I have been waiting to get back to the original problem too but now I have this alternator that I bought off eBay. I need to make a decision to keep it or return it and try something else. Probably a rebuilt one.

Could you give me your opinion? after driving to work the battery measured 13.2 (engine off). So it is charging. But is it going to wear out my battery or is it worth keeping?

Thanks.

The battery is reading 13.2 so the alternator is charging so my gut instinct is to keep it. It is possible that a new regulator is needed but I would be chasing the intermittent stopping problem first and keep watch on the state of charge.

I have just gone through the whole thread.  I think you have a good alternator and battery.  You mention on an earlier post that the voltage rises as the engine warms up.  That is a normal reaction of the alternator to the charging of the battery after a start.

When the engine is started, a very heavy load is placed on the battery for a short period of time.  A lot of electrons are removed from the surface of the plates.  As soon as the engine starts the alternator starts putting out current.  Because the surface of the battery plates has been depleted of electrons and the surface is the most attractive to the electrons from the alternator, there will be a lot of current flow into the battery.  This high current through the resistance of the alternator's coils and the connecting wiring causes a voltage drop per Ohm's law.  As the battery replaces the electrons that were used by the starter it will draw less current from the alternator.  As the current to the battery drops, so does the internal voltage drop in the alternator, and the output voltage will begin to rise.

Normally, the amps removed from the battery to start are fairly small since even though the current flow to the starter is quite high, it is only being drawn for a few seconds.  My rule of thumb is that with a healthy battery and an easily started engine, the battery will be fully recharged in the time it takes to reach normal operating temperature.  When you said the voltage rose as the engine warmed up it sounded to me like everything electrical was functioning properly.  The starter was taking a normal load on the battery, the battery was replacing the removed charge in the normal time and the alternator was responding to the battery charge being replaced in the expected way. 

Having said that, I wonder about the condition of your ignition coils.  I just recently read some posts on another forum about early model K100 ignition coils and the problems related to them.  It appears that there are two coil designs.  Early coils were prone to problems and many were replaced by the new design that corrected the problems.  I wish I had read the postings closer and could find the thread now, but my brain isn't quite up to the task as it was many years ago.  Suffice to say, that all black coils are the old style, and the appearance of orange somewhere on them indicates the new style.  I think that it wasn't until 1986 that the new coils were phased in on production bikes so unless your coils have already been replaced they may be the old style and prone to problems.

You may want to take a little time to do some research on early k100 ignition coil problems.  There may be something there that helps.

The early coils are a dark grey and many go almost black with age and the later type, which are totally interchangeable, are a medium brown with orange towers where the leads attach. I would say it is unlikely to be the coils because the tacho follows the cut out where if it was a coil it is unlikely the tacho would cut out but with an intermittent electrical fault never ever say never.

I only mentioned the coils because I had just read of the problems they allegedly caused on the early K100's.  I have no experience with them, and suggested some research to see if there might be something they could affect.

Coils don't last forever on other engines, and if there is a marginal design on a 30 year old part it might make sense to look into it if it hasn't already been upgraded..

I am gathering parts so as to build an original 83 and in near 5 years have only found one grey coil that is good. It sure is a good idea to replace them as they will give you grief.

Thanks guys. I appreciate the advice about the alternator and agree with it. A way back in this thread, before I fried the diodes on my old alternator, I had purchases a regulator and brush set for it. I have been holding off on returning that in case I might need it for the replacement alternator. I'll probably keep that too now.

I'm almost certain my coils are black. I'll check tomorrow and look into it.

Rick, I don't think I ever said that the tachometer drops off when the bike cuts out, did I? I may have, but right now I can't say if it does or not. I don't think it does but my brain is a bit fried from all the directions I have been pushing it. I know that I did say that I get very erratic readings on the tachometer but they are not associated in time with the cutting out. I think the occassional erratic tach is a problem with my Acewell speedometer. I mentioned it in case it was a clue but it seemed unlikely to be related because I had the cutout problem also when I had the stock instruments  and on the stock tach it always read true.


Thanks again to everyone for the help and knowledge. One reason I like to fix things myself, without turning the bike over to a professional is because I learn so much.

You said about erratic readings with the Acewell but not dropping off. That to me still points to an erratic pulsing from the Hall sensor because the tacho needle is well damped and would probably not respond quickly to some missed pulses.

