De-mystifying and Troubleshooting BMW Airhead Boxer Motorcycle Electric systems

© Copyright, 2012, R. Fleischer

boxerelectrics.htm-14A

Please also see my other electrical articles, in particular, ElectricalHints.htm

This is an article to furnish THREE types of information:

(1) Some BASIC and ADVANCED INFORMATION on electricity and Airhead problems. The approach used here is different than in manuals and troubleshooting guides that you might have, or are contemplating obtaining.  Although some hints are given later in this article on some common faults, this article should be used in conjunction with other articles on this website.   particularly:   ElectricalHints

(2)  Common problem areas, explanations of some of the circuitry, etc.  A discussion of such as batteries; starter motors, voltage regulators, etc.....a goodly amount of good technical information.  Other articles will get deeper into some of these things.

(3)  An addendum that may discuss some particular point that has come up, or some topic of interest.  This is at the very END of this article

Available to you are certain helpful booklets, etc.  Guides such as the ones from Motorrad Elektrik, Chitech, and your Haynes and/or Clymers manuals (and, perhaps, a schematic in the rear of your owners booklet or on this website), may well be necessary items for you, and are actually recommended...and if you are anal enough,  get them all.  In MY OPINION, the Chitech electrics manual and the owners book or factory schematic, or schematic on my website, are THE BEST sources for electrical information for the Airheads.  I recommend you purchase at least the Chitech Electrics Manual.

The Chitech (Chicago Region BMW Owners Assoc.) Manual (BMW Electric School Manual) is THE BEST manual for BMW electrics, from basics to full-blown technical details, components, diagrams, etc., and includes the singles and all boxer airheads. It is VERY complete. Only a few errors are in it, and I have an article I did on those errors. Here is the link to my Critique of the Chitech BMW Electric School Manual:    chitechelmnl.htm
See my url.htm page for more information on Chitech, and how to order their publication.  Some of the total-bike schematics are not reproduced well, that is the only big problem with that manual.  Get it anyway.

Much of the following on basic electricity and its use in the airheads is rather simplified.  Please, no flaming from fellow engineers!

Electricity might be easiest to think of as a flow of atomic-sized particles called electrons. These little bits have a 'charge'. Get enough of these little charges (zillions) moving through a wire, and you have a flow of CURRENT. Apply the flow through something like a lamp, and if enough are flowing, the lamp will heat up, and will put out light. Too much flow, the lamp burns out just like a fuse which blows for excessive current flow.

Current flow is measured in amperes, and in many cases, very tiny parts of an ampere, such as milli-amperes or micro-amperes.   Milli-  means thousandth of; and micro-  means millionth of.   Current may be listed as A for amperes, ma for milliamperes, when some device has a current rating.   Confusingly, the symbol used to express current is capital letter i, which will look like a Roman Numeral if I put it here, which I will not.    If a diagram showed current flows to be expected, it might show something like   12 A for 12 amperes, or 12 ma, which stands for 12 milliamperes.   12 µa  would stand for 12 microamperes.   This same convention is used for resistance and voltage.

It is still popular to use water pipes to explain electricity. I find that this often VERY confusing for people.  In MY opinion, it is OK to think that water pressure is the force (voltage), that ALLOWS more flow from the faucet, at A GIVEN faucet opening. That adjustable faucet opening IS like resistance (ohms).   The symbol used to express resistance (ohms) is the omega sign, which MIGHT here show up properly on your computer:
Ω.   Resistance may be shown as R.

There is a relationship between amperes, ohms, and voltage.    The relationship shown here can be manipulated easily, so if you know any two of these items, you can calculate the third.   The relationship is simply that resistance in ohms is equal to voltage in volts divided by current in amperes.     R= E/I
In that formula, E stands for voltage.  The reason an E is used is that the E means Electromotive Force (which you can now forget!).

There are a couple other relationships you might be interested in.   Power, often just P, is equal to voltage in volts multiplied by current in amperes.   P=EI.
Since you also have R=E/I, then you can now determine volts or amperes or ohms or power....if you just know TWO others!
If you are curious about this, you would see that by manipulating these two simple formulas you can see that Power is equal to amperes squared multiplied by ohms......etc.

In order to have a CURRENT FLOWING, electrons must begin someplace, travel 'through a circuit' AND BE RETURNED to the source.  Please do not think of that idly!   Many folks have a problem realizing that a COMPLETE circuit is necessary.   Circuit here means the same thing as a closed racetrack, or some other analogy......you start at one point, and must continue ALL the way around.  If the electrons do NOT continue all the way around, there is NO CURRENT FLOW.   A battery may have an excess of electrons at one terminal, compared to the other terminal, but NO current (other than internal leakage) is flowing.  You need to have the device to be powered, a lamp for instance, connected (somehow, which means directly or through other items) to BOTH battery terminals, for electrons to flow THROUGH the lamp & other items, and through the battery too. Again...this idea of a complete circuit often eludes folks.

When electrons flow through something allowing such a flow (usually metallic), the properties of the CONDUCTOR are such that the conductor itself always (unless at or near absolute zero temperature...said here for those purist nerds out there reading this) offers SOME 'resistance' to the flow. A thinner wire would offer much more resistance to flow to your starter motor, than a much thicker wire. Resistance is generally undesirable in our motorcycle wiring, switches, and so on. You can't get away from it, however.  As more and more electrons flow through the conductor, the conductor will heat up.  This is how a fuse works.    If the connection is not perfectly tight, a teensy bit of resistance might allow enough heating to cause problems.   This is quite often seen at the alternator output terminals...where the insulators or wire ends that push over the spade terminals are often seen overheated.

 An 'insulator' can generally be thought of as something that has such a high resistance that electricity in any appreciable amount does not flow through it. 

In many electronics devices, a part called a RESISTOR is used on purpose to restrict electron flow.  There are places in your motorcycle that this is also done on purpose; such as in resistors in your tachometer electronics circuit, or in a /5 starter relay circuit, etc.  However, usually much more important to you, is that in your motorcycle you have resistances in the wires themselves, diodes, contacts in switches and connectors, relay coils, lamps, ignition coils, alternator rotor and stator windings...even the carbon brushes (close to 3/4 of one ohm for both of the brushes together), internal parts of some things like voltage regulators, ETC.   Resistances can add up, from various causes, and if excessive, you will have problems with your motorcycle.  You may gain such unwanted resistances from such as in corroded connections, or poor contacts in a switch, or a few broken strands of a wire; loose connections, etc.... all of which may cause the resistance to get high enough so you will have problems. 

