|
New to Ham Radio?
My Profile
Community
Articles
Forums
News
Reviews
Friends Remembered
Speak Out
Strays
Survey Question
Operating
Contesting
DX Cluster Spots
Propagation
Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers
Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net
|
|
1-10 of 13 messages
|
Page 1 of 2
Next
|
|
Regulating a motorcylce generator
|
Reply
|
|
by KC2MMI on October 23, 2005
|
Mail this to a friend!
|
OK, so here's a horse of a different color. My friend has an older Suzuki "touring" bike which apparently has a rep as a good bike with a lousy electrical system. We've been working on improving that, and we've gotten down to what seems like just a terribly clever (or lousy) design problem.
The generator is a real generator, using field magnets not coils. Suzuki runs the three stator leads out of the generator. One through the 55W headlight, and the other two directly, into a potted regulator. The regulator apparently dumps excess power as heat and does a lousy job of charging the battery, with wide voltage fluctuation to boot.
When I say clever or lousy...the generator runs *inside* the oil bath in the engine, apparently using the headlight and the oil bath to dump excess heat (from back EMF) when the regulator is not supplying output to the battery or other loads. That probably seemed ingenious to whoever designed it, maybe it was.
But...we've got a generator that puts out up to 100VDC (3 phase) at full speed, 280 watts (20 amps) max output, that needs to be regulated down to a nice tight 14.4 VDC so he stops chewing up a battery every two years, and gets power for the running lights, etc. a little more reliably.
I've worked with 3-terminal regulators and bootstraps...but I'm wondering, when we're starting with raw power from 20-100VDC, and need to cut that down to 14.4 VDC at 20A...is there some clever way to do this effectively?
Installing a "real" alternator is simply not possible, because the only place to do it is internal in the engine housing. What's a modern, robust, reliable way to do this?
The only constant load is the 55A headlight, that's on all the time and makes a substantial drain on the 14AH rated battery if he's stuck in traffic. (Then he's realyl got to come back into neutral and idle high to keep a charge up.) We've discussed using one or two auxiliary lamps (55-75W rated) if needed, which could be turned on to "dump" extra power at highway speed. That would work...I'm just not sure about how to dump excess power efffectively with those voltage and amperage criteria. And, the need for a small (paperback book, max) size package.
Anyone up for a challenge?<G>
|
| |
|
RE: Regulating a motorcylce generator
|
Reply
|
|
by KC2MMI on October 23, 2005
|
Mail this to a friend!
|
|
I ahould have clarified--that's an AC generator, producing 20-100VAC. The rectifying and regulating both happen internally in the existing "puck". So what I've got is AC that needs to be regulated and rectified...possibly a good opportunity for a PWM approach?
|
| |
|
RE: Regulating a motorcylce generator
|
Reply
|
|
by KB1GMX on October 24, 2005
|
Mail this to a friend!
|
The only way you have available is PWM or switch mode controllers.
However you also have another problem. Total wattage
at the cruse range may be below the load or above it.
If above then regulation is possible. If below consider load shedding.
Years (decades) ago I had a similar problem with a
Honda 160, the solution I used was instead of rectifiers I used SCRs and the general idea was to
not trigger the SCR if the battery voltage was above
14V and trigger the scr if it was below. Fairly
simple and worked well. At that time I was working
with maybe 20A total load. The basic circuit was
from the RCA transistor handbook as a battery charger
that cut back to trickle mode when battery was charged.
Make sure your scr can handle the load currents, voltages and can work at high RPM (high frequency AC).
For 1972ish that was a advanced design.
With modern POWER MOSFETS I'd rectify the alternator
output and filter with a large cap (energy storage)
then use a switch mode controller to regulate the
power(current) to the battery. The bigger problems
are space to put it in, vibration and temperature.
FYI: that puck is really rectifiers and a honking large zener diode..
Allison
|
| |
|
RE: Regulating a motorcylce generator
|
Reply
|
|
by KC2MMI on October 24, 2005
|
Mail this to a friend!
|
Thanks, Allison. I had some suspicions about that what might be in that puck.<G>
<However ...Total wattage ... below the load or above it...consider load shedding.>
Yes. I was thinking that while it was ingenious of Suzuki (who used all untinned wire and plain brass connectors, which all punked out) to just dump power back into heat and then use the engine oil to dump that...that a better way would be to dump extra power into extra running lights at highway speed, making the bike more visible and safer. Which is how this whole thing got started, because my friend wanted some extra lighting and the bike wouldn't carry the load of what he got it with, for a start.
<<I used SCRs and the general idea was to
not trigger the SCR if the battery voltage was above
14V and trigger the scr if it was below.>
I was hoping it wouldn't have to be that way. IOW, anytime there's more than 14.4V coming out of the system, all the rest of that voltage is just waste?? So when the bike is running at cruise with 80 VAC coming out of the generator, I've really got to dump 65V, huh? (A perfect application for stored flywheel power.<G>)
<With modern POWER MOSFETS ... The bigger problems
are space to put it in, vibration and temperature. >
Yeah, shame there's no battery or supercap that would fit in my palm and be happy with a 100V range and hold many amp-hours.<G>
A honking big zener...ugh. I suspect that's what I just found in an old Marchall (circa 1976) alternator's "diode" frame, six critters with 1.5meg-ohm resistance one way, near infinite the other way, nothing like what a modern PWM car alternator would have.
UGH. Thanks!
|
| |
|
RE: Regulating a motorcylce generator
|
Reply
|
|
by KB1GMX on October 24, 2005
|
Mail this to a friend!
|
Is that 80V loaded or no load?