An alternator as apposed to a generator, requires power in to get power out, so perhaps its got a high resistance issue on the in side, just enough to give you low readings whether idling fast or exceeding 2000 rpm. Charlie Rick where is a safe place to look for an issue like that?

yes you did mention erratic tacho ...I had thought it was a polarity thing ...the way the pulse arrives at the tacho (high going low or low going high kinda thing )

might pay to read up the posts on hall effects for this current issue ....as rick suggests
there have been many of them
a few are very similar to each other   but there does not seem to be a reliable fault diagnosis for a few cases  and occasional failure repetitions after testing seem to be more common than one would think .

the main observation that seems to be consistent is that they fail as they are heated  

that's possibly worth investigating keith

the safest way to pick that up might be to monitor the blue wire going to the field of the regulator ..(voltage ) I would do it just before it connects to the alternator (with another wire at the spade connector soldered in place just for the monitoring process which could easily be removed without damage later  )
heres just a suggestion ,,
if the film connection for the bulb holder for the charge indicator was intermittent that could prove to be a possible candidate ...or the pin for the instruments connector to the film circuit , likewise if the cable in the loom had been stretched (around near the steering head ) and likely cause becoming high resistance ???
even the bulb becoming high resistance might affect this ?

but on an already running engine (read generating power ) might be hard to pick up the fault - see the affect

cheers

Lots of new things to consider. First I am still verifying that my battery is ok. After sitting overnight unplugged it read 12.99 in the morning. So I think it is ok but i am giving it another day because I couldn't drive today anyway. 

Once i know the battery able to hold a charge I will see how it does when plugged in overnight. From the readings i was getting on my old battery I am not so sure there is not a parasitic drain. Anyway, that will be easy to tell if there is.

After that I need to just drive it and see if I can get the issue to occur. I am guessing that it will but you never know. Maybe a fully charged battery will keep it from happening. Or maybe not but i want to find out before i start making other changes.

I checked my coils and they are the black ones. So that is something to look into in the future.

The Hall Sensors is something I have never looked at because the bike does not need to cool down after the problem occurs. But it is something i should look into and learn about so I will.

charlie wrote wrote:heres just a suggestion ,,
if the film connection for the bulb holder for the charge indicator was intermittent that could prove to be a possible candidate ...or the pin for the instruments connector to the film circuit , likewise if the cable in the loom had been stretched (around near the steering head ) and likely cause becoming high resistance ???
even the bulb becoming high resistance might affect this ?

but on an already running engine (read generating power ) might be hard to pick up the fault - see the affect

When I installed the Acewell I cut off that multi-prong connector with the film and soldered and heat shrank wrapped each wire individually. I had the cutout issue before with the stock speedo and after with the Acewell so I have not suspected that wiring in regards to the cutting out. Maybe there is some relation to the alternator charge. Also, the Acewell does not have a charge light. It had a hazard light, which seemed worthless to me, so had the choice to wire in my temp light or my charge light to it. I chose the temp light. I can not remember what I did with the charge light wire but I must have terminated it.

yeah sorry about that ...

I had thought  about the acewell and the internal wiring of the standard instruments after I posted that ....another  (alcohol induced  ) wtf  moment I guess   (been a big day and my alternator in the car packed it in ..but made it home ..just...  on a completely discharge arrangement ...through the city )

you need a resistor in the field wire going to the alternator ...from a switched 12 feed ....this will excite the alternator at about 1200 -1500 revs and give you charge out of the alternator ...other wise the charging will happen but not till  a higher revs range ...

there were a few posts in here about that ...not too long ago and I dont remember the value at the moment
but I does seem that your battery is getting charge and holding it
...which is a bonus .

what concerns me is if the system could be in discharge fully when at above idle revs ? is this going to be an ongoing issue ?

charlie99 wrote:and I dont remember the value at the moment

68 ohm

thanks inge

Wow!!! I had a feeling that comment about terminating the charge light wire might bring up something. 

So, I need to either connect a 68 ohm resistor or a 3-5 watt globe. I'll do that.


I don't understand this comment

what concerns me is if the system could be in discharge fully when at above idle revs ? is this going to be an ongoing issue ?

Also, from this post link  Rick G says

An 85 K without the warning light will not charge. It needs the current that lights the charge light to make the alternator start charging but once started it will self excite.

Mine was charging, just not high enough. Although it would take a good minute of running before I got a charge reading over the battery voltage. Also, I have been driving like this for over 2 years. It may be that by adding back in the globe or resistor that I just get the charge started quicker. We'll see. I will do it as soon as I can.