A very commonly seen problem is at the plug-in STARTER relay, where even a small amount of corrosion or poor contacts at the relay male spades and/or the socket female connections, would cause big electrical problems...quite often the entire bike goes dead electrically.

Some resistances in Airheads were unusual or unexpected or not thought of at all.....and unless one knew about them, you'd never know about the problem such was causing.  One such example is that there was sometimes unwelcome resistance between the timing chest and the engine case, due to factory paint, and BMW dealt with that with a modification to the grounding of the diode board.

In our BMW Airheads, the resistance of the GEN lamp is used on purpose to not only let the lamp provide an indication of charging, but to supply the current for the initial magnetization of the alternator rotor (via the battery, ignition switch, and voltage regulator internals). 

Your alternator must have a certain number of turns of wire, in order to obtain proper VOLTAGE output. If we wanted to reduce the resistance (the unit of measurement of resistance is the OHM) to allow more CURRENT output (current times voltage is WATTS), we need either a lower resistance copper wire (via a larger diameter copper wire), or a metal that flows electrons with less resistance.   The existing alternator physical size is fixed, so we can't pack more volume of wire into the alternator, it is already fully filled.   Both increasing the wire size (which would give less turns, less voltage) or changing the copper to the only thing better, silver, are either impossible or impractical. If you are thinking about this here, you will realize that to get more WATTS of alternator output, at any given rpm, you need some conflicting changes...heavier wire, which means less turns, so you need a larger physical alternator to pack in more turns to get back the voltage the fewer turns left you with;...etc. BTW.  The only conductor that is better than copper, that could conceivably be considered, is silver, a costly metal.   The aftermarket Omega alternator gets its higher output by physical changes in the alternator part sizes, and making them fit the existing timing chest casting.

So far I have mentioned amperes, volts, & ohms, and a brief mention of WATTS.  When current (amperes) is flowing through a resistance (ohms), due to being forced through the conductor by pressure (voltage), HEAT is produced. In some cases the heat is desirable or necessary, like in an incandescent lamp (or, is used to open a fuse if the current is excessive). In other cases the heat is not desirable. Semiconductor 'things' like diodes and transistors, do NOT like heat, generally.  They particularly do not like excessive heat, and also do not like to be cycled, cold/hot/cold....this cycling tends to bring about failures from molecular-sized faults in the manufacturing process. Microscopic cracks, if you will.  In many types of electronics equipment, excessive heat causes the circuitry to fail, sometimes in intermittent ways. This happens to the ignition module under the tank if it overheats due to lack of regular replacement of the heat conducting paste. USUALLY in THAT instance, fresh paste fixes things ..without replacing the $$ module, but letting it overheat a large number of times may well lead to permanent failure.  

Heating and cooling cycles can be responsible for diode board failures. The heating is caused by the engine heat itself, as well as the current flow through the diodes.   Diodes have an internal resistance, and the current flowing through them adds more heat.  That was a problem on some early Wehrle-brand diode boards, as they did not have the large diodes outer wires bent-over before soldering, so the soldered area got too hot, and the solder melted.  That was in the early 1980's.  The Bosch boards were always OK.

When current flows, heat, or maybe work is done...whichever way YOU want to think about it. Work being done is called WATTS. It just so happens that there are some very specific, dyed-in-the-wool relationships between amperes, volts, ohms, watts....two of which are repeated here...and one is added.

volts multiplied by amperes equals watts. 
If you divide volts by amperes, you get ohms.
746 watts is one horsepower. 
 

A thousand watts is called a KILOWATT, often abbreviated as Kw.  You may see, at times, engine output rated in Kw.   NOW you know how to convert Kw to horsepower!   Whilst there are some differences between horsepower measurement standards in various Countries, the relationship between watts, voltage, current and horsepower is fixed.

If the flow of electricity is restricted by such as a too thin wire (like maybe some broken strands!), badly corroded connections, sulfated battery...etc....then we can say that there is 'excessive resistance'.  That literally means that there is excessive resistance to current flow.

Voltage is typically measured by allowing a small amount of current to be diverted from a circuit being tested, and applying that diversion to some sort of meter, in such a way as to have a calibrated reading.  Depending on the circuit under test, most meters divert so little current that the voltage is not noticeably changed by attaching the meter.  This is usually true for situations where the source being measured is of low internal resistance, such as almost every area of a vehicle.  This is NOT quite so true of the electronic ignition, which has some areas you should not even try to measure with a meter (Hall device, as example).   The older meters with actual meter movements...a needle physically moving....usually draw far more current, but it is still NOT a problem with almost all areas of vehicles.

Resistance in ohms, or kilo-ohms (thousands of ohms) or meg-ohms (millions of ohms) is typically measured by applying a small voltage to the part under test by internal meter circuitry in such a way that the current flow is indicated on the meter, but the meter is marked or displays for the effective resistance in the circuit. That is why ohmmeters contain at least one battery, to produce that small current flow through the part under test. Some meters contain another battery for higher resistance ranges, and possibly a battery for powering the digital display, if it is that type. Some devices, such as diodes, are often quick-tested by means of an ohmmeter.  NEVER EVER connect a meter, on the resistance function(s) to a circuit that already is powered.  You will very likely damage the ohmmeter extensively.  At the least you will blow an internal meter fuse.

