Apparent Regulator Failures

[Magnum Magna]

That is similar values other then the 3 minute idle but you had it at 3000 and I had it at 900 RPM

[The_Raven]

Howyadoin,

OK, done deal!  I replaced the regulator, and things appear to be in order.  13V at idle, 14V at 3K, no sparks, smoke or other drama...  Nice to be back on two again!

Not an easy job, since removing the rear fender wasn't practical; the mounts for my Leatherlyke bags are tough to get off, so I unbolted the battery box and was able to manipulate it enough to get a socket on it from below for the left nut, and a side angle for the right.  I got out of it for $50 for a used OEM piece off eBay, which is nice considering my local dealer quoted me $205 for the part, and probably another $150-200 in labor!  (Did I mention that I'm a cheap bastid?)  Oh, and a new battery, but she was due for that anyway.

Thanks everybody for the timely responses... tomorrow's our club BBQ and I would have felt like a putz rolling in in the cage!

Now there's still the matter of the GL1200 with the toasted stator, but that's a story for another time (and place)...

Off to polish!

[Charles S Otwell]

Glad to hear your back up and running . The stator problem has turned me off some nice looking Goldwings, I've heard ya got to break the case apart to get to the stator, is that correct? The stator on the Magna is actually easier to get to than the regulator. Anyway, like I said glad ya got er fixed..

[Len Averyt]

I was givin an old regulator at on of the wrench sesions. I have the intention of disecting it to get a full understanding of what is in it and how its asembled.
My "plan" is to develop a hardened version of this unit using more robust parts.
As an Electrical Engineer I have access to and use PC board creating software and can easily make a more robust unit.

[The_Raven]

Glad to hear your back up and running . The stator problem has turned me off some nice looking Goldwings, I've heard ya got to break the case apart to get to the stator, is that correct? The stator on the Magna is actually easier to get to than the regulator. Anyway, like I said glad ya got er fixed..

Basically, you have to pull the engine to replace the stator on the GW.  I'm opting for the "Poorboy" mod, which involves mounting an alternator from a Geo Metro and driving it off the front crank pulley.  It doesn't require engine removal.  I'll be doing that some time next month.  Big fun!

[drkngas]

My "plan" is to develop a hardened version of this unit using more robust parts.

Sign me up for the "hardened version"!  SInce my battery goes dead rather quick after idling with the cooling fan on i'm sure I might be needing one soon.

[Magnum Magna]

IMy "plan" is to develop a hardened version of this unit using more robust parts.

"hardened version"  Will it be protected from EMP?

[lragan]

I was givin an old regulator at on of the wrench sesions. I have the intention of disecting it to get a full understanding of what is in it and how its asembled.
My "plan" is to develop a hardened version of this unit using more robust parts.
As an Electrical Engineer I have access to and use PC board creating software and can easily make a more robust unit.

Len, my specialty is rf circuits and systems, and integrated circuits, so redesigning the regulator looks a little more formidable to me.  However, if I can help in any way, I will be glad to.  The  manuals show some schematics which I can scan and send to you. 

I see two large problems:
1) How do I unpot the unit to see what is really there and maybe identify a single "weak link" so we can proceed with minimal risk redesign?
2) How do I test new designs?  I have envisioned a system which has an alternator somehow geared up to an electric motor so I can run bench tests off the bike where it is easy to instrument.

Your thoughts on these would be appreciated!! 

[Sledge Hammer]

Len, Lawrence,

Maybe you two can answer a question I have. Did Honda do anything other than simply hand us a circuit which impresses loosely regulated voltage across the battery terminals? I spent just long enough with an emergency lighting manufacturer to be dangerous, but I do recall that lead-acid batteries are touchy about how they are charged and will quickly degrade when overcharged even with trickle currents. They also really hate being partially charged, so not sure that running with all your accessories going will help the battery. If Honda is just dropping voltage across some diodes and a resistor/pass transistor, that is a recipe for early battery failure.

I had a long blurb ready about how we designed chargers for lead-acid batteries at an emergency lighting company (the first thing I noticed about the batteries was that they were the same size as typical motorcycle batteries), but if you have already designed lead-acid battery chargers, you already know what you're getting into.

For what it is worth, I designed a lighting-management circuit to automatically handle my bike's auxiliary lights because I didn't want to discharge the battery at a stoplight. I haven't yet physically built the circuit yet (time for a Digi-key order!), but the PSPICE model works quite well. If it works as well in practice as it does on the computer, I'll be happy to pass it along to you so that hopefully it can relieve the rectifier/regulator redesign of having to supply a full load for all accessories at idle.

