An Apparent Charging Problem

[MonoRT]

Still Getting Experience...

Kurt, thanks - I will perform these checks when I next get a chance. This is a long-haul effort as my nest is far from empty and I am still involved in 24/7 systems support. Also, weak charging is annoying, but survivable compared to no charging.

Based on the almost 16 volt unregulated output, I donÔÇÖt WANT to believe that I have an alternator problem. However, when you are trouble-shooting, the root problem doesnÔÇÖt care what you WANT to believe.

Lew, I followed your thread with interest as it played out. My original battery was not known to have a problem, but it was old enough that it seemed wise to eliminate it as a possible source of trouble. When I first put in the new battery I did not charge it, but rode the bike right away. The voltage displayed by the meter was pretty low and I thought shoot! this is even worse than with the old battery. Then, I charged the new battery and went for another ride. Right off the charger, things looked good  for a bit and then the voltage settled back down again. The current battery reads about 13 volts after being on the trickle charger overnight. It always spins that old Bosch starter with authority  even when Ive been running lots of stop and go errands  but, so did the old battery. I put in a Westco sealed unit, so I cant test for a bad cell

I keep talking about running a heavy wire back from the diode board to the battery. Rick Jones at Motorad Electrik suggested that and Ive not tried it yet. I kind-of want to change one thing at a time. The last time I changed the timing chain, I put in one of those silicone gaskets between the timing chest and the engine case. I wonder if that could contribute to a poor ground path? Still, everything is BOLTED together and Id expect a good ground path through the bolts

 

[lmo1131]

Kurt, do you know if the R80 had rubber diode board mounts? The MAX fiche does not identify either way (in fact the reference identifier #8 isn't listed in the BOM)

If they are rubber mounts, current tribal wisdom says to replace them with solid ones (Rick has them). Perhaps combined with the silicone gasket you're not getting adequate grounding ... .

Paul Yank that tank off and make a ground cable to run back to the battery. Either way it won't be a wasted effort. I ran a 12ga wire up through the rubber "D" grommet that the diode board loom runs through. It distorted a bit, but when the front cover is tightened down it's fine. As I mentioned, I ran to the frame ground at the coil mount, thence, to the battery with a second wire. The reason I did that was to reduce the number of ring terminals that the battery (-) bolt had to try and hold down (mine seems a bit short) - better practice might be to run a continuous wire from diode board to battery.

BTW... personally ... I think using the speedometer cable bolt to disconnect the ground cable is asking for trouble ... it's easily stripped. Better, IMO, to use the battery terminal ... it's a pain to pull a saddle bag to get the seat open, to get the tool tray out, to disconnect the battery .. .. but it's one less worry I have about my old /5.

 

[20774]

As for rubber mounts, Snowbum has this on his website:

"2. BMW used rubber diode board mounts on MANY models. From about 1979 to about 1993, with some exceptions. Airheads that ALREADY SHOULD HAVE solid metal diode board mounts are: /5 models; /6 models; 1978-1987 R65 and R80 models."

http://bmwmotorcycletech.info/diodebds&grdgwires.htm

Seems to me that when I replaced my mounts on the /7, they were rubber...my bike was a September 1977 build.

 

[Walking Eagle]

Great Thread!

This is one of those "print-worthy" threads to put in the Clymer for future reference. Golden info on troubleshooting from everybody, and THANKS.

On two different '78 R100RS (MS) "late '77" builds, both had RUBBER Diode mounts. Changing them is EASY if you have hands the size of a four-year-old child; otherwise, a PITA, but easy to do as long as you don't drop any of the NON-magnetic nylocks down into the darkness. (Good luck with that, but they don't go into dangerous places.)

Walking Eagle

 

[MonoRT]

I'm Getting MORE Experience

Today, I started my ride to work with 13 volts indicated at 4K RPM and up (battery had been on the charger overnight). I had an (unusual) 16 mile run on the 405 where traffic was bombing along in the 70-80 MPH range. While moving at this speed, voltage slowly dropped to 12.4. The GEN light does not come on, and voltage does vary with RPM.

I was planning to go on a 300 mile ride with some of my Harley buddies tomorrow, but unless I can get somewhere this evening, I’ll be messing with my Prince of Darkness electronics tomorrow instead of riding. I take enough flak from the Harley crew – I don’t need to conk out a couple of hours into the ride.

I plan to run the checks that Kurt suggested and I’m going to take a very careful look at the brushes and rotor again. I have new brushes on-hand. I really like the look of the Airhead exhaust headers as they curve forward and then back down, but when you are running the engine and reaching in around the front wheel to poke at the electrical bits… I generally hear the hiss before the pain registers.

