My 1987 K100RT has been a reliable and trouble free bike for years and many, many miles and I try to do my part in keeping the maintenance current to include keeping most of the switches clean, corrosion free, and slathered in dielectric grease.

Recently the bike has taken to going dead as it idles or, more commonly, as I come to a stop. The engine doesn’t falter, it is as if I hit the kill switch. Up to this point this seems to be a typical short or poor ground (and either of these could in fact be the issue). But I want to mention a couple of things I noticed to aid one of you gurus towards pointing me in the right direction:

The problem occurs once in ever 20 or 30 stops
When the bike dies, the gear indicator goes blank - and then after 10 seconds or so, the indicator comes live and the bike will restart.
And today’s secret clue is this: the low fuel light comes on. The low fuel light has not worked in years and the tank was about 3/4 full (and verified by a fuel gauge and a recent receipt from Chevron).
During the last two occurrences, I refrained from panicking and flipping switches and pushing buttons, rather just I waited ten seconds or so until the gear indicator came live, at which time the bike restarted and I continued on my way. It was during these panick-less stops that I noticed the low fuel light burning. 

The purpose of this post is hopefully get advise from you on where to start, as opposed to spending my time in a direction which doesn’t make sense.

Dave

Check the startermotor brushes, should be longer than 7mm.
Place some new ones and I guess the problem should be over.
Leaves you as gremlin as you post this in "buy and sell"

I'm reaching, but since the problem has suddenly started and the fuel light has also suddenly started, if unplugging the white wire to the fuel light fixes the problem, that might be a place to look.

It's three quarters full because you looked in the tank,  or at the fuel gauge or both? Maybe alternator brushes wanna look see too.

Wiggle some wires, like say at the instrument cluster or tank, to see if it affects the gear indicator, fuel level light or something.

How's the previous momentary miss with turn signal problem going?

Actually, ignore my previous post until you try this; clean the socket for fuse one (top fuse) and replace it with a new fuse, even if it looks good, and see if that makes a difference. It's probably the quickest, easiest check. Probably the cheapest, weakest link in the chain too.

Then do the other fuses. It's a good thing to do, even if you don't have problems.

Okay. I have a couple of “to dos” and I will report back.

Fuel level was verified...about 3/4 full.

On the turn signal intermittent problem. I have ordered another flasher relay to see if that may be the issue.

I was thinking fuse one might be the cause of the turn signal problem too.

Arlina wrote:Check the startermotor brushes, should be longer than 7mm.
Place some new ones and I guess the problem should be over.
Leaves you as gremlin as you post this in "buy and sell"

On the chance that I just might learn something, why would the condition of the starter motor brushes have something to do with the bike dying?

On this forum iv read of a few Strange hiccups being traced to the starter,how many Km/years is it since it was serviced.Worth a look and you might learn something you can share with us .

Because a large amount of the electrical system earths through the starter motor brushes and the starter motor coils. Is why the load shed relay drops out when you hit the starter button and turns the lights and other bits off. Normally pin 85 on the load shed earths through the aforesaid starter bits but when you hit the button, the starter circuit back-feeds 12v to pin 85 and the relay drops out.

Not a terribly good idea on BMW's part but then who knows how much schnapps Hans drank before he designed that system.

I have not died, it just has taken me a few weeks to find the time to address the starter motor brushes. The first thing that struck me was the difference in length. The brush going to ground was 10.2 mm while the brush from the relay was 7.9 mm.

I installed a new Denso assembly and will see if my stalling problem has disappeared. Also, with the OLD brush assembly, the brushes fed from the outer edge while the replacement assembly are ‘loaded’ from the edge nearest the commuter. The functional difference is with the old assembly the wire connected to the brush would restrict forward travel such that the brush would begin to lose positive contact with the commuter as the brush neared 7mm-8mm. The new brush assembly loads the brush from the end nearest the commuter so brush movement is not restricted.

I will post again in a couple of weeks with an update on the stalling.

Had a very similar problem with my 89RT about 6 years ago. I'd pull up to a stop and she'd just stall. It would then just start back up and all was good but it was intermittent.
The problem was very tight exhaust valves. Re-shimmed her and never had that problem again.
You can read the thread here...
https://www.k100-forum.com/t8442-engine-cutout-on-throttle-off

I checked the valves - both intake and exhaust - and they were spot on. Given my problem appears to be electrical (I.e. gear indicator goes blank and speedometer drops to zero), I suspect that valves are not the culprit.

