This post describes my work inspecting and renovating the charging system on my ’84 R100RS.  The project falls short of an overhaul because only rudimentary testing and inspection was performed.  Indeed, this project is more in the category of “preventative maintenance.”  The diode board and voltage regulator seemed to have been working normally, but I replaced both as a precaution due to their age.  Where easily accessible, wires and cables were replaced for the same reason – although some appeared to have experienced heat stress or corrosion.  A great deal of thought was put into diode board mounts and grounding.  I hope my efforts result in continued trouble-free operation on the cross-country trips I have planned.

A Little Background

My experience with this motorcycle’s charging system goes back over 30 years ago when I bought a new ’84 R100RS from a BMW dealer in Walnut Creek, California. Within the first few years of ownership I replaced the diode board two times. The diode boards malfunctioned because solder melted out of them.  The ’84 R100RS in this project is one I purchased 3 years ago from Mary and Perry Bushong in Fort Worth, Texas.  It’s identical to the one I bought new in 1984 – both are “Last Editions.”

After the second diode board replacement on my new R100RS, the dealer suggested I replace the front fairing with a “police” version.  The diode board sits behind the front engine cover and the “police” fairing allows much greater airflow for cooling the front of the engine.  I was told that the stock fairing on my R100RS provides less airflow because it was designed to work with the addition of the new oil cooler intended to manage engine temperatures.

• LEFT – “Police” front fairing
• RIGHT – Original ’84 R100RS fairing

While conducting research for this project I found a BMW Service Information Bulletin that documents BMW’s acknowledgment that motorcycles were delivered with defective diode boards.  The bulletin authorized replacements under warranty. It seems I paid for at least one diode board replacement unnecessarily back in the 80’s.  After installing my second replacement diode board and installing a “police” fairing, my diode board never failed again.  I don’t know if the better cooling of the more open fairing or if the improved diode board resolved my problem.

The Scope Of This Project

I bought my “new to me” ’84 R100RS with the “police” fairing installed.  Luckily the previous owners kept it and passed along the original fairing to me.  In addition to ensuring that the charging system does a good job of keeping the battery charged, I hope to have the diode board working well enough to use the original fairing again – even if I can only use it in cold weather.

This tune up project includes:

1. Installing a new OEM diode board
2. Installing new OEM rubber diode board mounts
3. Making and installing new diode board grounding wires
4. Installing new alternator and diode board wires
5. Inspecting the alternator brushes for wear
6. Measuring the resistance across the armature rings for faults
7. Installing a new voltage regulator
8. Installing new battery cables
9. Inspecting and cleaning the starter relay
10. Replacing the engine breather hose

A comprehensive overhaul of the charging system would include more tests of the  alternator’s stator and armature.  Since the charging system is delivering a reliable 14 volts or so – I decided not to do those tests now.

Diode Board Mounts and Grounding Wires

Some years after I sold my first R100RS I began reading about new metal mounts for the R100 diode boards that cured their problems. The new metal mounts reportedly provide better electrical grounding and heat transfer than the OEM rubber mounts. They also have the advantage of never failing – eliminating the “pain in the ass” task of replacing a broken rubber mount.

After opening my engine’s front cover and removing the diode board I discovered that a previous owner installed two home-made metal diode board mounts. They were installed in the top two positions, grounding the diode board to the timing chest cover. The bottom two mounts were still OEM rubber mounts. The diode board’s two bottom mounting positions are electrically insulated and don’t benefit from grounding.  No diode board grounding wires were presently installed on this motorcycle.

• LEFT – The two home-made metal mounts previously installed on my motorcycle
• RIGHT – The four mounting positions on the timing chest cover

My research relied heavily on Robert Fleisher’s comprehensive article on Diode Boards and Grounding Wires. To read it: Click Here

Fleischer’s article has a partial history of the Service Information Bulletins pertaining BMW’s efforts to eliminate diode board failure. His article omits the first bulletin that authorizes dealers to replace owner’s diode boards with a newly manufactured version.  This bulletin also explains how to identify the new and improved diode boards by their different paint on the back of the circuit board.