Common types of simple diodes (which are one-way devices as far as electron flows are concerned!) must pass current in one direction, and not in the other (or, very very little). If the ohmmeter does not apply enough voltage and also current to the diode being tested, the diode may well not 'turn on' in the so-called 'forward direction'. This DOES happen on some [usually expensive] digital meters THAT ON PURPOSE use super-low currents to avoid damaging extra-sensitive devices that might be connected to the meter. Do not purchase a meter unless it tests diodes adequately. The readings on a meter that do NOT turn on diodes properly might be so weird as to be unusable.     It is common practice to use an ohmmeter to check diodes for forward and backwards resistance.  Many digital meters will have a specific diode-testing function.  In that function, the meter, proper polarity connected, reads the voltage drop the diode exhibits with the small meter-supplied current flowing.   This may be in the neighborhood of half a volt.  This test is useful to those with a fair amount of electrical/electronics knowledge....but what YOU are likely to be more concerned with is the forward to backwards resistance ratio.  That is, the forward resistance (the conducting direction) of a diode should show a very low resistance.  The exact value depends on your meter design and the type of diode, and it might be 30 ohms or it might be a couple hundred.   In the reverse direction the resistance on the meter should be VERY much higher....in the millions of ohms is typical.     If anything else is connected to the diode, the readings might be faulty, and it may take some experience and knowledge to know exactly what you should expect.

The applied voltage to the diode must be at least half a volt for most common diodes to 'turn on' in this 'forward' direction. Some types of diodes are specifically made for some 'strange' functions. A Zener diode is used in your electronics type voltage regulator, and some tachometers, to regulate a voltage to some set value...or provide a reference for that type of function. There are diodes used in your CD or DVD player, called laser diodes. Some forms of those laser diodes are specifically manufactured to be indicators. These types both emit a beam of  light. Laser diodes are used for all sorts of things, including vehicle tail lights, backlighting on TV and computer screens, etc.  Besides the small and and also quite large diodes in your Airhead's diode board, you may find, depending on year and model, other diodes in your motorcycle...in the headlight relay, starter relay, connection board in the headlight shell, and in the wiring harness near the coils if a R45 or R65.  

Diodes, in the forward, turned-on direction, can be thought of as having an inherent internal resistance; which causes a voltage drop as current passes through the diode. With enough current flowing, diodes develop a lot of heat.  That is seen from the formulas you learned about earlier in this article. The forward voltage drop of a common silicon power diode is fixed by atomic properties at roughly 0.5 to 0.6 volt.   Therefore, at 10 amperes, there is about 5 or 6 watts of heat to somehow be gotten rid of.  There are 6 of those larger power diodes in your diode board. Thus a goodly amount of heat must be cast off, which is done by the L metal ends of the diode board, to which the power diodes are pressed into.  Some is also absorbed by the passing air.  However, the hot engine also radiates to the diodes.    Later model Airheads with the rectangular air-box don't have so much air flowing, unfortunately.   The RUBBER mounted diode boards, that were used on some models of Airheads, can NOT throw off the heat quite as well to the timing chest metal, it being already hot from the engine being run.  This is just ONE of the reasons I HIGHLY recommend that the rubber mounts be changed to aftermarket metal ones from Motorrad Elektrik, http://www.motoelekt.com  or Thunderchild, http://www.thunderchild-design.com
The other reason to get rid of the rubber mounts is that the necessary extra grounding wires are then not absolutely needed, and with solid mounts the alternator almost always operates better as far as output power and slightly better on voltage regulation too.  You won't have rubber mounts failing in the future either.
  There is NO! negative to using solid mounts.....except for the hassle of installing them; and maybe $10 to $20 to purchase a set of 4.

Although your motorcycle may have a lamp marked GEN, it is really an alternator lamp indicator. Generator, the name, has been used for a very long time as sort of a generic term for a source of energy, typically means from a mechanical-electrical source, and not a storage battery.   A REAL Generator typically uses carbon brushes and an armature windings with a commutator.    Note that the Alternator in our bikes GENERATES electricity.  So, it is perfectly OK for the alternator lamp to be a GEN lamp as marked.

A little aside story:   When the world was first being electrified by Edison (for street lamps, home lamps and industrial motors), current flowed in one direction, this current was called DC, Direct Current.   This was very limiting, as when you had enough homes and factories using electricity, the wires must get larger and larger, as more and more current must pass in total, and more homes are connected to a pair of wires from the generating plant...and soon the wires are very unwieldy. It is almost impossible...or totally outrageous in cost.... to move lots of electricity if it is DC ...for long distances. That is where Edison personally failed, from stubbornness, insisting on DC.  As a matter of fact, Edison had his ego on the line so strongly in this area, that he lied about the dangers of A.C., and was quite a nasty guy in some respects regarding AC versus DC.  Edison lost, as we all know, since our homes, factories, etc., are all run on A.C.     The major exceptions are in vehicles.    Note that moving A.C. over long distances would have the same problems if the voltage was low, but for A.C., we can TRANSFORM voltage/current.

SO.....the electricity in your home is AC (alternating current). What this means is that over a portion of time, the power at the wall socket is such that its VOLTAGE is constantly varying, going up and down, and in fact becomes ZERO as it follows a CURVE that mathematically is called a SINE WAVE. When this 'WAVEFORM' goes from zero to maximum positive, back down through  zero and back to maximum negative, and then back to zero, that is called 'ONE CYCLE'.  Of course, ONE CYCLE could mean starting at ANY place on that sine curve, and advancing in TIME until it reaches the same place on the sine curve that it started from (later in time, of course).  Conventionally one just thinks of it starting and ending at the zero points.   Cycles per second (cps) gave way many years ago to the term HERTZ (Hz), to honor a Mr. Hertz who was a famous scientist involved in magnetic fields.  In your home, the number of Hertz (cycles per second), is 60. This value is kept very accurately by your power company...so accurately that your electro-mechanical and some other clocks, run correctly.   In many things like some TV sets, it is critical that the 60 Hz be proper.   In some areas of the world 50 Hz is used.  For technical reasons dealing with magnetic fields, 50 Hz devices will usually be heavier.   There are actually places that power is sent through wiring that is of higher frequency.  400 Hz is common in large aircraft.

Alternating current has a HUGE advantage over Direct Current, it can be EASILY transformed.    There is a very widespread use of an electrical item called a transformer.  A transformer is a specially designed magnetic steel structure, with some turns of wire on it called a coil, and another such 'COIL' of more (or less) turns of wire, the two generally being electrically separated (that means insulated from each other) but magnetically coupled.  This 'transformer' can very efficiently can change an A.C. voltage to a lower or higher voltage...and there are NO moving parts to wear out.   Transformers CAN be built in which there is only one winding, with multiple taps, but that usage offers NO isolation between input and output.  That would be VERY dangerous for homes, so is NEVER used.

