I don't get along well with electricalcity issues, so I could use some advice.
The bike is placed on a float charger every night, but starting can be an issue, especially if I ride to work (only about ten miles) and the bike sits for twelve hours. "Work" is right on the water's edge of the Pacific Ocean and the temperature hovers in the mid-forties, low-fifties most nights.
On a daily basis, once started it never fails to restart regardless of the distance between "start/stop"; the "state of tune" is excellent and generally lights off in one or two revolutions of the engine.
With the engine at around 3,500 rpm, my multimeter indicates 12.8 volts with the probes on the + and - terminals of the battery (which is about nine months old).
So, I (think) I need to check the alternator output...
With the multimeter set on A/C, and the expectation that the voltage readings (regardless of what they are) should be close, do I [I]alternately[/I] touch the probes to:
W - U
W - A
A - U
A - W
U - A
U - W
Guidance would be appreciated.
Usually the order of checking is:
- rotor -- what is the resistance across the slip rings
- voltage regulator -- when bypassed, does the charging improve
- diode board -- one of the outputs at the side of the board takes the power, runs it through heavy cables connected to the starter and then finally to the battery
- stator -- usually pretty rock solid, but the brushes need to be checked to ensure they're not getting hung up, resulting in no contact on the slip rings
I'm a little confused... A/C multimeter setting? U-A-W?? Is that the United Auto Workers?
Also: when measuring AC voltage, polarity does not matter - so U to W is identical to W to U, W to A is identical to A to W, and A to U is identical to U to A. Saves half the steps.
12.8 volts is a bit low. I'd be leery of totally bypassing the regulator - that might allow too much voltage to the rest of the system and smoke something (not a suggestion).
The rotor, brushes, and an open (or lifted) diode are the typical maladies.
By a "lifted" diode, I refer to the connection of the diode to the board. Some of these were assembled with little or no fold on the diode's leads where it lays against the board, forcing that branch of current to travel through the solder only. Solder is tin and lead, so that can melt (just like a fuse) or break away from the board.
[QUOTE=Pauls1150;782852]I'd be leery of totally bypassing the regulator - that might allow too much voltage to the rest of the system and smoke something (not a suggestion).[/QUOTE]
Actually this is a standard approach and I think Rick Jones writes about it in his Classic Charging booklet...obviously, don't let the system run on it's own for very long, but if the regulator is bypassed and the voltage climbs to 15v or so, then the alternator proves good. The circuit should be able to take that for a short period of time.
This article seems appropriate for /5 charging. There is even a link to measured charging voltages.
Edit: if one were to read the above article linked measured voltages - the voltage you measure at 3500 rpm is 0.1 volt higher.
It's never anything glaringly obvious, is it? At 12.8 volts that's only 0.5 volts off of the 13.3 observed by buchanan. For some reason I was thinking the voltage should have been closer to 14.something at "speed".
But still, a half volt is a half volt.
Battery disconnected prior to removal of the front cover.
- rotor -- what is the resistance across the slip rings[/QUOTE]
And this is measured how? :whistle
[QUOTE]- voltage regulator -- when bypassed, does the charging improve[/QUOTE]
The voltage regulator can be temporarily by passed by ... taking the next off ramp? Or... ? ;)
[QUOTE]- diode board -- one of the outputs at the side of the board takes the power, runs it through heavy cables connected to the starter and then finally to the battery[/QUOTE]
And I do what here.... check all connections for tightness and corrosion?
- stator -- usually pretty rock solid, but the brushes need to be checked to ensure they're not getting hung up, resulting in no contact on the slip rings[/QUOTE]
This seems like fairly straight forward mechanical procedure...
[QUOTE] ... an open (or lifted) diode are the typical maladies.[/QUOTE]
Requires removal of diode board for inspection ( I assume...); looking for obvious cracked solder joints (?)> Repair by simply re-soldering, or replace entire unit with something from Motoelektrik?
What precautions must I take to prevent any damage to the components on the diode board (beyond initially disconnecting the battery before beginning) ?
Thanks for the support gentlemen. I ordered Paul's book this morning, but of course it won't get here until after the holiday. :(
the slip rings are the copper discs that the brushes bear against (?).
NOT MY ALTERNATOR... REF. ONLY.
Rotor resistance - carefully slip a business card under each of the brushes to isolate them from the slip rings. With the multimeter set to ohms, put one probe on one slip ring and the other probe on the other ring. Resistance is ___________.
