Took my 1994 K75RT out this morning. Two previous 100 mile rides this spring were flawless.   First thing, starting out the ABS fault lights started flashing.  This was the second time in about two weeks it did this.  Shut off the bike and checked the connector to the brain which was tight.  Restarted the bike, and rode away, no flashing.  Same thing happened last week on a previous ride.  Reason for checking the connector was that both times I had been digging around in the stuff I keep in the tail cowl prior to going and I suspected I might have disturbed the connector.

Rode about 4 miles, and ABS fault starts flashing.  Ride another mile or so and flashing stops.  Another mile and the alternator light comes on and the tach needle drops.  The bike is losing power but the tach is dropping faster than the engine rev's.  After a second or two, everything returns to normal.

Ride a bit further and the taillight bulb warning light flashes for an instant.  It flashes randomly an additional 7-8 times during the 10 minute ride home.  When I got home I did a couple quick checks to see if the battery had a good charge.  Not having a lot of time or a meter handy, I just did a couple of starts.  With the engine hot, the starter had no trouble turning the motor over, and connecting my maintainer showed that it went immediately to float mode.  I left the maintainer connected and left for work.

This evening when I got home I tried to do some checking and unearthed some new, possibly related problems.  

Since it's too late to take it out for a test ride, I just tried starting again to see if the battery would spin a cold motor.  Several cold starts in succession and no problem, then suddenly, on one of the starts the starter would only turn the engine part of a revolution, and then stop.  Next try, everything starts normally. 

OK, I am thinking the battery is good, so I test the tail lights to see if the BMU was trying to tell me a bulb was out.  I turn on the ignition, touch the rear brake, and the instrument cluster goes dead with only a faint glow coming from the alternator idiot light.  Turn off the ignition, turn on and try again, everything normal and brake light works. 

I do this another 20-25 times and find that touching either the front or rear brake will randomly kill the power to the Instrument cluster.  Additionally, (this only happened twice), just turning on the ignition switch caused the cluster to die.

The last time I got the cluster to die it was when I turned on the ignition switch.  I wanted to see if there was power anywhere else so I touched the front brake to see if the brake light would come on.
 Not only did the light come on, but the cluster came on as well.

I suspect the BMU, but have no idea of how it is wired or how it might be able to kill the bike's power.  The fact that I seem to be dealing with several possibly related intermittent problems has me pretty confused right now.  Have spent some time looking at the essentially useless wiring diagrams in the Clymer book and am now hoping that someone here might have seen a similar problem.  

I have tried to make this as clear as possible given how random everything is behaving right now.  I'm supposed to be riding out to Colorado in a week, and would hate to have to cancel the trip because of this.

Update, when the instrument cluster goes dead it is only the idiot lights and the gear indicator.  The dial illumination still works as does the parking lights. 

When the idiot lights are dead, starter does not work either.

Have you considered that it might be a carbon dusted starter motor which either needs brushes or just a clean out with contact or brake parts cleaner?

Worn brushes was the first thing I thought of, except that the starter runs really strong.  I'm just wondering if anyone had any other suggestions of things to look for and tests to try before I tear the bike apart and can't run it.  

I guess the plan today is to get at the starter and the BMU.

Point-Seven-five wrote:Update, when the instrument cluster goes dead it is only the idiot lights and the gear indicator.  The dial illumination still works as does the parking lights. 

When the idiot lights are dead, starter does not work either.

Battery?. I had issues on a car like that and the particular forum identified loads of issues that could arise from a failing battery even though the battery would appear to work as normal. Trust nothing until you know for certain the battery is good.

kinda sounds like the instruments connectors corroded - intermittent

the starter button gets a feed from there as well

but before you do anything ...wiggle the fuses ...or clean the sockets that they fit into ....then test each fuse for voltage drop with ignition on
lastly visually inspect the fuses removed before testing each one with a meter

you have cleaned up all the under tank connectors haven't you ?

as an aside  was thinking of asking that "stickers man" to consider an English version fuse naming sticker that overlays the fuse cover


hope that helps a little

Update:  Starter pulled and cleaned.  Brushes are 10.5 and 11mm long as measured to the low point of the curve where they bear on the commutator.  As would be expected with such light wear, there was virtually no carbon dust to be found.