Since you have learned that POWER (watts) is voltage times amperes, this means that we can TRANSFORM the electrical energy output of a power plant to a super high voltage, many thousands (in fact up to half a million and more is in use), and send that power someplace ...which is obviously at a much lower CURRENT (amperes). Remember, the current carrying capacity of a wire is a primary function of the wire physical size (cross-section actually).

Thus, for a given WATTAGE of power plant (major power plants are typically in the millions), we can use THINNER wire to send the SAME power plant output hundreds if not many thousands of miles... if the VOLTAGE is high enough. This thinner wire might still be very thick for large power plants, but it can carry a lot of power at half a million volts.    That high voltage can be TRANSFORMED downwards...usually in steps...first at a local power distribution center...and then dropped farther in your neighborhood by a transformer on a power pole...until it enters your home at 115 or 230 volts, or both.   Whether or not your house is said to have 110, or 115, or 120 volts, theactual value is about the same, or, as we say, a NOMINAL value (probably around 118).   It is common in the USA today to use a nominal 120, 240, 440, and 660 volts, the first two for homes, and the last two for big machines used in industry.

A special form of this transformation idea is actually done in your alternator, magnetically and mechanically, from the induced field from the rotor, and some other effects in the stator...but this is far too complicated to explain in this article.     A simple explanation would be that the rotating rotor transforms by moving the magnetic field induced in the stator.

The only other place electrical-magnetic transformation is done, fairly directly in your Airhead, is in the ignition coil, which by trickery, has a DC voltage applied that is made to ultimately act like a form of AC. The DC from the battery is applied to a moderately low number of PRIMARY winding turns of reasonably large copper wire.  The important thing is the low number of turns and the wire large enough so that the proper current will pass through it. The current in those turns, from applied battery voltage, produces a large magnetic field, very quickly after being applied.  The SECONDARY winding has many thousands of turns of much thinner wire, so it will fit into the coil enclosure. It is a property of transformers that a one turn primary and a 1000 turn secondary is a multiplication of 1:1000 in voltage step up (and a corresponding DROP in CURRENT). If the secondary voltage is high enough, it can break down the resistance of human skin, and pass into the body, dangerously in some cases. It is hard to give cut and dried values, but generally you will not get an electrical shock if your skin, even if wet, comes in contact with a voltage under perhaps 30 volts.

BUT....our ignition coil(s) output will be MANY thousands of volts, and 40,000+ is NOT unheard of and quite possible. So you have the primary winding of your ignition coil (the one with the spade lugs) having a relatively small number of turns of a wire that is relatively thick, and the CURRENT is relatively high (perhaps 4 to 8 amperes).   The secondary then must have wire that is much thinner in order to fit in the case, since so many thousands of turns are needed. Since VERY high voltages are being developed, insulation must be quite good. The voltage coming out the high voltage terminal(s) of your ignition coil(s) will do all sorts of bad things if the wire insulation is not good, if the spark plug cap is not in good condition, if the 'tower(s)' of the ignition coil(s) are not in good condition....and you can RUIN the coil by having the secondary circuit OPEN (NEVER remove a spark plug cap! unless you KNOW what you are doing).
 

When the PRIMARY current is interrupted (assuming enough battery connection time to charge the magnetic field) by opening that battery circuit (points or module), the magnetic field collapses, and the trickery continues, and "induces" a very high voltage in the secondary winding of many thousands of turns.   The TIME for the voltage to rise to the jump-the-spark-plug-gap can be very short.  The shorter, generally, the better.

Once the coil secondary winding (that is the term used to mean the output winding) voltage rises to the point that it will jump the spark plug gap (with the spark plug seeing compressed mixture under pressure), the spark begins and the voltage output of the coil starts to decrease very rapidly.  The spark itself has a very short over-all duration.  While I personally think that the resistance cap (do not use resistance plugs in an Airhead) has an effect of somewhat LENGTHENING the TIME that the spark exists, helping ignition (and reducing radiated radio energy), this is not universally believed by all "guru's".  In any event, whether I am correct, or not, adding series resistance makes things worse if the resistance is large enough.   But, to ME,  there is an optimum value of resistance, that includes the resistance of the Secondary coil winding, and spark plug cap resistor.   It is not critical, but needs to be within certain parameters.

 IF the spark plug cap was off the spark plug, just laying there, the coil voltage might just rise high enough to cause an insulation breakdown inside the coil...or someplace else....and in the electronics ignition models (1981+), you could damage expensive items besides the coil(s).   

Sparks jump easiest in low gas pressures.  The easiest jumping would be in outer space.   The next easiest, for our illustration purposes here, is for a spark plug cap to be off the spark plug and dangerously just lying wherever it might be.  In our Airheads, one cylinder fires at a time.  There is compression pressure of the air-fuel mixture in the cylinder about to be fired (gases ignited) by the spark plug. It takes LOTS more VOLTAGE to jump the spark plug gap if the firing spark plug is the cylinder under compression....than if the spark plug was NOT under compression pressure.  Hence the ignition must be capable of producing a lot of voltage to overcome the spark plug firing gap.    Some airheads use TWO coils, one for each cylinder; and some airheads use ONE coil, with two outputs.  In THOSE with ONE coil with two outputs, the output must jump TWO spark plug gaps....that is, the voltage/current from one end of the single coil goes to one spark plug, jumps that plug, returns to the engine case, travels to the other spark plug, jumps that gap, and returns to the other tower of the coil....remember, you must have a COMPLETE circuit for current to flow.      Note that the cylinder NOT being 'fired' will be much easier to have the spark jump across the spark plug gap, since there is no appreciable gas pressure...and in fact, may be a slight vacuum.  STILL, a single coil may be better performing.