Lew – I just replaced the wiring harness between the VR and brushes. I noticed that extra hole in the rubber grommet that goes into the top of the timing chest. I’ve got 10 and 12 gauge black wire on hand, so I may try a ground run back to the battery. Who did the load test on your battery? Can you talk Sears or an auto parts store into doing that?

My 85 R80RT does have solid diode board mounts and I do have a daughter with smaller hands than mine who has helped me remove and replace that board in the not-too-distant past. If I determine that the Motorrad Electrik diode board is a problem, I still have the original Wehrle...

I’ve really benefitted from the experience and experiences of others on this forum. Now, it’s my turn. Even if I find the problem to be something really embarrassing – something that I’ve repeatedly dismissed as a potential problem, I’ll post my findings.

 

[lmo1131]

A guy across the street from my shop operates an automotive electrical shop. He has a hand-held device that he attached to the battery that "loaded" it. Sorry I can't be more specific with the details. I do know that the battery MUST be fully charged for the test (leave it on the charger all night and take it in ).

I believe any NAPA has a similar device. Interstate Battery does here on the central coast, I would imagine it's standard equipment for all their shops. Sears probably does. O'Reilly's, probably. I would think BatteriePlus would have one.

 

[MonoRT]

Positive Wiring?

On the way home, voltage was all over the place, varying from 12.2 to 12.9. I wondered about the brushes.

When I got home, I replaced the 11.6MM long old brushes with 16.5MM new brushes. Voltage at the battery was 12.2 @ 4K RPM. Brushes don't appear to be a problem, but I won't have to worry about them for awhile.

I was running out of light, but I checked the 3 phases out of the stator in AC mode. I got 8.9, 8.8 and 9.8 volts AC. Does a one volt difference between phases indicate a problem?

I quickly checked the B+ to ground (timing case) and got 13.8 volts!

Sounds like a wiring problem... Tomorrow, I'll pull the tank plus starter cover and start tracking the plus cables back towards the battery. I guess the ground path is worth checking also.

 

[airraceaddict]

(Thanks Lew for adding the drawing with the wires for the system). I fought and fought and fought with my 78 R100RS over a similar issue. It finally gave up it's secret. I replaced item 11 (which now comes as 3 separate cables with a black sheath). DAMN if all my charging issues didn't go away and has been excellent ever since. There are a lot of places to check voltage and when you mentioned the difference in voltage you were getting for the 3 different phases, BAM! That's when I thought I would jump in. I had disconnected and reconnected that #11 harness several times but was not relating the immediate results with that particular piece, (slow learner I would get the requisite charging only to have it drop off again just like you are talking about. What I would do is disconnect that harness at both ends and check the resistance on all of these wires. Chances are you have a bad connection at one end or the other of 1 or more of those wires. You can get the connectors at Radio Shack or a good auto parts store and build your own harness.
In my opinion only, I think you are getting too wrapped up with the voltage you are seeing. The voltmeter is a good reference but not the ultimate determination of correct operation. If you got rid of the voltmeter, I seriously doubt you would think you had a problem. If the gen light is coming on when it is supposed to, and going off above 900-1100 RPM, and you weren't having dead battery issues, and the alternator is putting out voltage like a raped ape with the VR wires jumpered, you should be good to go. My voltage indication on my Voltmeter is about 13.2 at all RPM's above 1200 RPM. And I have not been hooking it up to the charger except for the winter months. However, you are not wasting your time. If you go to the trouble to go through and clean all of those connections carefully, you will not regret it. (If you don't induce more problems by breaking something that was working before). The upside is you are getting a good idea of how this thing works and will gain valuable knowledge for troubleshooting and fixing a problem on the road.
Back to your regularly scheduled programming.

 

[20774]

Reading 13.8v at the output of the diode board is a good thing...that means that everything about the system to that point is basically as it should be. Now, getting that voltage, without significant losses back to the battery is the trick. Hopefully you'll find that one of the wires has significant corrosion or that one of the terminals at the starter has corrosion or is loose.

 

[MonoRT]

Well, charging remains poor for the most part. I commute 40 or so miles a day and the battery loses charge over time because it is not getting quite enough output from the alternator. Every third day or so, I put the bike on a charger overnight. This is OK in commute mode, but I canÔÇÖt afford to try an all-day ride.
HereÔÇÖs what IÔÇÖve done so far:

1. Unplug, clean, reconnect charging subsystem terminals and connectors (MANY times).
2. Battery ÔÇô replaced because it was old, not because it appeared to be a problem.
3. Battery positive and ground cables replaced. Again, they did not appear to be a problem.
4. Wiring harness from VR to brushes replaced. Insulation was brittle and cracked.
5. Brushes replaced. New brushes were only 5mm longer than the old ones, but why not?
6. Voltage regulator replaced. Voltage jumped dramatically when the VR was bypassed. However, the new VR appears to have had no effect on the overall charging situation.
7. Inaccurate Motometer voltage meter replaced with a fairly accurate VDO unit.
8. Added a 10 gauge wire from the diode boardÔÇÖs ground to the negative terminal on the battery. This also appeared to have no effect on charging behavior.
9. Based on some other threads in this forum, I purchased a Datel LED voltage meter and used it to monitor voltage at various points while I rode the bike. I wish I had done this sooner.