The current from the charge light will initially make enough magnetism in the rotor to start the charge. You cant rely on having some residual magnetism in the rotor but some do retain a bit which at high revs (3-4 x idle speed) will start charging and once a few milliamps is generated it is able to self excite. You were just a bit lucky that yours apparently did do this.

I would be replacing the Hall sensors. Some years ago the K75 I am building into an adventure bike had a slight miss and after new plugs, leads etc I persuaded her to put new Hall sensors in and no more miss.
Someone in the US may be able to point you to where to get them there as I only have a source in Oz at $20 each. It may be Radio shack.

If you are considering them, here are some coils just posted on eBay.  Seller is open for offers.

http://www.ebay.com/itm/BMW-K100-Coils-/271827577783?pt=LH_DefaultDomain_3&hash=item3f4a2fabb7&vxp=mtr

Need some cleaning in Dawn dish detergent followed by a good spray with WD-40.  Polish up the terminals with some 600 grit.

I had some good news today. I replaced the regulator on the eBay alternator and am now getting 14.11 V with the headlight on! Schwing!!

Now I just need to ride it and see if the problem is still there. Most likely it is but before I start looking into the coils or hall sensors I want to confirm it.

In reality it could be any number of things causing your poor running symptoms but with a new bat and the alt charging properly now, I wouldn't be surprised if you've seen the last of your stalling issues... here's hoping anyway and good luck. 

Interestingly, over this past long weekend I needed to move my old girl under cover due to an approaching storm. At low revs and at idle she was running very rough and had poor response to throttle. I also heard the starter relay chattering. Turned her off immediately, removed the battery, alt regulator and starter motor. Cleaned and checked everything. Cleaned alt connector pins, all the earths including the base of the starter where it bolts onto the GB. Starter brushed are only just within limits so will be replacing those asap (I think that may have been the problem. Not enough pressure on the brushes to make good contact). Checked starter, fuel and LS relays to ensure no damage had been done. Cleaned fuse contacts. Applied dielectric grease to all connections.

Not only did she fire up beautifully and run very smooth with good throttle response but I'm getting an extra 0.2V charge at the battery now.

kringb wrote:I had some good news today. I replaced the regulator on the eBay alternator and am now getting 14.11 V with the headlight on! Schwing!!

Now I just need to ride it and see if the problem is still there. Most likely it is but before I start looking into the coils or hall sensors I want to confirm it.

some good news at last

woohoo !!

keep at it

This problem is a real tease. After getting the alternator chargining properly I wasn't able to ride much until this past week. On the short 10 mile rides to work and back there were no problems. Yesterday I had to go to a place about 80 miles up the highway. All the way there I had no problems but forced myself not to get too confident. Coming back, about 30 miles in I was litterally just starting to think about the "case closed" post that I was going to write on this thread when the bike cut out. So, I am back to searching.

Here is what happened:
Riding along at a steady 70mph the bike suddenly looses engine power. It still has electrical power. The gauges including the tach still work and I can still hear the fuel pump running. Twisting the throttle does nothing. I put my blinker on and start to head onto the shoulder when it suddenly regains power. It did this a number of times and by keeping it at about 60mph I am able to manage it and contine on without too many cut outs. I drove this way for about 15 miles and then decided to pull over to see what would happen after letting it rest. I left the bike sitting for about 20 minutes then restarted and drove the rest of the way home without problem.

I know the fact that it ran ok after cooling down might point to the hall sensors. I am prepared to look in that direction but here are my doubts. If one of the sensors was failing wouldn't two cylinders still be firing? From previous tests I know that the fuel injectors are not receiving a signal to open. Would the hall sensors cause this. And also, I thought that I had read that the hall sensors tell the ecu when to turn off the fuel pump, is that correct.

Anyway, while my uneducated guess is still the mass airflow sensor which I don't know how to test, I would like to test the hall sensors.

This morning I took a look at them and visually the sensors and connector all appear clean. Because this problem is intermittent and only occurs under load (I have never had it happen by just revving the engine) can I test it while driving? Can I plug an led into the connector and watch it while riding or is there a better way? I tried heating it with a hair dryer a while back and could not get the problem to occur.

Thanks to everyone for sticking through this with me.

Sounds like its not cutting out completely???
You say it's loosing power and then recovering?
Could be it's loosing spark on 2 cylinders. Hall effect sensor or maybe a bad coil. I think you mentioned you have the black coils.

