Starter won't turn -- intermittently

[billday]

Here's my story.

First, the engine abruptly cut out at 75mph on the NY Thruway. Luckily I'm in the right-hand lane, so I coast to the shoulder. Pushing the starter button elicits a gurgling noise I've never heard before. Pushing it again: silence. A passing motorist stops and offers help, I ask him to drive me home (15 minutes away) so I can figure out what to do next. We get the bike as far away from the road as possible (not far, since we're on a massive 8-lane overpass). Then I give it one last try and -- surprise -- it fires right up. I thank the guy and ride home.

The overpass where the engine cut out has several very bumpy expansion joints. After I pulled over, the bike was on and off the center stand a couple of times. So I'm thinking that a short or bad connection cut in and out with the bumping. I dig out my hard copy of Bert Vogel's "Bike Won't Start" opus. I take apart and clean the starter switch. I clean every connection I can find. The upshot: Sometimes the starter spins when I push the button, and sometimes it doesn't.

The harness from the right-hand controls has had some frayed and pinched-looking spots since forever. Hoping that an intermittent short there is my problem, I bought a replacement from EME (only $150!) and installed it this morning.

As I reassembled everything (putting the instruments and the headlight back on), I tried spinning the starter at each step of the way, and it worked. And then suddenly it didn't any more. I swapped the starter relay with a spare I happened to have, no difference. I disconnected and cleaned the ignition module connector, that made no difference. When I hit the starter button, the starter relay clicks but the starter motor does not turn.

A few more pertinent facts:

1) The starter relay clicks every time I hit the button.
2) The starter _should_ be good -- I replaced the brushes and cleaned it just a few winters ago. The screw terminal and the cable there are clean.
3) My battery is good, just a couple of years old and it holds a charge no problem.
4) The switch in the clutch lever is working fine: When the starter wants to turn, it will turn with the transmission in neutral and it will turn with the transmission in gear and the clutch lever pulled in.
5) The fuel pump is spinning. The sock and the fuel filter are new this season.

Sorry this post is so long but I'm trying to avoid the Twenty Questions Syndrome. Any ideas, suggestions or related experiences will be deeply appreciated. Thanks!

[Martin]

Check and clean battery connections both ends. When it cut out on the highway what was the situation with the instrument lights were they functioning? Check voltage drop across the battery when hitting the stater button and post results.
Regards Martin.

[rbm]

If possible, attach a voltmeter to the positive terminal of the starter, and to ground.  Continue to conduct your diagnosis.  When the starter fails to turn, observe the voltmeter to see if potential is being applied to the starter or not.
- If yes, then internal problem to the starter motor
- If no, then intermittent connection at the starter relay or battery

[billday]

Hi Martin, Instrument lights have always been completely normal thru this whole thing.

Hi Rob, funny you mention that. After lunch today I was thinking that Bert's algorithm was pointing me to the starter. So I pulled it and ran through the Haynes diagnostic routine. It appeared that the tiny O-ring that insulates the terminal post might have slipped out of position, so I fixed that up, put everything back together, and -- nothing but crickets (in the form of a clicking starter relay) when I push the button.

Frustrating! tomorrow I'll do the voltage checks you suggested.

Thanks, Bill

[rbm]

Bill,  That was exactly the problem that was running through my mind ... an internal short on that terminal due to failure of that insulator.

[The Mighty Gryphon]

Dirty ignition switch causes whack.

http://www.eilenberger.net/K75S/IgnitionSwitch/

[billday]

Yep, been thinking about that.

My ignition is embedded in my airbox -- a little harder to get at than most -- so I'm been hoping something easier could be the problem.

[rbm]

But the starter relay is always clicking, Bill says.  So, that kind of eliminates ignition switch problems, no?

[Laitch]

 :oldguy: I foresee this thread concluding in a "for fecksake, it was only this" kind of way. The death by road shock has me in the not-quite-tight-enough-connection camp.

[Rcgreaves]

I'm a newbie here so my interpolations of the diagnosing questions may not be solid:. Anyway in my case (87 K75) starter brushes wore out causing intermittent starter and the ground created by said brush.  I could shift the position of the starter motor slightly by dropping the clutch whilst rolling the bike- and that was enough to get the brushes in contact and the starter spinning. 

Sent from my A573VC using Tapatalk

[Rcgreaves]

PS u managed to bugger the insolator in my reassembly...not Good. I note you rebuilt it a few winter's ago....brushes OUGHT to be ok.

Sent from my A573VC using Tapatalk

[billday]

After lunch today I was thinking that Bert's algorithm was pointing me to the starter. So I pulled it and ran through the Haynes diagnostic routine. It appeared that the tiny O-ring that insulates the terminal post might have slipped out of position, so I fixed that up, put everything back together, and -- nothing but crickets (in the form of a clicking starter relay) when I push the button.

Like I said, I was into the starter yesterday. Brushes have 8mm left.