Subsequent Service Information Bulletins provide instruction on grounding improvements for the diode board.  I presume that even the improved diode board would fail without improvements to the diode board’s grounding. None of the bulletins reference changing the front fairing to a “police” version. And none of the bulletins reference changing to metal diode board mounts.

The original diode board grounding design uses two fairly light gauge (I estimate 16 AWG) wires to ground the diode board to the timing chest cover – one to each of the two top mounts. The last Service Information Bulletin directs adding a new supplemental grounding wire.  Fleisher’s article refers to it as a “spider” wire.  The “spider” wire is also only about 14 or 16 AWG wire. It provides two additional ground wires, both to the engine case.  Additionally, it has a wire to ground the timing chest cover to the engine.

Both metal mounts and the original two grounding wires only ground the diode board to the timing chest cover.  This series R100 has a black coating on the timing chest cover that may affect its electrical conductivity to the engine.  The new “spider” wire adds grounding from the diode board directly to the engine.

My “new to me” R100RS had metal mounts but didn’t have any grounding wires.  Since the diode board hadn’t failed, any combination of things might be contributing to its success:

• The timing chest cover may have good electrical conductivity to the engine.
• The diode board may be one manufactured to withstand high temperatures.
• The “police” fairing may be keeping the charging components adequately cool.

Another factor is that metal diode board mounts transfer heat and rubber mounts provide some heat insulation.  Robert Fleisher shared with me that he had measured diode board temperatures using a highly accurate thermocouple thermometer.  He did not share his specific results or testing conditions – but he fully supports using metal diode board mounts.  It seems counter intuitive to me that a diode board rectifying no more than 280 watts using appropriate wiring and connections would generate enough heat to achieve higher temperatures than the engine itself.

So whether the metal mounts allow heat to flow from the diode board to the engine or the rubber mounts insulate from heat moving from the engine to the diode board remains to be tested and reported by someone interested with a bit of time on their hands.  Frankly, I’d like to do this test myself.  The only other factor I can think that might be relevant to the diode board’s longevity is if the rubber mounts provide any significant vibration damping.

Because BMW never specified using metal diode board mounts, I decided to use four new OEM rubber mounts.  Many respected Airhead experts advise against this. (To the point of being described as “dogma.”)  This meant I had to use grounding wires.  I bought a complete set of OEM grounding wires but found the terminals poorly connected and made with light gauge wire.

To ensure good diode board grounding I decided to make my own version of the “spider” wire using heavy 10 AWG wire and heavy .8mm brass ring terminals.  I also sealed the ends with waterproof heat-shrink tubing.

• LEFT – The heavy .8mm brass ring terminals required using a vice to crimp.
• RIGHT – The sealant from the waterproof heat-shrink tubing required a bit fo trimming to make the full surface area of the ring terminal available.

I connect one of my ground wires to the engine and the other to the timing chest cover – and a third wire grounds the timing chest cover to the engine.  This duplicates the function of the “spider” wire – but provides about twice the electrical load capacity over using all of the OEM ground wires.

• LEFT – My 10 AWG grounding wires.
• RIGHT – The grounding wire attachment points to the timing chest cover and the engine.

Inspecting and Replacing the Diode Board

A separate blog entry contains photos and descriptions of the diode board previously installed on my R100RS and the new OEM diode board I installed. To view it Click Here.

The diode board I removed shows no sign of heat stress other than some blistering of the paint coating the back of its circuit board.  I replaced the diode board with a new OEM part in the hope that improvements in the diode board’s components and manufacture might make it even more resilient to heat stress.

Removing and Replacing Diode Board Mounts – A Few Tips

One of the most common reasons I’ve read supporting metal diode board mounts is that they are unbreakable and never have to be replaced.  Removing and replacing lower diode board mounts in particular are said to be a very significant “PITA” or, pain in the ass.  In addition to figuring out how to get a wrench on the diode board mount’s nuts inside the timing chest cover, there was the fear of dropping nuts and washer into the timing chest.