Might be worth pulling the spark plugs just after its 'spat the dummy' to see if any are wet. If a coil failed you'd get 2 cylinders with no spark but still getting fuel. They fire in pairs. 1+4 and 2+3. Just a thought

I have just reread this thread again and my conclusion is that the hall sensors are most likely the problem with the coils in second place and a good way back.
I don't think that testing them would serve any purpose because by the time you get them out they probably have cooled enough to work.
You can get the individual sensors for a fairly good price as compared to OEM or even EME. I paid around $20 each for the ones I bought. I bought them through Jaycar but I don't think they are in the US so Radio shack would be a good place to start the hunt or maybe someone there would know where to get them.
You would need to fit them onto the plate and connect the wiring. They are riveted on so that needs to come off to remove them and small bolts with a countersunk head are needed to fasten them back on.

I was reconsidering whether it might still running on two cylinders when i remembered that the fuel injectors are not being turned on when it cuts out. It is my understanding that they all fire at once so when it cuts out all four cylinders are loosing combustion. It may be getting spark in only two or none at all or maybe its still getting spark in all.

I wil look into replacing the hall sensors but I sure wish I could confirm whether or not it is getting spark on any cylinder before undertaking the task of drilling rivets and all. 

There must be a way to monitor the spark going to the cylinders while driving, any suggestions? 

Otherwise, would plugging an led into the back of the HES connector cause any problems or show me anything useful? I am assuming I would see a flickering light if it was working properly. If I understand the plug correctly then i would need to insert the positive lead from the LED into Pin 3 (red wire) and the negative lead from the LED into pin 2 or 5. Is that right and safe to do without doing any damage?

I wanted to write in to give some closure to this post I started a while back. After my last posting, I connected an led test light to one of the hall sensors so that it would flash whenever the sensor was sending its signal. The problem went into hiding for a bit but eventually returned after a few months and I was able to confirm that I was loosing the signal from one of the hall sensors when the bike was cutting out. Rick G had correctly suspected this but it felt good to have proof in that led test light. My symptoms were not exactly typical for the sensor failure. Heat would not make them fail and they fail temporarily and then be fine (not needing to cool down), for example. It is possible there was some short in the wiring at the sensor or between the sensor and the wiring harness connector, I don't know. As I was researching how to grind out the rivets and install new sensors onto my plate I saw an inexpensive unit on ebay and bought it. I put it in and the problem is now solved. 

I don't usually like swearing in internet forums but Freakin A that was a long time chasing that problem! But I did learn a shitload about my bike in the process. 

Anyway, thanks to all on this forum for so generously sharing your knowledge.

Congratulations on finally finding the problem and kudos for the dogged determination and persistence in pursuing the fix.  Thanks for the update, and here's wishing you many trouble free future miles on your brick.

Great to know you have solved the problem. I always say there is nothing harder to sort that an intermittant electrical fault. Many a man has ended up in the home for the premanently bewildered over them. I had an interior light fault in a 3 series BMW car and in 3 years I never figured it out. Often I have found that the harder they are to find the easier they are to fix.

In the interest of full disclosure and the edification of others who have as yet not gone down this path, a question or two.  Did you investigate deeply enough to determine if the fault lies in the sensor or the attendant wiring? If you bought a used OEM set-up do you know its age?  From what I've read, it's as likely the wiring has disintegrated as the hall sensor itself has died.
I've bought a pair of the Chinese sensors and intend to fit them during the winter lay-up of my K100RS.

Brickrider: No, I didn't investigate any deeper to pinpoint the fault. i just know that one of the sensors was losing signal. The connector to the harness was good so it was someplace after that. I eyeballed the wiring of the HES and from what I could see (not much) it looked fine. It was not hard or brittle or anything like that. But from my symptoms I would suspect that it was an electrical short problem someplace and not the actual sensor. But i really don't know. 

The HES I bought was off a 1987 k100Lt and had something like 40K miles supposedly. The seller was asking $55 and I offered him $45 because buying used electrics is such a risk and he agreed. It was cheap enough that if the unit was also bad i figured I could use it for practice in grinding out and installing the new sensors. It may fail at some point but at least now i will have two to mess with.

When the wiring fails its the insulation properties that fail when exposed to high temperatures. Their insulation properties return when they cool. However you said in your post "Heat would not make them fail" so I suspect it was the sensor itself that has failed (not unheard of).

Interestingly, newer sensors now have cross-linked polymer insulation which withstands heat much better than the old OEM sensors