In your Airhead, the primary source of electricity is the battery, which has an INTERNAL RESISTANCE which is very low, a very small teensy fraction of an ohm. This low resistance is why dangerous currents (like melting things type of currents) can flow with short circuits at the battery, or in too low a resistance in circuits connected to the battery. Your battery stores energy NOT as electricity, but as CHEMICAL energy...or potential chemical energy.   Upon a circuit being connected and completed to the battery, the chemical relationship changes in a way that produces electricity.  The type of parts inside the battery determine the nominal voltage of the battery.  Lead-acid batteries have a nominal industry-speak rated voltage of ~2.2 volts per CELL at rest.  That is not the voltage you will be dealing with, necessarily, consider that 2.2 volts as a sort-of value for conversation.   You have, however, SIX cells in your battery, hence 2.2 x 6 is a 'nominal' 13.2 volts.  In practical terms, the battery voltage, fully charged, but the engine off, and no substantial load on the battery, will be AROUND 12.6 volts, after some minutes of resting.

When you are riding down the road, the alternator keeps the battery fully charged (hopefully!), and the battery voltage will be in region of about 13.6-14.5 volts.  About 14.2 is a good optimal value for most batteries.   I won't get into just why, as I do that in a battery article. 

As soon as the alternator output is below that needed to 'float' the battery voltage at the voltage regulator set value, the battery voltage will drop.  If the battery was 100% charged, the terminal voltage will be in that range mentioned, and if the engine is then shut off, the battery voltage will decrease rapidly to under 13, then fall very slowly, until it stabilizes at about 12.6....and will remain there, until, over time and any drain, it is slowly discharged.  The voltage drops intially very slowly during this discharge.  For practical purposes, a battery that measures....in its well-rested state.....below 12...has a rather low charge.  There are published 'tables' of battery charge level for all types of batteries, for voltage and temperature.    If the battery will seem to charge properly (13.7 or more is obtained), and then the voltage drops under about 10.5 during engine cranking....then the battery has little life left (assuming the starter is not excessively drawing current).   More information on voltages later in this article.

"Flooded" batteries are the type where you can see liquid sloshing around, so I call them SLOSH (or flooded) batteries.  These are the types of batteries where you must add distilled or purified water to ...occasionally.  In quite hot weather, this type of battery can self-discharge as much as 1/3 every month, unless recharged. If not recharged at least nearly fully during a ride, this type of battery tends to fail faster due to somewhat IRreversible chemical effects.  NOTE, however, that this type of battery is typically longer-lasting, than most sealed batteries (non-slosh).

A battery fails chemically as well as failing if INternal connections break or partially break. Once a battery fails chemically, it MAY be impossible to recharge it very much at all.  Failure of any one or more cells can cause a type of failure that is sometimes hard for amateurs to determine.   There are many other types of batteries, one interesting type is called Valve Regulated, typified by the Panasonic brand version.   I prefer the original, more properly descriptive name, Absorbed Mat.      As a general rule you should automatically replace your Panasonic battery every 3 years, and your flooded battery at 4 or 5 years.   This schedule assumes you take care of the battery, and it is being charged properly.   A vast number of Airhead owners try to get every last usable day from their batteries, and may brag about it.  Stories are legion with owners getting 7 to 9 years of service.   That can be penny-wise, pound foolish.  As the battery ages without catastrophic problems, it requires more and more alternator power to maintain it at a reasonable charge.  That becomes harder and harder on the alternator.  If the battery is barely usable, the engine may be harder to start, that is, it may require longer (due to slower) cranking, particularly when it is cold.  All this is harder on lots of things, battery, charging system, starter motor, etc.....all causing excessive wear.   Consider also if you are willing to have a battery 'suddenly' fail on you, if you are a long way from where you can obtain a replacement....and, perhaps it is a cold rainy night.....!   Any battery can have a catastrophic failure of course, even if nearly new.  I treat sudden battery failures elsewhere's.   No matter what I say here, there will ALWAYS be PLENTY of vehicle owners priding themselves on getting every last bit of usage from their batteries.

The Airhead Alternator...in more depth:
First, a VERY nerdy, hardly known bit of information:  Earliest alternator rotors were 73.4 mm in diameter.  From approximately early 1975, they are generally mostly 73.0 mm in diameter. I prefer the 73.4, rewound for low ohms.  They will perform better (slightly more power  at lower rpm).

Earliest /5 type alternators were 180 watts, had 105 mm diameter stators where they fit into the engine case, and had Bosch part numbers on the outer housings that ended in -001 or -002.  The R90S only had a -003 stator, it had a slightly larger inside diameter...with slightly reduced output.  The later stators, such as -004 and -005, are all 107 mm, and do NOT fit the /5.  SOME EARLY /6 also had a 105 mm stator, and thus a /5 180 watt system is easily changed to a 280 watt system, by simply changing the stator and the diode board.     If you want the most powerful airhead stock alternator, use any version besides the -002, and use a late or rewound rotor of approximately 3.4 or 2.8 ohms (either works fine)....and use a 73.4 mm rotor, if you can find one....and have it rewound for the thicker wire and less ohms.  Use an electronic regulator.

BMW elected to use in the Airheads a type of generator called an alternator. The name means that its output is alternating current...that sine wave thing again.  The output frequency (number of CYCLES per second, or Hz) varies with engine speed.  

Another very nerdy bit of information: The frequency output of an alternator is a function of the number of pole-pairs, and the rpm.   The formula is:
F = P x N       Where P is the number of pole pairs; F is frequency in Hertz; N is rpm.
         60

When the rpm/frequency is high enough, and your attached radio is not filtered well, some of the alternator noise may show up on your radio as a whine that rises and falls with rpm (and may get louder as you load the system more), as the alternating current is not totally and perfectly converted to DC by your diode board. In fact, due to inefficiencies in the diode's actions, the diodes themselves can create some types of radio noise, that can be difficult to filter out.   There are other sources of radio type electrical noises in your airhead....switched contacts, relays, mechanical voltage regulator, and especially the ignition system.  The contact noise is often heard in a radio as clicking, the ignition as static varying with rpm, and the alternator by a whining that varies with rpm.  All can be filtered out, with some effort.


In order to charge the battery, the AC must be RECTIFIED, that is, converted to DC (Direct current), which is done PRIMARILY by the six large diodes in the diode board [/6 and later have some additional small diodes connecting to a center tap of the stator winding, and have SLIGHTLY improved output, and SLIGHTLY smoother waveform, due to them]. One diode section of three large diodes, the LOWER set, allows only the positive half of the AC to go to the battery positive post, and the other diode section of three large diodes, the TOP set, allows only the negative half of the AC to go to the battery negative post, via the engine structure metal.   Because of this, the engine structure must be electrically solid...of very low resistance due to the large currents that will flow.   Remember what I said much earlier about painted case surfaces?

