire or bundle of wires is bound so tightly that it compresses the insulation around the wires. When heated a cross coupling can occur corrupting the integrity of ground, voltage or signal wire, without any sign of a physical short. A harness that I would look at provides the wiring around the top of the engine that connects to the MAP, CMP, TP, IAT sensors and wiring to the injuctors.

 

[wjbertrand]

The other thing is that most EFI engines run in open loop mode when cold. In this mode the engine runs on an internal pre-programmed map and does not depend on sensor inputs. Once warm, the ECM switches to closed loop mode but only when running at below about 1/2 redline and under part/light throttle. Even when fully warmed, the engine will run open loop at RPMs above 1/2 redline and/or heavier throttle application. Since the redline is 8500 RPM, 4250 is right at the half way point, which is where the cutting out occurs. The engine runs perfect under any condition where it's presumably in open loop mode. I think there's a problem when the ECM is changing between modes. Three different ECMs, all sensors replaced and all three harnesses replaced doesn't touch it though.


Sent from my iPhone using Tapatalk Pro

 

[DavidR8]

I know you're at the end of the line with this problem but your mention of the 4250 being where the switch from open to closed occurs got me thinking.
When it switches from open to closed loop the O2 sensors come into play. I wonder if some other sensor(s) also come online at the same time. And maybe one of them is malfunctioning.

 

[wjbertrand]

Decided to post an updated liST of attempted repairs: