Background - for about the past 30 months - my bike tended to have a crappy idle. It would idle roughly enough that the transmission input shaft clattered at idle if the clutch was engaged (and the shaft spinning.) Gear-rattle sort of noise. Not unknown certainly, but more of it then I'd expect, or remembered when the bike was new. The bike also had a bit of surge in 3rd and 4th gear at lower RPMs - again not awful like the R11XX bikes did, but enough to be noticeable and make it a bit harder to ride smoothly at lower RPMs. And 5th gear had a periodic buzz/drone - at about 1 second intervals. Could feel it through the footpegs. Again - not awful, didn't get any worse, I pretty much just ignored it figuring it was some odd resonance perhaps from the exhaust system.
Finally - I got annoyed at the idle, and decided to take a look at the bike with my GS-911. The current GS-911 software has the ability to watch engine readings in real time - from all sorts of sensors, and from the BMS-K (think ECU..) I had hooked it up when the bike was idling crappy in my garage, and somehow spotted that the O2 sensors seemed different. Luckily - the new software also has an O2 plotting function - so I switch that on, and what was happening became quite evident. The #2 (which is the left side) O2 sensor was dropping out - signal loss. When it lost the signal the bike idled crappy. When the signal came back the bike idled smoothly.
So.. I started trying to track down what was causing this. My initial thought was a lazy O2 sensor. I disconnected the sensor and ran the bike, and saw that when the BMS-K didn't see any signal from a sensor - it set the mixture value to a default value (about 0.4V - a good O2 sensor swings at one second intervals between 0.1-0.85V or so.) It didn't use this value when the sensor was plugged in - it lost the signal completely.
I then reconnected the sensor - and lo! and behold! it started working again. And the idle smoothed right out. Took it for a ride - the surge was gone, the 5th gear drone was gone, and the bike felt much different - less vibration (it never had reached the level my old RS was at.. but there was some..), overall - felt much better.
SO - I figured it was the connector giving problems, and disconnecting and reconnecting it fixed it.
Not so. Went for a ride last Saturday - about 150 miles over 4 hours or so - mostly all back road stuff - and about mid-day I could feel the character of the bike change (I was now "attuned" to it..) And the crappy idle came back, the drone came back, some buzz came back. So - it wasn't the connector.
Saturday night - next try - a new O2 sensor. It just happened that I had several "spare" R12R O2 sensors kicking around (part of my obsessive buying of spare parts off Ebay - I got two sensors and a complete harness for $50.. about $2,000 at BMW prices..) So - Saturday night - I replaced the sensor. Checked with the GS-911 and all appeared well.
Sunday's ride was longer - bike felt wonderful for the first 130 miles or so - then it changed character again, and was buzzy, crappy idle, drone for the next 120 miles.
SO - again - looking at it in the garage, I decided that since twice - disconnecting/reconnecting the O2 sensor "fixed" it - I'd use some of the magic "Stabilant-22" contact enhancer (stuff is VERY - VERY - expensive, but good) on the O2 sensor connector. While I was moving stuff around to dose the connectors - I spotted something:
See the bright spot on the rib on the engine? In front of it is the harness coming from the left side of the bike. It has the O2 sensor, the throttle-position-sensor and the idle-stepper-motor wiring all in it. The harness didn't used to have the plastic corregated sleeve over it - I put that on. It had been simply cloth tape around the wiring. And it was pressed against the rib by the crankcase vent hose going from the left-cylinder to the airbox.
I think I found the problem. I didn't see any bare wire, but I intially did see the insulated wiring between turns in the cloth tape. I suspect the O2 sensor signal wire had rubbed through the insulation and was grounding out on the engine rib. When I disconnected the plug for the O2 sensor it moved the harness enough to temporarily eliminate the grounding out. I put the plastic sleeving (it's split so it's easy to put on) over that section of the harness so this wouldn't happen again.
So - what I'm curious about - does anyone else have the same bright spot? You do have to remove the crankcase hose, at least at the airbox end to really see it since it's right behind the hose. The crankcase hose uses clamps that are easily released with a set of slip-jaw pliers (you pull the two tangs sticking out toward each other - opening the clamp then wiggle it down the hose past the end of the stub the hose goes to.)
I'd also suggest that you check the tie-wraps holding some of the harness to the bottom of the throttle bodies. As is very common with BMW - those tie wraps were WAY too tight. I snipped mine and replaced them with new ones, tightened just enough to keep the wire in place - not so tight as to pinch it.
Curious what other might find..