The two rows of 3 large diodes are NOT the same part, although they look identical.   One set of three diodes is internally reversed in direction of current flow from the other set.  These 6 large press-fitted diodes used on the boards are identified by numbers: 1N3659 and 1N3659R.   The R means reversed from normal polarity inside the diode structure.   YOUR diodes may not have these generic numbers printed on them.   All six are PRESS-fitted to aluminum heat sink material.   

The alternator does not produce just one sine wave output; but, for efficiency, is designed to have THREE....'THREE PHASES'. As you learned well above in this article, one speaks about single and multiple cycles of waveform.  Let us call one cycle as being 360 degrees on a linear time chart.    The alternator produces, via its three phases, overlapping outputs.  Each phase is 120 degrees different (very simplified here).   Those waveforms are constantly rising and falling in sine-wave form. With overlapping waveforms at 120°, there is more constantly an actual output of good value...that is, much LESS time is at lower sine-wave levels, then these three waveforms produce more power, than if there was only ONE phase....and, the three phases are rectified by the three positive and three negative large pressed-in-place diodes.

    (aside note: If you have one of the EnDuralast alternator conversions, that is a ONE phase alternator.  It would be more efficient if it was 3 phases, but that would complicate the system they use)

Let me state this 3 phase idea a bit differently:  If you were to draw these three phase waveforms on a piece of paper, and eliminate the lines below the crossovers, you would see a positive-going part of a sine wave, and a lower-going part of a sine wave, each with three peaks.  The vast AREA between them is increased.     If you think about the AREA of the waveform, you can easily see that more OVER-ALL power, per unit of TIME, is available with THREE phases.....compared to ONE phase.

Besides the /6 and later stators and diode boards center tap connections and associated small diodes mentioned previously that increase the stator output a bit, ALL the diode boards have yet another set of 3 diodes, again these are small ones, which do exactly the same thing as the three very large positive-going ones.  However, these particular small three diodes are used to provide a relatively small amount of current, which has special functions:  (1)  for driving the voltage regulator's  'sensing'  function; and, (2) to drive the several amperes needed for the rotor; and, (3) to extinguish the GEN lamp after the alternator spins up fast enough to need more rotor current than that provided through the GEN lamp. 

If a large diode in the diode board shorts [short circuits], it allows the AC waveform applied to it to pass through it, causing a huge current flow, and perhaps charring/burning, and perhaps a gross failure. If, instead of shorting, that diode OPENS, you will lose somewhat more than just 1/6th of the alternator output, due to complex interaction of the waveforms, diodes, and magnetic fields. This type of failure is hard to diagnose with the diode board still in the motorcycle, and still connected.  Typically, the system charges nicely...until the headlight is turned on.     It is possible for a device called an oscilloscope to make a definite determination, but few own those instruments.  SO.... symptoms might be a battery that MIGHT seem to fully charge up, with correct voltages....yet, when enough load is put on the bike's system...such as the headlight, or heated clothing...etc....the voltage will not come up nearly far enough. Since other faults can mimic this one, it takes some sleuthing. It is a relatively rare event, but does happen.   USUALLY when it does happen, it is not a bad diode, but a bad diode solder joint!...this can be seen visually.

Diodes boards are best checked when OUT of the motorcycle.   Using just an ohmmeter will give reasonable results, for forward and reverse diode readings (BE SURE TO DISCONNECT THE BATTERY if in the bike!), but the best test is using an AC (Alternating Current voltage source) with a lamp...and the board must be OUT of the bike.   See Oak's June 1999 article on Diode Boards, in Airmail, which deals extensively with the diode board OUT of the bike.   On a practical basis, since diodes can act funny if hot or cold, a really anal person would check them at room temperature, and then repeat all measurements around boiling water temperature, and at freezer temperature. I am NOT quite that anal, but admit to doing this a time or two when faced with a seemingly intractable problem.


Specific diode and other problems in Airheads, OTHER THAN in the diode board:

1. If a single diode in the headlamp relay shorts, the motorcycle engine will not turn off with the key switch, only by stalling the engine or by disconnecting a battery wire. The process repeats after the next start.  Later headlight relays may contain TWO diodes.  The function of the diode inside this relay, at pin 85, which is in series with the coil, is NOT well understood by me, I've only seen one, it was 1.244.411, and I have proposed that it might be in case the starter locks up.   The original diode, still there, is between pins 86 and 87b.   The pin 86-87b diode's purpose is to leave the tail and dash lamps on during starting.  Probably a German requirement.  In the monolever bikes, from 1987, if the lights come on when the starter is used, or PARK is selected, you have a bad diode in the lighting relay (one of those two in there).  NOTE that the headlight relay pin 85, a black wire, the grounding end of the headlight relay COIL, returns to ground via the starter motor hot terminal.  This 'clever' arrangement means that when the starter is energized, the headlight relay has +12 on BOTH sides of its coil, and the relay DEenergizes, turning OFF the headlight, but the mentioned relay's diode keeps the tail and instruments lit.   

ALSO note that BMW has used TWO arrangements for the wiring to the 3 position left bar headlight switch.  Most schematic diagrams show the green wire going to the key switch, but some have it to the hot always side, some do not.  Thus, on some bikes the momentary switch works without the key.  

2. On models /6 and /7 and up through 1984 (for the challenged folks, this means 1974 through 1984), if your neutral light has weird things going on, such as being OUT in neutral, ON in gear, unless clutch lever pulled in, and maybe ON at every clutch usage....or some one or more of these indications, you have a bad diode that is located on the BACKside of the board inside the headlight shell. This applies to all twin shock models, EXCEPT the R45 and R65.  In those two models the diode is plugged into the wiring below the starter relay (on pre-1981 R45/65, .....and inside the starter relay in post 1981 R45/65 models).  NOTE that on MONOLEVER models the diode is also inside the starter relay.   Some have not understood these differences, have installed the wrong relay type.

 If your bike will crank the starter ONLY if you also pull in the clutch bar lever, you may have a bad starter relay diode. Did you notice here that the starter relay MIGHT have a diode? So, what do you think may happen if you don't use the proper relay?? YEP.