Update on electrical gremlin: the problem persists. If anything, it seems to be getting worse.

Following two recommendations, I replaced the starter motor brush assembly and verified the exhaust valve clearances.

One additional observation today is after the bike died the speedometer drunkenly floated to around 1500 rpm before coming back to zero (this was within a second or two after the engine had died and was at a standstill). Also, during one attempt to restart the bike immediately after dying, the idiot lights were lit, but dim. The starter turned very slowly. After waiting maybe two seconds, the bike started normally.

When I feel the bike dying as I slow to a stop it feels differently from a normal decrease in speed - it as if the kill switch was on. Before coming to a standstill, I can quickly bump start the bike and it immediately comes back to life.

Check and Deoxit everything.

You've looked at the likely culprits and not found the problem. So you're at the stage where you just have to eliminate stuff. It'll not be wasted effort, checking and cleaning all the connectors is a very useful thing to do on a bike of this age anyway.

Bad earths tend to be responsible for the really weird faults, so tank off and do a first class job on the frame earths.

I'd clean the fuse connections and put new fuses in. That might do it.

I'd unplug the fuel level wire.

I'd unplug the wire from the flasher to the instrument cluster (coil-tacho?)

Sposing the flasher wire is drawing heat (rusty connection, teeny weeny short, plug not fully pushed home,  or loose, sumnat like that) that might explain why it only stalls at idle, why the speedo chucked a wobbly, why you had a flasher related problem, and that.

A lazy way would be a piece of paper on a pin before plugging a connector back.
I guess removing those wires could eliminate the problem, if so then just plug one back in to find which one.

Also, during one attempt to restart the bike immediately after dying, the idiot lights were lit, but dim. The starter turned very slowly.


Earth issue has been highlighted:
1: Bad earth terminal connection from battery?
2: Bad connection inside ignition wiring or the connection.
3: Bad battery [get it load tested]. 


Does the warning triangle light come its 1 2 or 3 AND does the clock also  go blank, if yes its 1 or 3.


Starter you have eliminated.


A bad relay can eat juice and a bad flasher relay can do just that. Been there.

chris846 wrote:Check and Deoxit everything.

Bad earths tend to be responsible for the really weird faults, so tank off and do a first class job on the frame earths.

+1

+1 on checking grounds.

The clock is fed from fuse 3. Fuses 3, 4 and 5 have common feeds. The same circuit also feeds the ignition switch, the load shed relay and the starter relay. The source of power to these components is directly from the battery. I'd be checking the wiring in this area. Check for loose connections on the battery positive and starter relay pin 30. Same goes for loose or intermittent connections on fuses 3, 4 and 5.

chris846 wrote:Check and Deoxit everything.

You've looked at the likely culprits and not found the problem. So you're at the stage where you just have to eliminate stuff. It'll not be wasted effort, checking and cleaning all the connectors is a very useful thing to do on a bike of this age anyway.

Bad earths tend to be responsible for the really weird faults, so tank off and do a first class job on the frame earths.

BIG +1

Electricals doing weird things is a sure sign that there's a ground/earthing issue. In a K-bike the most common issue here is the starter. So it looks like you've sorted that but don't forget to clean the mating surfaces where the starter beds down to the GB casing.

Pretty much nearly everything on a K-bike earths via the main ground point on the frame spar under the tank. You can easily check whether this is working properly by measuring the impedance between a wire on that point and the negative terminal on the battery. You should be reading <5Ω. If >5Ω then test between there and the GB casing and then also between the GB casing and the battery neg terminal. If there's something impeding the current on the earth side then this will tell you where it is.
• The earth connections at the frame can be corroded.
• Earth continuity through the frame to engine mounting points may not be good.
• The earth cable from GB to battery might be physically failing. The strands can stress and start to break at the ring connector.

You can upgrade your earthing system by adding an additional cable from the main ground point on the frame spar to the earth lug on the GB. This bypasses the earth connection through the frame to engine mounts. It's called an Engine-to-Chassis Ground Strap.
Upgrading the battery earth cable to a heavier gauge wire is a good idea also.

I can't see where you mention your battery (may have missed that) but a bad battery will have the same effect. The fact that you can start the bike is no indication that the battery is ok. Nor does seeing 12 point something volts across your battery terminals indicate a good battery. It really needs to be tested under load. Go to a battery shop and get that done even if just to eliminate it from the equation.