Starter and battery terminals cleaned and dielectric grease applied. 

Battery voltage at rest with surface charge removed is 12.9v.  Battery voltage with ignition switch and lights on but engine not running is 12.4v.

Battery voltage with starter cranking is 11.5v. 

Battery is good.

After servicing starter and battery terminals the application of brake with engine not running but ignition on caused entire electrical system to die.  Holding brake, the system came back in approximately 3-4 seconds.  Only got it to happen once in about 30 tries.  Odd in that it hasn't blown any fuses.

Next test is to read the battery voltage when the electrical system is dead.

This may take some time...

Edit:  Changed "engine off" to "engine not running".

Update:  "Normal" battery voltage with ignition and all lights on, engine not running and brake applied 12.2v.

Battery voltage with electrical system "dead" 12.2v.  Note:  I am testing with a digital meter, and the update time is not showing transients well.  No analog meter available.  I suspect the 12.2v reading I saw was after the system came back.  I am thinking I need to connect a light bulb that might indicate what is going on by dimming.

So far, problem doesn't appear to be loading down the battery which might indicate the something is being switched off somewhere. 

Regarding corrosion in the instrument cluster connector, I will check, but the electrical system on this bike is in excellent condition with little or no corrosion anywhere.  I went through and cleaned and greased almost all electrical connections in the past year.  Especially the instrument connector which was done just last winter.  Being an RT I am loathe to go after the instrument connector as I will probably have to pull the fairing to get at it to do any worthwhile service to it.

Point-Seven-five wrote:
After servicing starter and battery terminals the application of brake with engine not running but ignition on caused entire electrical system to die.  Holding brake, the system came back in approximately 3-4 seconds.  Only got it to happen once in about 30 tries.  Odd in that it hasn't blown any fuses.

hope that top fuse was only rated for 7.5 amp

my wiring system nearly caught on fire when my hand brake switch failed (internal short to earth) and I tried to put in a 20 amp to locate the issue  ( I just happened to have at hand ) before getting the correct one and disconnecting  the switch from the loom
I thought myself lucky that the ignition switch still worked after the event
I wonder if the current limit caused by the brake switch fault is consistent with the condition you are experiencing now ?

additional info

to get at the instruments no need to pull the fairing ...just remove the left hand pocket ...then you may find that removing the 2x  6mm allen screws that hold the instruments bracket will let you get at it
else if a little difficult, you can pull the screen and inner liner and other pocket  to get better access  without a major pull down
hope that helps

Wondering about the occasional kill switch fault or a loose/duff connection inside the switch gear. Front brake causes some very slight movement, could it be enough to cause a problem?  Electrics seem to read good in order areas.....

Charlie, top fuse is 7.5 amp and hasn't blown.  None of the other fuses are blown or bad and their terminals look good.  For what it's worth, the starter button works properly every time and the starter spins the engine effortlessly.  Engine starts nearly every time in less than 1 second of cranking.  Ambient temperature is approximately 20C.

I have pulled the BMU, and a visual inspection can't find anything suspicious.  It is the new style with the metal strips instead of the reed relays.  All solder joints are solid, clean and shiny as are the connectors. 

Presently, I have pulled the tail light.  It has a 20w halogen brake flasher retrofit which I am tempted to replace with a conventional bulb.  Will be replacing all bulbs in the tail light.  Not a bad idea since I don't know how old they are.

While I have the ABS brain out, I will be cleaning the connector again with DeOxit, just to be sure since the original symptom was an ABS warning. 

Don't know if I have fixed the problem, because I haven't been able to make it fail once in the past two hours of messing with things.  In a way I hope not as I haven't really found anything wrong yet.

Just went for a 35 mile ride and everything worked properly.  Wish I knew what I did.  3000 mile trip starting next Monday.  This sure wasn't a confidence builder.

Your symptoms appeared to be very erratic and initially it sounded to me like a condition like corrosion, rather than a fault or fail somewhere. I think that by pulling the BMU you have maybe fixed the problem. But I think Charlie's suggestion to clean the instrument connection would be a good move.