4. There are some peculiarities in the BMW system here and there.   One is that from 1977 the starter relay has two RED wires that are essentially jumpered via a link inside the relay.  Corrosion here can cause the entire lights, etc., system to act weird or act disconnected.  This is a relatively COMMON PROBLEM.   Just exactly why BMW did that particular wiring arrangement is open to conjecture, but I solve is permanently by NEATLY joining the wires, but leaving them connected to the relay!!

Batteries:
A LOT more can be said about batteries.  Here are a few somewhat useful bits of information:

Voltage regulator settings:
The voltage regulator should not be checked unless the battery has first been charged. Voltage regulator settings are BEST checked with a thermometer on the voltage regulator. However, what I do is to simply start the bike after sitting all day or night at a roughly known air temperature, and then I rev the bike up within two or three minutes, and measure the voltage at the battery terminals, with a known accurate digital meter, BEFORE the regulator can be heated by engine heat. With the battery previously being fully charged, it takes only a minute or so at 3000 to 4000 rpm for the battery to recharge from starting and reach its voltage regulator setting. Temperatures below are VOLTAGE REGULATOR temperatures. It is FAR better to have the voltage regulator and the battery both at about the same temperature, which is why the testing should be done from a fully cooled-off engine within minutes of starting.   Values below are for flooded batteries.

47°F  13.8-14.4 volts 
68°F  OPTIMUM setting for MAREG batteries at this temp. is 14.1 volts.
70°F  13.7-14.3 volts
93°F  13.6-14.2 volts
117°F  13.5-14.1 volts
140°F  13.4-14.0 volts
163°F  13.3-13.9 volts

NOTE: voltage regulators are internally temperature compensated...and you can expect your fairing, or other voltmeter to DEcrease in reading as the engine warms up and radiates heat to the voltage regulator. NOTE that the voltage you are interested in is at the battery, not at some other place on your bike. NOTE that if connections, especially to the voltage regulator, alternator, diode board, and battery, are not good, clean, solid, the readings and performance will likely suffer.  Note also that the stock fairing or dash area voltmeters will normally read a bit lower than the battery terminal voltage, perhaps 0.3 volt lower.  In excess of 0.5 lower means that you should check that voltmeter for accuracy, and if it is accurate, start looking for poor connections, etc....as you will have charging and other electrical problems soon enough.

Charging: 

There are a few things to know about charging a battery.    Initially, on a very weak battery (low charge), you especially want to simply limit the current flow.  Typically and usually recommended maximum is a rate of charge equal to 10% of the battery capacity in ampere-hours. That is, a 28 ampere-hour battery should normally not be charged at a rate over 2.8 amperes. On a practical bases, about twice that value is usually acceptable for the short term, just do not allow the battery to get over a slightly warmish feeling. After the battery voltage comes up to near 14 volts, which charges it fully....then....the battery can be 'floated' at a much lower level, to keep it fully charged, and the float charger can be left on indefinitely if the voltage is 12.8 to 13.2 (at nominal 77°F). This is what 'smart chargers' do, although some are smarter than others in HOW they do it, and the exact voltage they are set for.

Slosh batteries, often called FLOODED or conventional lead acid batteries, can endure a sustained charge of their rated ampere-hours, divided by 18.   They will, however, need the water replenished more often...and this type of 'floating charging' is NOT recommended by ME.

Never allow the voltage to exceed 15.5.  Some batteries can handle this, short-term, others can NOT. In fact, I advise against going over 15.0.

Hydrometer readings on slosh (flooded) batteries, corrected for temperature, are fairly accurate, but some battery faults are such that readings are NOT overly useful. Still, the test is useful at times.   Lower liquid capacity hydrometers for small batteries are available cheaply.

If a battery has been charged fully, then let sit and stabilize over time, the battery voltage will very slowly drop, after a much larger initial drop from fresh charging. The following information assumes 77°F, and that the battery has sat for at least a few hours:

Fully discharged: 11.89 volts or less. NOTE that SOME books will say that this is 10.5 volts, SOME will say that a 10% charge is left at 11.31 volts; and 20% at 11.58 volts, 30% at 11.75 volts. These variances are due to the type of battery construction, and method of measuring and the amount of capacity left. 

The following are generally accepted 'official' values:

100% of charge at 12.7 volts and 77°F [UNofficially, your battery is PROBABLY going to read about 12.6 volts for fully charged, at around 65°F, after it sits for some hours].  

The below values are for 77°F:
75% of charge at 12.5 volts
50% of charge at 12.27 volts
25% of charge at 12.06 volts

Absorbed Mat (Valve Regulated) (Panasonic and other similar types) batteries need somewhat higher charging voltages....and I like to see the voltage regulators set for about 14.3 or even a tad more, at nominal 'room temperature', at the VR.
 

Typical electrical usage in the STOCK airhead:
Headlight 55 or 60 watts; ignition 40 watts; miscl small lamps, etc about 15 watts; charging after battery fully charged...about 10 watts.
 

Starting:  

Problems with starting can often be traced to a starter in need of replacement or overhaul or lubrication of the Bendix drive, a bad starter solenoid, and occasionally a problem in the starter relay circuitry.     Some stock relays have a diode inside.  Unconfirmed information is that some have substituted a 0-332-014-118 relay, perhaps a DF005 'Blazer' relay from AutoZone stores.  DO keep in mind what I have posted earlier in this long article about diodes inside of starter relays!    The Bosch starter relay uses two #87 terminals, and may sub to Bosch 03 32 019 150 for 1977+ bikes.  That is a common Bosch accessory use relay.  Connector, if you need one, is 0 334 485 007, while the spring loaded  terminals are 1 901 355 917.

Bosch starters up through 1974 were 8 tooth 0.001.157.007, rated 0.5 hp and 290 A.  The /6 bikes for 1975 and most of 1976 used an 8 tooth 0.001.157.015 rated 0.6 hp and 320 A.   The 8 tooth starters are used ONLY with the 93 tooth flywheels. 

For 1977 and later, the starter has to be 9 tooth, for use with the 111 tooth flywheels (or later clutch carrier).  The starter is 0.001.157.023, rated 0.7 hp and 320 A.    Solenoids, unconfirmed, seem to be the same as EARLY air cooled VW.