" lt's as if the kill switch was on"

Maybe give the kill switch a look and clean.

+1 on the Kill switch.

My bike started to cut out periodically, although it restarted every time.  My Kill switch was in poor conditon,

so I fitted  a good used unit  and have now done  20k  problem free.

Bob

Should all of the suggestions posted, not identify the problem, there is one other possible cause, Ignition Control Unit.After hours of trying everything,Crazy Frog suggested the ICU and joy of joy problem solved with a S/H unit.Hope you get lucky

Yikes! I was hoping to travel and see the world and now I find I will be spending my golden years in the garage. Grounds, grounds, and more grounds...one would thing we were talking coffee!

Going down the proverbial troubleshooting rabbit trail, I read the TPS might be failing and/or dirty and the reader had disconnected his TPS and his (similar) problem disappeared. I pulled off the TPS connector and took a 20 minute ride duplicating the various conditions which I had (earlier today) led to repeated stalling issues. Smooth sailing! This does not mean the problem has been properly identified, only a solution was found.

Be grateful that you don't own a motronic equipped bike as they will not run without the TPS in place.

On your Jetronic model the TPS is but a two position switch to detect 'off - no throttle) and WOT Wide Open Throttle.

Mike

I mentioned in an earlier post that disconnecting the TPS seemed to resolve the stalling problem - but alas, it has not (although taking the TPS out of the equation has rewarded me with a super smooth engine response around the 2000 rpm point).

The bike stalled this morning and I sat five minutes before the idiot lights came back to normal (rather than dim) and the bike restarted. Yesterday it stalled and I sat two or three minutes waiting for the gremlin to move along. So it seems restarting after stalling has progressed from a second or two to five minutes. So....

I spent this morning cleaning frame and engine grounds (the grounds looked good). The primary frame ground had slight discoloration on the bare metal of the frame. The dielectric grease from past cleanings was still present.

I will have the battery load tested tomorrow.

Holister wrote:Electricals doing weird things is a sure sign that there's a ground/earthing issue. In a K-bike the most common issue here is the starter. So it looks like you've sorted that but don't forget to clean the mating surfaces where the starter beds down to the GB casing.

Pretty much nearly everything on a K-bike earths via the main ground point on the frame spar under the tank. You can easily check whether this is working properly by measuring the impedance between a wire on that point and the negative terminal on the battery. You should be reading <5Ω. If >5Ω then test between there and the GB casing and then also between the GB casing and the battery neg terminal. If there's something impeding the current on the earth side then this will tell you where it is.
• The earth connections at the frame can be corroded.
• Earth continuity through the frame to engine mounting points may not be good.
• The earth cable from GB to battery might be physically failing. The strands can stress and start to break at the ring connector.

You can upgrade your earthing system by adding an additional cable from the main ground point on the frame spar to the earth lug on the GB. This bypasses the earth connection through the frame to engine mounts. It's called an Engine-to-Chassis Ground Strap.
Upgrading the battery earth cable to a heavier gauge wire is a good idea also.

I can't see where you mention your battery (may have missed that) but a bad battery will have the same effect. The fact that you can start the bike is no indication that the battery is ok. Nor does seeing 12 point something volts across your battery terminals indicate a good battery. It really needs to be tested under load. Go to a battery shop and get that done even if just to eliminate it from the equation.

Humor me, what is GB casing?

Gear box casing, the lug above the gear lever near the coils. The earth terminal from the battery is normally screwed into that lug.

Deox-It’d the connectors on all the relays.
Removed the turn signal/flasher relay (possible problem relay)
Tested each fuse
Deox-it’d the fuses and fuse holder

Checked wires, connections, connectors, and grounds under the tank
Disconnected the speedometer and gear indicator connectors (to alleviate possible problem points)
Cleaned and sanded frame where frame connects to GB. Cleaned bolt, bolt threads, and washer for same.

Slathered (technical term) dielectric grease on all connections, my shirt, and a good pair of pants.
 
The plan is to ride it for a week. If after a week, stalling is only a distant memory, I’ll reconnect speedometer and gear indicator, followed by the turn signal/flasher relay another week down the road.

Holister wrote:You should be reading <5Ω. If >5Ω then test between there and the GB casing and then also between the GB casing and the battery neg terminal.