Regulator check - [url]http://bmwmotorcycletech.info/testingvoltageregulators.htm[/url]
Diode board check - [url]http://robfrankham.co.uk/bike/testing_the_diode_board.htm[/url] -- one of the pictures shows two spade terminals off the side (T). When the bike is running, there should be voltage here as the output of the diode board...I would expect to see 13+ to 14v at this point.
here's what I've found thus far:
I checked the voltage at the battery as a baseline: 13.3 volts [img]http://www.freesmileys.org/smileys/smiley-confused004.gif[/img]
I jumped the voltage regular and backed off the throttle at 15.3 volts (measured at the battery anodes). :)
I removed the jumper and reconnected the voltage regulator and got 12.7 volts (measured at the battery anodes). [img]http://www.freesmileys.org/smileys/smiley-confused004.gif[/img] [img]http://www.freesmileys.org/smileys/smiley-confused004.gif[/img]
It's reassuring that the alternator seems to be okay and that (by deduction) the diode board is too. But what's up with the variable voltage... ?
Maybe time for a new voltage regulator? Perhaps an adjustable type?
A bit frustrating when things aren't black and white. Can you rig your meter so that you can ride down the road for some time and take glances at the readings? That would be a truer indication of what's going on.
One thing that might be going on are some dirty connections. Find all of the wires and pull them off and do what you can to clean the connectors. The three wires that go from the alternator to the back of the diode board can be corroded where they plug into the back. On my /7, the plug is a molded plug which tends to trap moisture...sure enough, when I pulled off a couple of years ago, plenty of green corrosion. I bought three new individual cables from Rick Jones.
What about all your grounds? I suspect your /5 has metal posts that connect the diode board to the engine case...they came up with the rubber ones later IIRC. Rick sells metal mounts if you need them. Plus there's a harness than can be added which improves the grounding of the diode board.
One thought and one question -
If you are still using your original voltage regulator, which if I remember right is mechanical, how about replacing it with a new upgraded technology one. Take a look at the bottom of the charging page on Motorrad Electrik's site -- $38.
What is the problem? It is hard to start after it has been sitting awhile at cool temps? Hard to start because the starter turns over the engine slowly as though the battery wasn't fully charged?
[QUOTE=Lmo1131;783067]For some reason I was thinking the voltage should have been closer to 14.something at "speed".[/QUOTE]
It should be just above 14V. 14.2V to 14.4V would be ideal.
Have you checked the battery? Have you taken it out of the motorcycle and charged it? Will it charge to 14.4V? Sometimes a battery can be the issue. It should not get hot, not even warm, during charging.
Excellent points Gentlemen.
The bike is continually on a battery tender, but I have not removed it and placed it on a charger per se; I will do that today while I clean all the contacts and terminations. The battery itself is not six months old (which does not necessarily mean that it is not at fault); +/- anodes are clean, electrolyte levels are spot on, and no sulfate buildup is visible on the plates.
The regulator is original to the bike, making it 39 years old.... If time is money, then replacing it outright would probably be a good investment, and $38 wouldn't be an outrageous sum (when spread over the previous 39 years of service of the old one). Being able to bump up the voltage to 14+ would be another plus given that the machine currently sees more "town" runs than open road work (although hopefully that will change this summer (and I could adjust the voltage back down for long trips)).
[QUOTE]What is the problem? It is hard to start after it has been sitting awhile at cool temps? Hard to start because the starter turns over the engine slowly as though the battery wasn't fully charged?
Exactement! That is precisely the issue that started this quest. I had the starter rebuilt last spring; new bushings, armature cleaned up, brushes, lubricated, etc.
I have also modified the infamous starter protection relay to eliminate the "cricket" issue (which helped somewhat).
But there is still a slight hesitation when the starter engages (and has been since the bike was new); and the starter has never really shown any "authority" when turning over the engine.
Under consideration ... the Nippon-Denso starter offered by Motorrad Elektrik. A friend of mine has one in his R90S and it's performance is dramatically better than my old Bosch unit.
Over 15 volts with the regulator bypassed points a strong finger at the regulator as your culprit. The changing voltage is because there's nothing "regulating" it any more - more RPM means higher voltage (up to the alternator's max). If it is the older mechanical regulator, Barron is right - get a newer version. The mechanical ones are just relays that shunt (re-direct) the extra voltage away from the system, and let it bleed to ground instead. I think you've had a couple of old HD's, yes? Same problems. Relay contacts will corrode, burn, and stick together; relay coils can open or short.
Diodes FYI stuff -
Yes the board has to come out to inspect and measure.
In certain model years (and I don't recall which ones), the assemblers neglected to bend over the wire leads from the diodes so that they physically laid against the copper printed-circuit trace prior to soldering. If the solder didn't melt (actually due to "current density", the number of electrons per cubic millimeter), cracking of the solder joint became common.
IF the diode is still good (they really don't like vibration or heat), and you've spotted a cracked solder joint (may take a magnifying glass), you may be able to bend the lead over & re-solder. The caveat here is that you need to use a high-temperature solder, either with less lead and more tin, or more silver.
Overheating the solder joint will make it brittle and likely to break again; and you don't want to let the heat of soldering travel up the diode's lead wire into the diode, since the heat can damage the diode. Use a heat sink - like a needle-nose pliers with a rubber band on the handle, or a forceps - to prevent the soldering heat from traveling up the lead.
If you can use an ohmmeter: use an ANALOG meter (with a moving needle, like a Simpson or a Triplett), not a digital meter, to measure the individual diodes. Most DVMs do not "source" enough voltage to turn the diode On, and will give you odd non-repeatable readings. If a DVM has a specific "Diode" checker function, that'll be OK. BIG CAUTION with the analog ohmmeters, though: DO [B][U]NOT[/U][/B] use the "X1" (Times One) range, [B]nor[/B] the "highest" range (X1000, X1M, depends on the instrument) on any solid-state device: the x1 range will let the meter's full current flow into the part being tested and possibly damage it; the Highest range will let the meter's full battery voltage flow into the part being tested and possibly damage it (some of those meters have 32 volt batteries - not good to apply to your bike).
Diodes are supposed to allow voltage to go in one direction only. In the "Forward" direction - with the negative side of the meter connected to the cathode ([U]striped[/U]) side of the diode - you should measure anything between 30 and 60 ohms with the analog ohmmeter. It's OK if they're not strictly "identical". In the "Reverse" direction, you should measure Very High resistance, over 10K (10,000) ohms. The kicker here is that if the diode is connected to anything else, like the alternator, regulator, or other diodes on the board (darn!), this will totally screw the ability to get a true reading - you have to be able to measure one element at a time.
Or you may get lucky and observe that one or more has been burnt.
[QUOTE=Lmo1131;783310]Being able to bump up the voltage to 14+ would be another plus given that the machine currently sees more "town" runs than open road work (although hopefully that will change this summer (and I could adjust the voltage back down for long trips)).[/QUOTE]
A newer electronic voltage regulator would be a plus on top of it being adjustable.
One of my most useful items on my motorcycle and car is a voltmeter and both are connected directly to the battery for an always-on and true reading of battery voltage. Consider installing one that is accurate and displays battery voltage.
[QUOTE=Lmo1131;783310]Under consideration ... the Nippon-Denso starter offered by Motorrad Elektrik. A friend of mine has one in his R90S and it's performance is dramatically better than my old Bosch unit.[/QUOTE]
Even the Valeo and similar pop-can starters are more efficient. They use a fast spinning pop-can sized motor and a planetary gear set to up the torque.
I would have someone measure the current draw of your original starter to see if it is "in spec". Then make sure every single connection is clean and even more important is to make sure every cable crimp is totally corrosion-free. That is hard to determine visually and is best measured with a 4-wire ohm meter. Think about it, with 100A starter current draw, even a 10 milli-ohm (0.010 ohm) cable resistance is far too high.
[B]To give you an idea of the importance of near-zero resistance connections[/B], I made up a new ground cable for the alternator on a local owner's Porsche 911...I guess they use a ground cable so as not to rely on the metal to metal contact of the alternator/engine castings.
The new (crimped and soldered) and old ground cable...
An Agilent 34420A nano-volt, micro-ohm (4 wire) calibrated meter was used.
The old cable measured at 0.11 ohms.
The new cable measured at 0.000714 ohms.
Specs of the wire used:
Gauge: 8 AWG
Diameter: 0.199 inch
Resistance @ 20??C per 1000 feet: 0.701 Ohms. (looks like that confirms my measurement at 13 inches = 0.000714 Ohms
Standard wall ethylene-tetrafluoroehtylene (ETFE) insulation ( also known as Tefzel ) designed for aerospace applications where weight, dimensional tolerance, and mechanical durability are required. This wire exhibits high chemical and radiation resistance.
If you up the ground cable by one gauge and use welding cable (multi-multi-stranded...and flexible), you'll get a lower voltage drop on that cable.