[lragan]

Gee, guys, I wish I was at home with access to the manuals, but I will have to do this from memory.  What I recall is that there is a six phase rectifier circuit going to the battery terminal.  In parallel with each leg is an SCR that shunts to ground, controlled by a mysterious box called, I think "regulator" -- which I presume to be an IC.  My perception of how it works is that as the output voltage goes up the SCR's fire earlier in each cycle, shunting current to ground.  This will increase the dissipation slightly, by the current X the forward voltage of the conducting SCR for the portion of the cycle that it is on.  I don't know how the "regulator" controller decides when to fire the SCR's, or even if the schematic in the manual is representative of what is actually in the box.  It wouldn't have to be, since their standard repair is to replace it.

Since this regulator was probably designed about 20 years ago, I am confident that better components exist to do this job.  I am thinking of the chips from Benchmarq (now part of TI) that provide battery management, for example.

[Sledge Hammer]

That's more involved than I thought it was, but it sounds very much like the approach which a friend of mine said Kawasaki used on his '77 Kz650, only with actual relays in place of SCRs  on the Kawasaki (don't ask me how it has lasted this long with real moving parts). Knowing the Japanese, it is probably a generic approach used across the Japanese motorycle industry. It sounds like it was something developed by somebody with more background in motor control than in battery charging.

[Len Averyt]

What I recall is that there is a six phase rectifier circuit going to the battery terminal.  In
parallel with each leg is an SCR that shunts to ground, controlled by a mysterious box called, I think
"regulator" -- which I presume to be an IC

Sorry for the long delay in responding, I have been very busy. Anyway the above is very correct.
The hardend version of the regulator circuit I'm thinking of doing is based on voltage regulation
systems we, the company I work for and I, are designing for one of our customers.

http://www.airtractor.com/at-802f

We are developing the second generation control system for the dump doors on this plane. My boss was
the head engineer that designed the first gen unit which is now becoming very "dated"
Anyway I have been doing a lot of research into voltage regulators for use and used in this project.
One particular part is very promising for both applications. Its what is called a buck regulator, it
takes up to 45 Volts of DC and regulates it to 12VDC +/- 1% a adjustable version can be "tuned" to run
at any voltage between 3VDC and 30VDC. We would use this part and set it for around 13.2 to 13.8
Volts. The part I'm using for bench testing is rated up to 3 amps. but the nice thing is that several
of these can be paralleled up to our target amprage.
There are also other components that can be used in the same way.

The six phase rectifier circuit won't really need any changes other than higher capacity units.
I also want to include another separate regulator that can run 12V aux items.
This regulator would get its power directly from the phase rectifiers and would provide power only when the engine is running. this would allow devises like extra lights or similar to be powered without a drain on the battery or the motorcycles existing system. No switch would be needed, however i will put a control line to allow a low current , small switch to enable or disable this regulator effectively allowing it to turn on and off the regulator output. or a jumper can just be installed to run anytime the engine is running.

This extra regulator can easily be built as a stand alone unit that can be implemented right now. it would just tap into the 3 wires coming from the "alternator" to pick up the supply power. The 3 amp part should be good enough for this.

Let me know what you think. Ideas, questions, etcetera.

PS the test part im using is the following
http://www.diodes.com/datasheets/AP1501.pdf

BTW the entire circuit minus the phase rectifiers is 1/2 the size of a postage stamp.

[Magnum Magna]

What will be the effect on the stator?

[Len Averyt]

What will be the effect on the stator?

Since these parts are DC to DC converters and are 90% efficient or better, less voltage will be shunted to ground or lost as heat.
The end result should be less draw on the stator at an equal output with the current "regulator" which i would guess is only about 80% efficient at best.
Of course there are no free lunches, so there is of course a maximum limit that the stator can deliver without becoming cooked
I would try to limit the system to about 80-85% load.
I will need to do some tests to find out how many watts the stator can deliver at around 4K rpm.
This is a fairly straight forward test just to look for when the output current begins to sag at a giving load.
Since the "altenator" is a permenent magnet design and not a feedback type like most, I would guess that the output will not be very linear.
As the rpms go up above a certain point the field frequency (Hz) will become higher than the windings are "tuned" too, less voltage will be induced into them.

[roboto65]

Well the stator throws A/C in the mix will this change anything being D/C to D/c converter