Several tips make removal and replacing diode board mounts relatively easy.

• Removing the breather hose and ignition sensor connection provides greater access to the right diode board mounts.

• An 8mm flex-head ratcheting box wrench provides easy access to the rear diode board nuts.  The right lower nut can be reached through the front of the timing chest cover.

• Tape on the wrench keeps the nut on the wrench until its threaded or after its removed.

• Using “Whiz” nuts make using washers unnecessary.
• There is a “shelf” on the back side of the timing chest cover that prohibits dropped fasteners from falling into the timing chest.  Dropped fasteners are easily retrieved with a magnet from the back side of the timing chest cover.  It is possible however to drop a fastener into the alternator housing.

New Wires

In addition to the diode board ground wires I made, I bought and installed new OEM wires connecting the diode board to the starter and the diode board to the alternator.  Two wires to the alternator and one wire to the diode board are part of the motorcycles wiring harness and appeared to be in satisfactory condition.  I cleaned their terminals with Caig Deoxit and reused them.

• LEFT – old wires
• CENTER – new wires and new diode board
• RIGHT – finished installation

Measuring the Alternator Brushes

I bought new alternator brushes already mounted in the brush holder to avoid having to solder new brushes.  Soldering isn’t to challenging – this is just one place I’d rather avoid making a mistake.

I was prepared to remove the existing alternator brush housing but found I didn’t have a proper wrench to remove its fasteners without detaching the stator from the alternator housing.  So before tackling that, I decided to simply measure the brushes to see if they actually needed replacement.

Specifications on acceptable alternator brush length varied according to different references, so I elected to use the one that specified that the alternator brushes should be no less than half their original length.

Comparing the length of my new brushes to the existing ones showed I was nowhere near needing to replace them.

Testing Voltage Across Alternator Rotor Slip Rings

The factory service manual’s technical specifications list the resistance across the alternator rotor’s slip rings as 3.4 ohms,  +/- 0.34 ohms.

Testing mine with a digital volt-ohm meter I found mine to have 3.15 ohms – within spec.

Several other alternator tests are available.  The alternator brushes and the slip rings are readily accessible – and moving parts.  Since my alternator has been  working normally, I limited my tests to those items measuring the resistance across the slip rings.

Inspecting and Replacing the Voltage Regulator

I describe the original and new voltage regulator in a separate blog entry. Click Here to open that entry.

This motorcycle still had an original “metal can” electronic voltage regulator – which is adjustable.  I replaced it with a new non-adjustable plastic case voltage regulator simply because of the age of the original part.

Installing New Battery Cables

My research revealed that battery cables the age of this motorcycle can experience corrosion within the insulation by creeping inside through the cable over time from the exposed end of the cable.  For that reason I decided to replace both the positive and negative battery cables as a precaution.

Before installing the new cables I treated the exposed cable ends with Caig Deoxit and then sealed them with liquid electrical tape.

Inspecting and Cleaning the Starter Relay

From my reading, it seems possible that the starter and the headlight relays, along with the ignition switches, can malfunction causing problems with charging and starting. I had replaced this motorcycle’s headlight relay earlier, so I elected to simply clean and inspect the starter relay.  As nearly as I can tell, it works normally.  I cleaned the relay’s contacts and terminals and it’s socket’s terminals with Caig Deoxit.

What’s Next

This motorcycle still exhibits two things related to the charging system that may indicate faults I still need to investigate.

1. The instrument panel voltmeter swings more widely when the turn signals operate than I believe may be normal.
2. Even with the new headlight relay, the headlights to not appear to dim when activating the starter.  The starter has no difficulty starting the motorcycle – so that fault isn’t yet too worrisome. When cold weather comes in a few months, investigating this may be a higher priority.