Tempted to plunk for this anyway. An expensive way of feeling like I am doing something:

http://www.euromotoelectrics.com/Starter-repair-kit-for-Denso-Starter-BMW-K-p/bmw-strk007tk.htm

[rbm]

Bill, can you remove the starter motor and bench test it off the bike?  Clamp the motor by its mounting lugs into a vice.  Use a battery or power supply to apply power to the starter positive terminal and vice jaws (convenient to clamp the starter and negative wire by the vice jaws.  This will validate the condition of the starter.

[billday]

Rob, great suggestion. Now I'm going to ask a question that will reveal the vast depth and breadth of my ignorance of electricity. I hope you're suitably impressed.

Here's the question. There is only one terminal on the starter motor. Is that terminal (a) positive (b) negative (it's grounded to the frame, right?) or (c) doesn't matter (because it's a DC motor)?

When the motor's clamped to the bench, do I just connect a lead from a terminal of the battery to the post on the starter? Does the other terminal of the battery need to be connected to a ground or anything?

Thanks in advance. Now I'm off to the garage to pull my starter again.

[rbm]

The one terminal on the starter is positive and goes to one of the two brushes.  The other brush is connected electrically to the case of the motor.  Attaching the negative wire from the battery negative terminal to any part of the motor frame will complete the circuit.  That means clamping the bare wire end along with the motor mounting lugs will serve both the purpose of supporting the motor against torque when it starts and completing the circuit.

[billday]

Got it, that makes sense.

So get this: I went down to the garage to pull the starter. As always, I tried turning on the ignition and pressing the starter button. The starter turned over. I did it 10 more times, each time it worked. So I threw on the tank, fired it up, and took a brisk ride around the block (five miles or so). Went home, shut it off, let it sit a few, started it back up, another run around the block and now it's parked in the garage again.

The good news is, I rode my K today.

The bad news is, I have no idea what the problem was and whether I fixed it.

[rezo11184]

Hi Bill and everyone else! My name is Greg, I have a K1100 and my profession is CMM technologist - NOT Motorcycle technologist and I was an Electrical engineering student.
the good: I believe you have solved your problem by rattling the starter motor while trying to remove it
the bad: I believe I have the same problem or worse -I think my contacts in the starter have welded to the brush or something along those lines because the voltage of my battery drops to almost 0 on start and same symptoms as you Bill: the relays click and nothing happen.
My symptoms:
The relays click because the voltage of the battery cannot sustain the starter motor current and starter relay electromagnet current therefore it resets, the load drops because the starter relay no longer completes the starter circuit so more ions in the battery become available, the voltage rises and completes the circuit again.
This meant that my battery is weak.
I jump started the bike, measured alternator output voltage: 13.5+-0.5V.
took the battery out to look at the acid level, it was good, took the bike for a ride(jump start), tried starting after an hour[/size] with exactly the same symptoms.
[/size]so yes my battery may be bad but I'm sure its not that bad as it's a little over a year old.I will post update of what I find. My mission is to remove the starter and turn it by hand then determine if its seized and why or whatever is downstream to it. My fear is finding out that the starter works fine but the sprag clutch has seized or something like that.

[Laitch]

so yes my battery may be bad but I'm sure its not that bad as it's a little over a year old.My fear is . . .

The way to be sure of that, Greg, is to measure its resting voltage, not count the candles on its birthday cake.  :hehehe It could have been made on a Monday. It's a little too early for fear unless you just enjoy that kind of thing.

Welcome to the fun house.  Consider starting a thread of your very own with photos of your bike and explanations of your trials, tribulations and remedies so we won't be distracted by billday's success.

[billday]

Hi Greg,

My bike's been running great since last report. I feel pretty certain that the problem was the tiny O-ring that isolates the screw terminal post from the body of the starter. On my list for the winter is to install this refurb kit:

http://www.euromotoelectrics.com/Starter-repair-kit-for-Denso-Starter-BMW-K-p/bmw-strk007tk.htm

Good luck!

[rezo11184]

Hey thanks for that link, once I get the starter out I will know whether I need that or not. The Battery resting V is 12.6 stable. I will see how that insulating O ring is doing... maybe same thing as you. Now That I think of it I will try to bypass everything by connecting the Positive of the battery directly to the starter to remove all that's between this circuit from the equation.

[rezo11184]

Took the starter out, spun by hand ok, connected it to a batter and it spins, resistance was 0.1 Ohm. took it apart - no sign of anything unusual, cleaned and lubed it, resistance between the positive terminal and the motor case is infinite, reassembled, connected directly to battery and it spins fine, so it wasn't the starter motor... :( I can only hope that the battery is totally dead. I'll keep you posted once the starter is reinstalled in the bike...

[Laitch]

:( I can only hope that the battery is totally dead.

Thanks for the update. It isn't often a death wish is infused with hope. :giggle