Oh no - smoke...

[Northman]

Both coils test the same, spark to all plugs... but the voltage (at + side of coils) doesn't rise above the battery level and it should go to 100v minimum when the starter motor spins up...

Huh? :???:

The positive side of the coils will be battery voltage, with the negative as your trigger from the ECU. The ECU will not produce a weak spark, just no spark.

Try swapping the coils and see if the problem switches to the other two plugs.

[Northman]

Are you using a "peak voltage adapter" for your testing, or just a multimeter?

[jrdxx]

I generally don't like to swap parts around as a troubleshooting method, but I don't believe the coil has been eliminated as a failure mode. Do like mentioned above, and swap the coils, and see if the misfiring cylinders moves to the 1-4 positions.

[ironmike]

Swapped the coils - no change (coils are also covered as "emission components).

I am using a multimeter. The voltage at the + side should be the same as the battery (12.8v) w/key on; when the starter is spun, the voltage should go to 100v (minimum). If it doesn't, the ECM is suspect (Manual, 17-3).

Then, per the manual, you test the ECM (gray) connector (yellow + and white/yellow -) for 0.7v when the starter spins - I got 0.0v...

Then you test the ignition pulse generator for peak voltage (0.7v) when the starter spins - I got 0.0v...

Then you're refered back to the trouble-shooting chart (17-3) that suggests the ECM is at fault.

As it happens, all of the suspect components are covered under the "emissions warranty".

[ironmike]

The dealership's service manager called and asked me to come in... they need to talk with me. I get with the tech who's actually working on my 'Bird and we discuss the diagnosis... essentially, he's stumped.

He ran all the same tests that I did (per the shop manual) and arrived at the same conclusion - ECM glitch.

Northman, you were on to something when you asked if I was using a peak voltage reader (PVR) or a multimeter. Per the tech, a multimeter won't give the desired reading; a "special" proprietary Honda PVR is required.. (BS, I think). Nonetheless, the tech thought it's the ECM, because everything else seemed to check out.

So, he ordered and installed a new ECM. The bike fired right up on all cylinders, but #2 & #3 quit within moments... and for whatever reason, he apparently kept revving it (or at least let it run) until raw fuel flooded cylinders #2 & #3, flooding the air box - fuel spurting from the vent hose, either washing past the rings or succumbing to vacuum through the case vent, and filling the crankcase (and cylinders #2 & #3), ultimately hydro-locking the engine with fuel. The tech explained all this to me as I viewed the gas-soaked engine (throttle bodies, air box/filter, etc.) and shop floor.

At this point, he acknowledged that he hadn't a clue... but he would drain the cranckcase (add fresh oil/filter), purge the cylinders and injection system, and start over. We both tended to agree that it's most likely an electrical problem... the fuel pump sure as hell works... He made a number of comments about the regulator/rectifier being at the root of a host of electrical issues that often don't really seem to make sense. I've had no discernable problem with the R/R, and only replaced the original battery (which was working fine) when this overall problem began.

The emissions warranty is good until June '05. The ECM and many other components are all specifically covered, as is labor and diagnostic procedures. What is conspicuously absent from such coverage is the R/R.

When I know more, I'll advise.

[Northman]

How about an injector problem?

Are you sure it's losing spark, and just not injecting itself to death?

The PVR was right out of the manual. What it does is store the highest reading it saw, which is something a multimeter just isn't capable of.

I would doubt an electrical issue, myself, or at least not he R/R type. It's a problem unique to cylinders #2&3, and even a misfiring engine won't build up enough fuel to hydraulic lock the cylinders.

I'd bet on an injector problem (could be electrical, too, but definitely isolated to the center cylinders), myself, with the additional info.

If you cycle the key, do you see any fuel leakage past the injectors into the throttle bodies? Could be bad o-rings on the fuel rail even.

Keep me posted on this, you've got my interest.

[Northman]

One more point to help you in your diagnosis:

Figure this. The injectors are now pouring fuel into the engine quicker than the pistons can pump it out the exhaust valves in order to hydraulic lock the cylinders.

You have a major over fueling problem. :shock:

[Pete in PA]

When you say hyro-locking the engine, you know this can bend con. rods right? NOT GOOD!!

I'm not too hip on FI, I've got carbs after all, but I also think that's a LOT of fuel! Sounds like a leak to me also.

[ironmike]

... over-fueling is at least a symptom, but I can't help thinking there's more to it. I understand there's supposed to be contant psi at the fuel pump, but I don't know if there's a default mode in the ECM should a drop in pressure (as would be the case in a leak or an open -constant- injector.. or two) occur... sort of a "limp home" mode of reduced psi... Nobody seems to have a handle on the exact programming of the factory ECM; everything I've heard so far is admittedly speculation.