Re: 3000mi trip. Make sure you have some way to connect with the forum and pack a good toolkit.

Have a safe journey

I reckon have enough tools to be able to remove the gas tank, fairing panels, battery and levers and wheels. Its not a lot and that gives you access to all your electrical connections plus allows a puncture repair. For what its worth I bring a clutch cable and after an incident now carry spare gear lever and a brake/clutch lever. If the bike falls or gets knocked over they can break and leave you stranded. The relay box has room for spare bulbs and fuses.

Just lost a long post.  Suffice to say that I have extensive spares with me along with enough tools to get rather far into her.  Also taking a small laptop so I can lean on the brain trust here if things awry.

Want to thank you guys for the help today.  Checked some things I might not have had it not for your suggestions. 

Knock on wood, but she was running quite nicely today at 65-70 mph so I hope she is as eager to get some miles behind us as I am.

Cheers

Enjoy the trip and safe travels.!! Use the laptop to post some photos too.

It's back.

Tried bike this morning.  First time ignition on, electrical system goes dead...for about 3-4 seconds and then comes on.  Did it twice in 5-6 tries.

Then tried the brake with ignition on and electrical system live.  Pulling the brake killed the electrical system.  Held the brake and after 3-4 seconds system came back.

No blown fuses and battery is strong.  System acts the same when dead as when the "kill" switch is off, except that the ABS light doesn't flash like it would with the "kill" off and things come back in 3-4 seconds.

I suspect something in the interaction of the cluster lights between the ABS and the BMU, but I can't find any electrical diagrams that show both systems together.  I also don't know where to look to find out the operational checks the ABS and BMU systems do on start up.  Right now I'm thinking something is hanging up during the start up checks.

I don't suspect any bad connections at this point.  This bike has been very well cared for and has been garaged for it's entire life.  I have yet to find a dirty or even lightly corroded connection anywhere on this bike and I have been into almost all of them so far. 

I added LED driving lights, manual fan switch, heated jacket controller and outlet, and headlight relays over the winter.  I considered they may be an issue, but as far as I can tell none of these mods are wired anywhere near the problem areas of the bike.

By chance is the manual fan switch wired through the coolant temp sensor? I recall that that may be a part of the circuit that effects the fuel injection upon start up. While this doesn't explain the erratic oddness of electrical conditions you've got going on...it might have some influence on the initial starting and running.

Yes, the manual fan is wired to tie the temperature sensor input to the fan relay to ground.  Doesn't appear to be a factor in that there is no discernible problem with starting or running outside of the one time the engine did the momentary loss of power which may be related to whatever is momentarily killing the electrical system at ignition on.

Wish I could find an ABS wiring diagram for a '94 K75.

I think I have found the problem.  My ignition switch is getting finicky in that turning to the "run" detent doesn't always power up the electrical system.  I need to make sure it is all the way to the end of it's travel to get it to turn on.  Earlier this morning while pulling out of a parking lot it shut down the bike and when I turned it hard to the stop, everything came back.  

So now I have a new question:  Can the key switch be disassembled and cleaned or can I just shoot some contact cleaner into it?

Is this something I can attempt 1500 miles from home?  The bike is running well now, but I want to know what I can do if it decides to die on the way back.

You can get it apart, after you get it out of the dash pad remove the plastic cover and you will find a small grub screw with a slotted end. It is usually covered with a spot of green paint (maybe red). Remove the screw and it will come out the bottom leaving the lock barrel behind. Then you can get it apart to clean it.
They are renound for giving this exact same trouble you are getting and I have found that any of the BMW ignition switches usually are best replaced when they start to fail.

I had mine apart a few weeks back. As Rick say, look for that tiny grub screw in the side of the electrical portion of the switch. Mine blew apart and had to be cleaned and 'greased' and carefully placed back together.

A new switch, part number 61 31 1 459 594, is not available in the US warehouses currently but is in Germany and they retail for US$83.88

Ignition switch cleaning instructions:

http://technik.flyingbrick.de/images/6/6f/61_zuendschloss.pdf

Gentlemen, thank you for the quick reply.  This place is great!

My knowledge auf Deutsch doesn't extend very far from from "verstehe sie nicht, bitte sprechen sie Englisch",  and I'm not even sure I got that right.