Measured 0.4 ohms.

davemadsen wrote:The plan is to ride it for a week. If after a week, stalling is only a distant memory, I’ll reconnect speedometer and gear indicator, followed by the turn signal/flasher relay another week down the road.

It's likely to stall. 

Resolved!
With some certainty I believe I had a grounding issue. 
While I felt I had done a good job of ensuring my ground points were clean, alas it seemed they were not good enough. My ground point under the tank was good but I think the frame to engine points were the weak point. I removed the 10mm (?) bolts securing the engine to the frame and with a tap and die cleaned the threads. I also ensured the washer and the head of the bolt was just shiny metal. With everything bathed in dielectric grease and torqued to spec, my problems were resolved. 

The various components I took off line as to eliminate possible causes have all been reinstalled or connected and the bike is back to where it should be. 

I want to thank everyone for their input.

A couple of my bikes have a heavy wire(4 AWG if I remember) running from the frame ground under the tank to the ground point on the transmission. I don't like relying on the frame and the engine to provide the ground for all the lights and instruments on the bike. Wasn't too hard to install.

Five months later and the bike again died as before. After waiting about 15 seconds, the instrument lights came back on, the bike promptly started, and I was on my way. This event brought me back to re-reading the comments to this post/problem. My ground points appear corrosion free, tight, and still smeared with dielectric grease. Because the problem has always resolved itself - taking as few as a second or two to maybe a minute (all without any intervention on my part), I was wondering if the problem could be relay-related. Hopefully you have more insight into how the various relays work and can comment on this being a possible cause.

FYI I have ridden 3,000 miles since this problem last reared it’s head.

Does your time clock reset if this happens? 

If yes then check the battery connections, main earth and earths under the gas tank.

If no, check for ignition fault issues. ignitions do go faulty on these. Been there.

If you haven't already, I think you need to inspect and clean both the ignition switch and the kill switch.

Laitch wrote:If you haven't already, I think you need to inspect and clean both the ignition switch and the kill switch.

Agree. Check the ignition switch.

+10 on the ignition switch. I chased an ignition switch problem for many thousands of Ks with various "experts" telling me repeatedly that the earth under the tank was the cause only to eventually have the bike retrieved from a long way from home when it died big time and the switch was replaced. That was on a R80RT but the switch is the same.

The ignition switch on my K75RT first went wonky for the first time 6 years ago at 62,000 miles.  Started acting up again a couple weeks ago at 98,000 miles.  This time I didn't hesitate to open it up and clean the contacts which were slightly burned.  Put it back together and it has been working flawlessly since.

Try this if the problem is easy to reproduce.

When it cuts out, turn the choke switch on. Look at the instrument cluster back lights. Are the bottom ones off, and do they come on when the neutral light comes on? If so, maybe you have an earth fail.

Did the choke light remain off when you turned the choke switch on, and did it come on when the neutral light came on? If so, maybe your chasing a power fail.

I think sometimes you had dim lights and sometimes they were off. Sometimes the fuel light was dim, and sometimes the tacho was drunkenly at ~1,500rpm with the bike stationary.

You might want to have a good look at pin 13 of the instrument cluster (one of it's two earths) I removed it which gave me a dim fuel light, dim neutral light, and a drunk tacho. That was with the headlight switch off, which I don't think you have in the USA, but was similar with the light on. Further to that, the bottom back lights were off, which is to be expected. When I removed pin 6, power (green/black) I had no lights to the cluster except the back lights and clock, which is to be expected.

I don't remember if you cleaned the instrument cluster pins, I think it would be a good idea. Don't lever the plugs out, pull them out parallel to the surface.

daveyson wrote:Try this if the problem is easy to reproduce.

When it cuts out, turn the choke switch on. Look at the instrument cluster back lights. Are the bottom ones off, and do they come on when the neutral light comes on? If so, maybe you have an earth fail.

Did the choke light remain off when you turned the choke switch on, and did it come on when the neutral light came on? If so, maybe your chasing a power fail.

I think sometimes you had dim lights and sometimes they were off. Sometimes the fuel light was dim, and sometimes the tacho was drunkenly at ~1,500rpm with the bike stationary.