I realized that single point in time measurements with the Fluke meter were not giving me the whole story. The readings that I would get with the bike partially disassembled and running in the driveway were different from what I observed on the road.

I had been trying for real-time voltage monitoring with my Fluke meter in my tank bag. This turned out to be sub-optimal because it is hard to look at the tank bag while riding and because I was too stupid/lazy to make ring and spade terminal adapters for my meter leads and alligator clips like to pop off while the bike is running. The Datel meter is great in that it is a 1 ounce block of water proof plastic that can be stuck just about anywhere with hook and loop tape. It is accurate and only requires two leads and, with a switch, you can wire multiple wiring junctions to the switch and then switch-select the point that you want to monitor. Direct sunlight on the face of the meter does render it unreadable, so you want to think about placement.

HereÔÇÖs what I now know: Voltage readings at the battery, at the starter positive terminal and at the diode boardÔÇÖs B+ terminal are virtually identical. Voltage readings in the switched circuits run about .35 volts lower than the un-switched junctions mentioned before. Most of my ride-time monitoring has been at the B+ diode board terminal and at the battery terminals. Poor charging in the 12.7 to 13.3 volt range is the norm when RPMs are at or above 4K. Occasionally (usually in the evening), I get as much as 13.85 volts but never for more than a few minutes and only once every few days. Sometimes, as IÔÇÖm accelerating and RPMs are rapidly heading north of 4K RPM, voltage actually starts dropping ÔÇô IÔÇÖm guessing that the ignition system is using more juice, but the alternator is not able to meet demand even though it is spinning faster. The GEN light works, it comes on at around 900 RPM and below. The Bosch starter always spins the motor over with no problems. Battery voltage does not drop below 11 volts when the starter is spinning ÔÇô that seems like a good test for the battery and primary cables.

So, IÔÇÖm wondering about the following potential issues (in no particular order):

1. Brushes making poor contact with the slip rings because they donÔÇÖt slide cleanly in their bores? When I pull them up by their tails and let go, they seem to snap back down onto the slip rings with authority.
2. Out of round slip rings that bounce the brushes off their surfaces at higher RPMs? How would I know?
3. Iffy connectivity in the rotor or stator? Resistance measurements are OK by my manuals. The last time I checked the AC output of the three stator phases, one phase was a volt higher than the other two.
4. Bum diode board? I have a 5 or 6 year old Motorrad Elektrik model.
5. Current leak in the switched wiring? The battery does not run flat overnight, so the un-switched wiring seems OK. It would have to be a small leak or something would melt or catch fire.

There was a wise suggestion to stop worrying about the voltage meter and just enjoy riding. If I was not a programmer (with the personality flaws that are commonly associated with this profession), I might have done that. In commute mode with occasional overnight charging, the bike works just fine. However, being a programmer, I just HAD to have more data ÔÇô thus the VDO meter and then the Datel unit. Unless I have a current leak in the switched wiring, IÔÇÖd have been way ahead if IÔÇÖd just bought one of the two aftermarket changing systems that are popular with the airheads. IÔÇÖve sunk a bunch of money and time into this and I still donÔÇÖt know what is wrong ÔÇô although I have eliminated a few things.

 

[lmo1131]

just curious...

my /5 puts out 14.1v with the headlight OFF. With it ON, I see 13.85 (as read with my voltmeter in the tank bag).

I can run without my /5 with head light OFF (aftermarket 55W H4 ) ... can you? Just wondering what your output would be if you disconnected the head light....

 

[Stan_R80/7]

With the basic charging system consisting of an alternator/diode board assembly, voltage regulator, and battery - obviously one (or more) of these components is not working properly. I have to ask: have you tried using an adjustable voltage regulator? While that probably will not fix the fundamental problem with the charging system, it may allow riding the bike until the problem becomes more obvious. The beemershop sells adjustable voltage regulators for $20.00.

 

[MonoRT]

I am coming to the conclusion that something is going wrong in the power generation department. The battery is new and seems good, the primary cabling from the battery has been replaced, voltage at B+, starter and battery is the same...