Valeo starters can be replaced with Bosch, you also need the forward end bracket that the Bosch used to secure it to the timing chest wall.    There is an $$ aftermarket Denso starter available that is quite high powered.  You can also replace a Bosch with a Valeo.  Be careful when installing ANY alternator!...especially a Valeo in place of a Bosch, or a Denso...you may have to do a small amount of metal filing.  See the boschvaleostarter article.

NOTE:  The "Airheads Beemer Club" has an account with Ace Houston Warehouse, a wholesaler/importer/remanufacturer, ETC.  The Club account is #700.  Call Bob Spencer at 1-800-392-3332  or e-mail to acehoutx@flash.net.  Mention account 700.  The Valeo starters are available.  The part was D6RA15, Valeo changed it to 432586.  The last price for these from Ace was $172.50 plus shipping.  This is a brand NEW starter.  5 or more are cheaper.  There is no core charge, but they will probably pay shipping to get your old one.    They have rebuilt Bosch starters, last price was $200 and a $100 core charge and shipping (core charge refunded).   Bob thought that the Bosch starters MIGHT be put back into production at some point.   The information in THIS paragraph was provided by Ken Kirk   kirkkw@yahoo.com.
NOTE:  See information in the URL section of this website for other electrical sources!!!...including oversize alternators, parts and service for Bosch and Valeo, ETC!!   In particular, see Euro Motoelectrics;  www.EuroMotoElectrics.com!...they have the correct latest version of starters and parts!...at very GOOD prices!  Highly recommended!  They also produce the EnDuraLast Alternator system.

  See my article on substituting starters!  16-A, Bosch-Valeo Starters.

Addendum:

1.  Eddy Currents.   Eddy-currents are a phenomena, or characteristic, of some forms of magnetics.   In our airheads, they are at several prominent places, including the solenoid on the starter motor, inside the relays, inside the speedometer, and the alternator stator and its rotor.   The ignition coil has eddy-currents too, but has a very special way of dealing with them.

An eddy current can be thought of as an electrical current that can be induced into a magnetic medium itself (steel laminations of the stator), or into a non-magnetic structure, such as the aluminum cup used in the speedometer.     Speaking more as an engineer, an eddy-current acts as if it is a one-turn coil of wire.  This means that laminations of the alternator stator, and the rotor, can have not just magnetic energy in them, but actual electron current in each lamination.  The effect is small, but quite noticeable under certain instrumentation tests.    Just to make this clearer, not only is current flowing in the wires wrapped in the stator and rotor, but the steel laminations have current flowing in them, not just a magnetic field induced into them.

In the Airhead alternator, eddy currents are UNdesirable; as they simply produce heat and detract from output.  Eddy currents are often generated in transformers, and are generated in the steel or iron cores.  To reduce the eddy-currents, which waste power, the cores have thin laminations, up against each other.  Each lamination has its own tiny eddy-current, but the eddy-current effect of many laminations is vastly smaller than if the laminations were replaced by a one piece, solid, construction.  Any deeper discussion is way beyond this article.

To combat eddy-current losses, thin, laminated strips of metal are used in the construction of power transformers, and your stator, rather than making the transformer or stator out of one solid piece of metal. The thin strips are separated by insulation, usually something quite thin, perhaps even just a thin coating of enamel or other substance, which confines the eddy currents to the individual lamination strips. This reduces the total eddy current power losses.  Do not scratch the laminations across the edges, which might cause currents to flow between the laminations in a way that they were not designed for.     This means that you should be very careful when removing and replacing the stator.  Don't scratch the enameled wires either....a scratch across them might enable bare copper wires to touch each other, causing gross inefficiencies.

A bit more Nerdyiness:  In mechanical speedometers (and many tachometers too!) that use eddy current technology, a rotating permanent magnet induces eddy currents into an aluminum cup (could be a plate).  The cup is spring loaded (usually a hairspring).  The faster the rotation, the higher the magnetic forces the aluminum cup experiences, causing it to rotate, and take the attached indicating speed needle with it.   Thus, even though the aluminum itself is seemingly UNaffected by you bringing a magnet near it, and you certainly cannot turn the aluminum part into a permanent magnet, it will be affected by eddy-currents.  Probably your home has an electrical power meter run by eddy-currents.  Alternating Currents flows in a magnetic structure, and the eddy currents induced in the aluminum disc cause the aluminum disc to rotate, and it is geared to dials that show the power provider how much electricity you used every month.   In this particular instance, there is not only eddy-currents induced into the aluminum disc/cup, but the aluminum, whilst not thought of normally as being a magnetic material, 'conducts' the magnetism of the associated copper wound iron structure, across a 'gap' in the structure.   This is a rather simplified and not quite accurate description, but good enough for this article's purposes.   In other words, while you cannot 'magnetize' aluminum to be a permanent magnet, and can't even as a non-permanent magnet, you can induce magnetic effects in the aluminum, of sorts.  Weird, eh?!?!

2.  Troubleshooting hints....using a test lamp.  This is vry slightly edited from comments I made on the Airheads List on June 3, 2011:

Revisions:
02/03/2003:  Add typical electrical usage values
02/04/2003:  add information on starting and starters; headlight circuitry; (2) at top; add to #1 in problems area; simplify some explanations, eliminate some redundancy's
04/03/2003:  editing for clarity
01/05/2004:  update the URL for Chicago Region club.
02/06/2004:  clarifications only
05/26/2004:  update contact information for Chicago Region Club
03/01/2005:  revise including hyperlinks, for the chitech electrics manual, as article 80 is now done
04/12/2005:  Add (3) at very top, add Addendum, and explain eddy currents
08/06/2006:  Revised and updated entire article
02/23/2009:  clarify rotors and stators
11/18/2009:  Revise article for better clarity, and a few changes to keep up with latest tech.
12/03/2009:  Try to clarify more details
09/14/2010:  Clean up article, many places.
12/10/2010:  Clarify some basics
05/20/2011:  More clarifications, nothing at all major
06/03/2011:  Add Addendum item #2

© Copyright, 2012, R. Fleischer
 

Return to Technical Articles List Page

Return to HomePage