Looking at the photos, it appears that the ring around the switch is pried out and then the switch lifts out enough to service it.  Is that correct?  Then it is on to the grub screw and the inner workings.

TWB, when you say "greased" I assume you mean with a dielectric corrosion preventive.

The procedures look fairly straight forward.  I might just give them a try this weekend before my return trip.  Again, thanks for the help.  This forum and the brain trust here has to be one of the best things about owning one of these great bikes.

Yep, always, that's why I emphasised the word.

Same here on the grease.

Dies ist in Englisch
http://www.eilenberger.net/K75S/IgnitionSwitch/

It's actually a fairly simple device, with divets for detents where the contacts move across each other. I wasn't surprised how much crud had built up, and gave it a good cleaning, but it's easy to see how the few bits can wear. I have a spare in my shed in Queensland, and will need to replace this one over here, eventually, but OZ is a long way away at the mo'.

have a look around for a fibreglass cleaning stick

these devices will clean up crud on metallic objects ...like the internals of the switch mechanism ...without scaring the metal

deoxit could well be a good aid at cleaning up the internals of the switch

phillips and several other companies used to market a "proper " contact cleaner and lubricant ...not just a wd40 equivalent which may also help in the "keep it going "  type refurbishment ....do not go with wd40 in there ...it does gumm up eventually ...except as a agent to dissolve crud ...before a proper clean with a good electronic cleaner to  remove any traces or residue of the wd40

but to me sounds like your issues are also with the wires connected to the switch terminals ...as you said  some movement of the steering ...brought back functionality (flexing cables ..or just pure mechanical shock to the switch mech ?)

perhaps the soldered joints ...or more specifically the copper cable may have fractured at the point where the solder has wicked up the cable at the joint ?
now causing intermittent conduction as well ?

it might pay to resolder the cables to the switch part ...after cutting off a 1/2 inch  or more of the copper cable away from the current soldered section ...maybe even an inch away



good luck with the exploration

Thanks for the information.  I am at my destination in Fort Collins, Colorado and will be here for the weekend, leaving Monday to return.

Had the power drop out yesterday morning pulling out from hotel.  Electrical system and engine shut down.   Came back when I quickly turned the key to the stop.  Was not a complete electrical shut down, but more like a "brown out" that caused a low voltage fault in the ABS.  Reset the ABS and the bike ran the next 350 miles flawlessly.

I am parked on the street with limited tools and place to work right now and leery of disabling it completely, so am thinking of soldiering on with the plan to pull the switch when I get back home.  It seems to work properly for now when I am positive about turning on, and the only problems occur with the first start in the morning.  For some reason, once I am out on the road things seem to work as they should.   Possibly, engine vibrations are helping the switch internals settle in to making good contact.

Well, I am back home now and ready to go into the ignition switch with the help of the great write ups you guys posted.  It has been dropping out at least once a day during the ride home from Colorado and I don't want to have it crap out completely so I think it should be attended to ASAP. 

Only a few more questions now.  I have tried to get the switch out of the RT dash pad with no success.  Does it come out of the top or the bottom of the pad?  What kind of tools should I use to release the tangs that hold it in place?  Are there any tricks to getting it out?

Remove the symbol plate upwards, then the retaining tangs is visible...
press them towards the switch housing and push the switch assy down.

Thanks, Inge.  Going in there tonight.

Got in and here is what I found.

Pretty clean inside with a very slight amount of burn, mostly on the center post.  I cleaned the center post with an eraser and Deoxit because I didn't want to disturb the silver plating on it.  The rest of the terminals were given a shot of Deoxit and a scrub with a paper towel because I didn't want to work the level of the fixed contacts below the plastic surface the moving contact rides on.  (From what I can see, this is the weak point of a crappy switch design.)

The rotating contact had some very shallow grooves worn in it where it rides over the fixed contacts.  I leveled the contact surfaces with 800 grit and coated everything with a very thin coat of silicone dielectric grease.

Reassembled and everything works, but being intermittent, only time will tell if the problem returns.  Seeing how little, if any, projection above the plastic surface the fixed contacts have I guess I should start saving up for a new switch.