You might want to have a good look at pin 18 of the instrument cluster (one of it's two earths) I removed it which gave me a dim fuel light, dim neutral light, and a drunk tacho. That was with the headlight switch off, which I don't think you have in the USA, but was similar with the light on. Further to that, the bottom back lights were off, which is to be expected. When I removed pin 6, power (green/black) I had no lights to the cluster except the back lights and clock, which is to be expected.

I don't remember if you cleaned the instrument cluster pins, I think it would be a good idea. Don't lever the plugs out, pull them out parallel to the surface. If some rough knuckle before you levered them out rough, pin 18 might be damaged cause it's the last one on the multi plug.

Your description is EXACTLY what I have been experiencing. The challenge will be to remember to check various lights on/off when this happens next time and I am in stuck in traffic with four people honking their horns at me.

I’ll clean the terminals on the back of the instrument cluster.
Thanks!

I have cleaned many switches on the old girl, but not the ignition switch. I have added this to my to-do list for this weekend.

Again, thank you for your suggestions.

Remember safety first have the switch in a plastic bag to save the Flying Bits making a bid for freedom 

Rather than attempt to pull the wires and the ignition switch connector through all the nooks and crannies around the steering head I perform the surgery on top of the tank.  A box and towel make my operating theatre.



One caution, when you are ready to remove the screw that hold the switch to the lock, make certain that you have a screwdriver that fits snugly in the slot of the screw. BMW used their goddam red Locktite on it, and because of the plastic you can't use heat to soften the stuff. Anything less than a good fit will probably break the slot on the screw. You needn't inquire as to how I know this.

...looking for an appropriate operating theater now...

The symptoms were a little different with the light switch at park or on. 

The tacho symptom was a little different too, drunk and AWOL. For a few seconds at 0, then 2000rpm for a few seconds, then 0 for a few seconds, then 5000rpm, and remained there. 

If worried about stretching the memory tendons, hop on the bike and have a few practice runs. 
1 - Choke light on with choke on, check. 
2 - Back lights on, bottom one's off.
3 - Back lights on with N light on.  
4 - Choke light on with N light on. 

Anyways, you might not have to do this test.

When your brick cuts out the oil and battery lights should come on (probably not very noticeable cause that's normal) and the neutral should remain on, but it doesn't. Do you remember which? 

All three depend on the cluster for power, but only the N light depends on the cluster for earth, so only neutral coming on could point to no earth, all three coming on together could point to no power.

When the bike conks out, have you noticed if the headlight also conks out?

Actually, could it be worn alternator brushes?

Hey,  sorry when I said pin 18 I should have said pin 13. I tested both those, just got mixed up. Will edit the crap out of my earlier post now.

daveyson wrote:The symptoms were a little different with the light switch at park or on. 

The tacho symptom was a little different too, drunk and AWOL. For a few seconds at 0, then 2000rpm for a few seconds, then 0 for a few seconds, then 5000rpm, and remained there. 

If worried about stretching the memory tendons, hop on the bike and have a few practice runs. 
1 - Choke light on with choke on, check. 
2 - Back lights on, bottom one's off.
3 - Back lights on with N light on.  
4 - Choke light on with N light on. 

Anyways, you might not have to do this test.

When your brick cuts out the oil and battery lights should come on (probably not very noticeable cause that's normal) and the neutral should remain on, but it doesn't. Do you remember which? 

All three depend on the cluster for power, but only the N light depends on the cluster for earth, so only neutral coming on could point to no earth, all three coming on together could point to no power.

When the bike conks out, have you noticed if the headlight also conks out?

Tacho signal comes from one of the coils......if there is mention of a coil issue here could an erratic tacho be sign of an erratic coil issue?

I don't think so cause my test bike doesn't have electrical issues. For a long time I thought Dave had a power problem, I was surprised by the results from removing pin 13 from the circuit. By the way, removing pin 23 (another earth) had the same results, so that maybe wants a look see too.

The tacho remained at 0 for about six seconds. Now it then goes from there to 6,000rpm then settles down to about 3,000. Since Dave's problem often doesn't last that long, I guess the tacho symptom only happens sometimes.

With my brick I got the following lights on the cluster:

Pin 13 removed, Key on, light switch off; low fuel, neutral, high beam, oil, battery, drunk tacho. 

Pin 13 removed, key on, light switch on; high beam, oil, battery, top back lights (so no low fuel light even if the fuel is low) 

The choke light worked in both cases, but not with pin 6 (power) removed, which is to be expected.