One sorta wacky way to take the alternator is three (it's three phase) series (before the rectifiers) chokes. One of the oddties is the alternator has to work from around 800rpm to over 8000. At the higher Rs I'd bet the frequency of the alternator output is rather high as in greater than 500hz. A series choke will soften the current flow due to reactance and make
regulation easier.
Fets in switch mode are likely the best bet.
I understand the problems, the older bikes (my
experience ends with my brothers CB750) the
electrical systems were primitive and those that
existed before the lights on rule were worse.
I also remember the Triumph Bonneville with the big
50W Zener under the fork mount for cooling.
Allison
|
| |
|
RE: Regulating a motorcylce generator
|
Reply
|
|
by KC2MMI on October 25, 2005
|
Mail this to a friend!
|
Allison-
<<Is that 80V loaded or no load? >> Both, I think. One of the diagnostics was to check for raw output voltage. "Nothing" was on, but the headlight is always on, so there's always that unbalanced load (55W) directly "in" one of the output leads from the stator.
<<One sorta wacky way to take the alternator is three (it's three phase) series (before the rectifiers) chokes.>> Or run those through the diodes ans work with six parallel supplies.<G> Don't tempt me.<G>
<<One of the oddties is the alternator has to work from around 800rpm to over 8000.>> Yes, but the one in my car has to deal with 800-7500RPM too. Of course, that's got field coils so it is somewhat easier, but the electronics are almost as crude. I'm a Delco fan.
<<A series choke will soften the current flow due to reactance and make regulation easier.>> And heat the chokes, instead of the generator coils?
<<Fets in switch mode are likely the best bet.>> Is that a more economical mode, or an expensive mode, as opposed to the brute force they're using?
<<I understand the problems, the older bikes...>>
Bank robbery can help fix that.<weg>
|
| |
|
RE: Regulating a motorcylce generator
|
Reply
|
|
by AA4PB on October 25, 2005
|
Mail this to a friend!
|
Why do you have to dump the excess power into a load? Why not just use a series switch mode regulator that disconnects the generator from the battery when it does not need charging? That's what the original automotive regulators did. I assume the new solid state regulators work the same way.
Some solar charge controllers work the same way. If the battery voltage drops below about 13V the panel is connected and when the voltage reaches about 14V the panel is disconnected. What you wind up with is current pulses going to the battery with the pulse width and repetition rate varying to provide the right amount of charge.
|
| |
|
RE: Regulating a motorcylce generator
|
Reply
|
|
by KC2MMI on October 25, 2005
|
Mail this to a friend!
|
<<Why do you have to dump the excess power into a load?>> It probably doesn't have to be, but if the power isn't dumped...doesn't that create a back-EMF that actually slows the engine down and heats up the generator? In either case, we'd rather "dump" excess power in the form of extra LIGHT to make the bike more visible, day or night. A pair of 55W auxiliary lamps would make a very useful way to do that.
<<Why not just use a series switch mode regulator that disconnects the generator from the battery when it does not need charging?>> Can those still be found? Practically the entire industry (bike & car) has gone to "real" alternators with all coils--no magnets--and switched away from that technology, AFAIK. I know, generators with magnets are extinct--but it seems like few people are familiar with them these days. I'd love more information on charging systems being used with them.
<Some solar charge controllers work the same way. If the battery voltage drops below about 13V the panel is connected and when the voltage reaches about 14V the panel is disconnected. ...>> Except, the solar panels are rarely putting out 80-100VAC going directly into the battery at 14.4VDC. The charge controllers tend to be packed solutions, targeted to a somewhat different purpose, from what I'd seen of them. Or do you have any specifics in mind, that could adapt to this size and range?
|
| |
|
RE: Regulating a motorcylce generator
|
Reply
|
|
by AA4PB on October 27, 2005
|
Mail this to a friend!
|
When you disconnect the load (battery) from a generator there is no back emf created to load down the engine. When there is no load connected to the generator it just free-runs and the only load it places on the engine is from whatever friction is in the generator.
I agree that the high voltage could be a problem. However, is that 80-100V under load? I expect that it is a no-load voltage and it drops considerably as soon as a load is connected. Of course it still means you have to have solid state switches (transistors or whatever) that are rated to handle the maximum no-load voltage.
|
| |
|
RE: Regulating a motorcylce generator
|
Reply
|
|
by KC2MMI on October 27, 2005
|
Mail this to a friend!
|
So in theory, I could take the three generator leads, run them to six diodes, and then I'd have six leads each possibly putting out 80-100VDC, limited by the generator's capacity of 280 watts, or some 100 watts per line. (Only three lines can ever be active at once, 100 watts on each combining to make the 280W capacity.)
And if I find some components that can drop 100VDC to 14.4VDC at what amperage? I could actually split the load via six diodes, each with a separate lower-rated regulator?
I'm not sure how the 100V will or won't vary with load vs. speed vs. amperage...But can I rashly assume that if the generator can't put out more than 100W per line...the regulators would be handling some 7-8A each max at 14.4V and only less amperage if the generator was indeed putting higher voltage during any point?
So I'd be looking for 6 stud diodes rated maybe 200V (twice the 100V service) and 10A+ to start with, then looking a bit harder for regulators that only would be dealing with under 10A each. (Or some combination like 3 x 20A instead of 6 x10A.)
??
Simple brute force redundancy might work well enough.<G> Any ideas for the regulator chips themseves?
|
| |
|
Email Subscription
You are not subscribed to this topic.
Subscribe!
My Subscriptions
Subscriptions Help
Check our help page for help using
Forum, or send questions, comments, or suggestions to the
Forum Manager.
|
|
|