I've told myself that if I have to replace one of the expensive parts (rotor, stator or diode board) that I'd just go for broke (literally) and put in an aftermarket charging system. Heck, I could install it in my sleep at this point.

I did wonder about using a variable output VR and just cranking it up. I have 3 VRs now, none of which are known to be bad and all of which yield the same results in terms of charging voltage. I did get lots of voltage the ONE time that I jumpered around the VR, but I've only done that once and given the variations I'm seeing in the output, I'm not sure that I can trust that test - but, the alternator sure did chank out the power with the VR removed - that test prompted me to purchase VR #3 - which had no observable effect. If I used a variable VR and set it to 14.2 volts when the system is in its "normal" mode, I'd probably end up frying something when the system has one of its random fits of almost good output.

The one last cheap replacement would be to replace the alternator/diode board external wiring. Those wires were included the M-E diode board kit, so I'm not running the originals, but it still would not cost much to replace the lot of them. I see that Euromotoelectric sells these wires as a kit.

Lew - I have not tried a headlight-off test yet, but I did think about it. My 85 has a flat round spot on the switch pod where the headlight switch ought to go. However, I do have an Eastern Beaver relay setup, so I can just pull a blade fuse out to turn off the headlight. I'll give that a whirl sometime.

 

[Stan_R80/7]

If you haven't read the Bosch charging article from the resources page, here it is:

http://www.buchanan1.net/charge.shtml

If the charging light is completely dark at night when running, then the basic charging system should be OK. Any intermittent variations (or problems) with the alternator output should be reflected in the charging light. Based on the intermittent voltage swings described, I think your bike has a (some) bad connections somewhere. Good Luck!

 

[140008]

That vintage bike is notorious for not being able to support it's battery in commute/stop and go traffic.Upgrade your EL system! Get the EME or Motorrad Elektrik upgrade,I got the Motorrad.You could unplug your head light,but that's really not a good solution.Bite the bullet,it's just money.

 

[20774]

Is it possible that you have a diode out or bad on the board? With no regulator, maybe the reduced output would be masked because you saw higher voltages with the VR bypassed. But with the VR in the system, things are working such that the weakness of a diode is pulling the total output down.

 

[MonoRT]

This Saturday, I went back through every connection and relay that I could find, checking for visible problems, cleaning with a Dremel wire brush (where practical) and contact cleaner. I did not find anything obvious. I did a simple check for bad diodes by checking the voltage with the bike running and the meter set to measure AC - no AC voltage.

At several points, I ran the bike with one of my three VRs or a jumper between D+ and DF in the VR plug. The Wherle and Transpo VRs are both rated at 14.2 V max. The VW unit that I bought when I first thought I had a bad VR is rated between 13.8 and 14.8. None of the VRs are known to be bad. I bought the 14.2 Transpo unit because I mistakenly believed the Wherle to be a 13.8V version (it has "14/2" molded right into the case). The VRs all give very similar driveway results - around 13.3 volts at 4K RPM and above.

The jumpered VR plug will give 16 volts by 2,500 RPM - I did not run like that for long! So, the alternator/diode board can make plenty of power, but the VRs want to be overly restrictive when they are calling the shots.

I was trying to figure out how the bike could make so much power jumpered and then get weak with any of the 3 VRs. I checked the D- circuit carefully, but that IS all-new as I had replaced the VR plug and its harness earlier in the story. I wondered about D- because it serves as a reference point to the VR and is the one input that is removed from the equation when you jumper D+ and DF. The D- brush holder, the D- spade lug and the brown D- wire to the VR plug all show excellent connections to ground.

I ordered a variable VR today. One, it is possible that I have 3 bad VRs. Two, it will be interesting to see if I can get to 14.2 volts at 4K RPM (that's all I'm asking for!) with a variable unit. It is also possible that the small diodes on the diode board are sending plenty of positive juice to the VR and rotor via D+, but that the power is jumpy, wavy etc... in a way that confuses the VR into acting as if the voltage is higher than it really is in the rest of the system.

 

[20774]

I continue to be impressed with your detailed evaluation of the problem...I know that you'll be greatly satisfied if you find the problem.

Have you considered testing the VR to see if they indeed to cut off at the right level? I believe Snowbum has instructions on how to do that. I tried that at one point with either/both of my VRs, the stock mechanical and the replacement Transpo.

 

[35303]

Tin Whiskers

Lets make a stretch here, tin whiskers causing intermittent shorts in the regulator. These things caused 3 satellite failures, pacemaker failure, Toyota Camry accelerator failure and others as NASA shows on its whisker site.

http://nepp.nasa.gov/whisker/background/index.htm

 

[lmo1131]

wull, golleeeee... and all this time I thought it was