It has spark, at least at start-up... when I next talk to the tech I'll ask if he's checked for spark when the two cylinders seem to die - an inductive timing light ought to provide that answer. I'll pass on your suggestions re: fuel delivery system as well.

Now as for any damage that occured when it hydro-locked... well, we know who did that.

[Northman]

... over-fueling is at least a symptom, but I can't help thinking there's more to it.   I understand there's supposed to be contant psi at the fuel pump, but I don't know if there's a default mode in the ECM should a drop in pressure (as would be the case in a leak or an open -constant-  injector.. or two) occur... sort of a "limp home" mode of reduced psi... Nobody seems to have a handle on the exact programming of the factory ECM; everything I've heard so far is admittedly speculation.

There is no sensor for fuel pressure, so the ECU has no way of knowing if the fuel pressure has dropped, or risen. My bike, for instance, is now running fuel pressure 7psi higher than factory. I did this to increase fuel delivery when the motor was modified.

There is no oxygen sensor on your bike, either, to tell the ECU that the bike is running lean or rich. Even if there was, the ECU does not control fuel pressure. That's the job of the fuel pressure regulator, which uses a spring & vacuum to control fuel pressure mechanically, not electronically.

Let me check into the wiring diagram, and post back later. You've piqued my interest here :grin:

[Northman]

Well, according to the wiring diagram, each injector uses a separate trigger wire for control coming from the ECU, and a common ground shared with coils, IAT, PAIR valve, fuel stop relay, etc. That would pretty much eliminate wiring as the problem, unless both wires were shorted somehow, which I doubt. The ECU was swapped out with the same result, so that eliminates it as the culprit here.

Try the leakdown test I mentioned before. Cycle the key, with the fuel pump connected, and check for fuel leaking past the injectors into the cylinders.

I'll bet on either bad o-rings on the injectors (you mentioned sitting a while), or they are stuck in the open position.

After all of this flooding, make sure to change your oil before running the engine again. A lot of fuel can wash the lubricant right off of the bearings, not to mention ignite in the oil pan! :shock:

[ironmike]

... although the shop is closed until next Tuesday, so I know it'll be mid-week before I get with the tech again... Thanks, Northman..

[Red J]

I'd disconnect the injectors, clean out the combustion chamber, and see if it's overfueling with the injectors disconnected. If they are, send them off to RC engineering or whomever, to get tested and cleaned.

No ECU control on injector + injector still injecting = problem with injector.

If they don't inject (or, in this case, pump) fuel into the cylinders, you can check the injector drive with a LED, possibly with a resistor wired in line. I've tested for this before, but I don't know who built the LED and determined that it needed a resistor.

I would definitely short-change the oil a few times after you have this sorted.

J.

[Red J]

What it does is store the highest reading it saw, which is something a multimeter just isn't capable of.

Some DMMs have peak hold function...

[ironmike]

The primary culprit was a diaphram membrane in the fuel pressure regulator (16740-MAT-E01 6112572) that was slowly deteriorating. Symptoms were not even noticable in the early stages; it wasn't until sufficient fuel began getting past the membrane that the problem presented as over-fueling cylinders #2 & #3 through the attached vacuum lines. In the early stages, the bike would start and run for some time at normal operating temps, but within less and less time (minutes) the plugs in #2 & #3 would start to foul and eventually cease to fire.

The ECM doesn't have a fuel pressure sensor, so it had no specific input as to the overfueling issue, but the rest of the senror array readings deviated sufficiently from the norm that the ECM went into its "fail-safe" (limp home) mode (Manual 5-2) but didn't record ("throw") a code. So, the ECM initially seemed to be part of the problem.

As the membrane lost more of its integrity, more fuel escaped into the vacuum lines runninng directly to cylinders #2 & #3... and the severity of the symptoms escalated until #2 & #3 wouldn't fire shortly after starting.

Kudos to Northman, he suspected the right track to investigate... while we on the scene were predisposed (and fully expected) to find a purely electrical problem/solution. It was the testing of individual injectors that led to the pressure regulator.

According to Honda USA, this is the first such regulator failure on a CBR1100XX... that hasn't been... uh, "enhanced" for performance purposes.

As for the hydo-locking, that happened when the tech was trying to start the engine and over-pressurized fuel filled cylinders #2 & #3 - the starter couldn't overcome that. The engine was checked for any damage, none was found; but if any such evidence presents later, it'll likely be related to this event and therefore covered as a shop fault/error.

And the best news - all parts, labor, and diagnostic testing was covered under the 5-year emissions warranty. Of course, I fully tested the bike upon its discharge from the shop today.... oh yes, it is quite healthy. :wink:

My thanks to all for your suggestions, feedback, and support. We have a great little community here! Happy Holidays!

[JasonW]

Thanks for the update. Glad to hear you got it sorted. :grin:

[Northman]

Thanks for the update. Glad to hear you got it sorted.

Same here, happy to help :cheers: