n-c alternator modifications: discussion and testing

[DewCatTea-Bob]

By: MotoMike...
" MM
Clearly there is some effect exhibited due to the low ohmic value. Not sure what you mean about the stator windings being part of the load impedance. "

____ The DC-resistance of the alt.windings is not of as much importance as their exact 'impedance' to AC-flow (which varies according to frequency), so to get max.wattage output from the alt.windings, the load must have the same impedance as the alt.windings.
The latest tests you asked for now indicate that the 1-ohm load seems closest.
____ The circuit you've asked to have tested provides current-flow the ability to take the path* that's most efficient, and so is possibly responsible for the superior wattage-figures obtained (with that circuit of yours).
(* Either the 'parallel-path' [returning to the same stator-winding which produced it],, and-or, the 'series-path' [back through both windings].)


" Here is my initial thought. the stock system will allow as low as .6 ohms of load before the fuse goes. So the system in stock configuration can handle one ohm and not be bothered by it. But if there is an opposing field condition at play here, maybe dropping down to one ohm allows the strength of these fields to become large enough, owed to the current through the stator coils to have a significant effect. In the stock system as we have mentioned, since only half the stator is used at a time, the strength of the fields generated from load current and their effect on the other stator half are of no concern. "

_____ Right, that's a summery of what's been accepted as possible previously, some of which is waiting for testing to help confirm or deny.
I've merely responded to your thought of 'what-if' the opposing-fields possibility is found to not be responsible for the lack of full power-output during the alt.windings in series, measurement-tests.

____ Do you think it might be worth testing the alt.windings for their 'impedance' measurements ?


-Bob

[MotoMike]

Bill

thanks. I am having trouble keeping all the data straight. Did you not test the 3 ohm series circuit before? Even the 3450 rpm figures would tend to show that you could charge a 12 volt battery. If you had a 60 watt load on the system, that would work out to 2.4 ohms if my math is right, in the ball park of what you have the system loaded with here. I think I spend most of my time above 3k just taking it easy driving around. That would explain Bevel Bob's account of his setup.

very interesting data, but I think I am seeing inconsistencies that maybe when put on a spread sheet might make sense as I am probably mixing up the load and or rpms.

Mike.

[wcorey]

Today I decided is the day to get to the bottom of alternator coil anatomy...

Yes! This is what I've been waiting for someone to take on, nice to clear up another whole area of speculation.

I'd be interested in obtaining the other two coils, assuming there are another two and if you want to part with them.

Tomorrow I may decide to unwind one of the smaller coil assys., cause now I'm not 100% sure that it's wound counterclockwise.

Wouldn't it have to be wound clockwise? For some reason I thought basic alternator theory mandated that every coil be counter to the next, or is that just magnet polarity?

...some of which is waiting for testing to help confirm of deny.

What particular testing is this referring to? If it's on my end, I'm ready for another round...

Even the 3450 rpm figures would tend to show that you could charge a 12 volt battery.

To get a decent charge, you need to be seeing close to 14vdc, I don't recall seeing that in any of the 3450 tests, though my head is spinning along with the rotor so I could be missing something.
<Edit> Oh Duh, I just went back and looked at today's series 3 ohm 3450 results, 15vdc. Guess I'm getting a bit saturated here....

<2nd Edit> I went back to see where I got such a notion, apparently I was reading Bob's quote of my results on Bruce's dual rectifier test. Got confused evidently...

This changes everything, now I'm very interested in a nice luke-warm alternator with adequate output rather than a high output toaster. Some load/no load temp testing is where I'm headed next...

Bill

[ecurbruce]

Bill,
there's one more configuration that I'd like to see, I introduced this earlier, but I never saw data from it. Maybe I missed it?
Thanks in advance,

Bruce

[DewCatTea-Bob]

By: wcorey...
" Still rather a disappointing result, from my particular practical perspective.
What concerns me is not so much the wattage but that at 6k it's just getting to the point where the voltage output will start charging a 12v battery. Obviously it wasn't designed/intended for that purpose, I'm sure it works great at 6v, "

____ This is a faulty conclusion (in more than one way, if your Duke's total-load isn't over the 107-watts), even without knowing the test-results...
For one thing, an alternator can't be 'designed' to produce 12 or 6 volts, when it's RPM can't be expected to stay steady.
(And it's already known that a 12v.battery can be charged even with much less voltage-output than the stock charging-system is capable of, [providing the load is not so great, that all the wattage gets swiped before the battery gets enough to charge with].)


" also seems to put out substantially less heat than it's contemporary counterparts, "

____ But wasn't your other testing (done with the non-Ducati alternators), performed with regulator-units (which may have contributed to the excessive heat, due to becoming like a load as well) ?


" Bottom line, the modern stuff intended for 12v operation puts out more voltage at less than half the rpm (and lots of different rotor weights) so for me it's still the direction I'll likely stay in. "

____ Right, it would be a waste of all your gained experience with such, if you didn't make much use of it !


" My next step there will probably be in figuring out how to down rate the outputs a bit. "

____ You mean cut-down the power-output of those modern alternators ? _ If so, a power-diode should do.


" 1ohm load 3450 rpm, 7.1a, 10.6vac, 6.9vdc "

____ These figures are concerning Mike's presented circuit. _ And the measured DC.power-output is 49-watts (with the 1-ohm load), which I believe is the highest measured (by at least 3-watts).


" 3ohm load 3450 rpm, 2.9a, 15.4vac, 11.9vdc "

____ The load-decrease (from 1 to 3 ohms) lowers the measured-wattage to 34.5-watts.


" 5ohm load, 3450 rpm, 2.4a 15.8vac, 12.2vdc "

____ Further decreasing of the load (another 2-ohms) brings the measured-wattage down to 29-watts.


" 7 ohm load, 3450 rpm, 1.8a, 16vdc, 12.7vdc "

____ Lowering the load-resistance still further -(another 2-ohms, to 7-ohms), brings the measured-wattage down to near 23-watts.

__ I wonder if Mike has any explanation for why he requested such further testing !?


Content-Cheers,
-Bob

[wcorey]

there's one more configuration that I'd like to see, I introduced this earlier, but I never saw data from it. Maybe I missed it?

Yep, you missed it ( or fell victim to it, lol) back on Saturday/page 11.
See the previous page (15), down about five posts, second picture...
I'll fix it once everyone 'gets it'.
Sorry for that...

____ This is a faulty conclusion (in more than one way, if your Duke's total-load isn't over the 107-watts), even without knowing the test-results...

Yes I'm aware of that now, see my previous post.

For one thing, an alternator can't be 'designed' to produce 12 or 6 volts, when it's RPM can't be expected to stay steady.

No, but it does need to at least reach the intended voltage it's obviously designed to operate for/within, at an rpm that also suits the purpose. Just happens in this case it's probably close enough for either application.
You can't tell me (I know, you're not) that the operating voltage range isn't a consideration when designing an alternator for 6 or 12 vdc.

(And it's already known that a 12v.battery will charge even with much less voltage-output, [providing the load is not so grate, that all the wattage gets swiped before the battery gets enough to charge with].)

I'm under the impression that to keep a lead acid battery from sulfating, it needs to be kept up over 12.6 volts, up in the 13's for AGM. To get there, the charging system needs to be producing around 14 or more vdc. Some of the new LiFePo4 batteries (that I'm considering to use) need over 14 to charge at all.

" also seems to put out substantially less heat than it's contemporary counterparts, "

____ But wasn't your other testing (done with the non-Ducati alt.s), performed with regulator-units (which may have contributed to the excessive heat) ?

A load is a load, and the R/R's don't necessarily consume multiple amps by themselves. Though actually the alt's put out heat even when run standalone with no R/R. The pitfalls of permanent magnets...


Bill

[MotoMike]

there's one more configuration that I'd like to see, I introduced this earlier, but I never saw data from it. Maybe I missed it?

Yep, you missed it ( or fell victim to it, lol) back on Saturday/page 11.
See the previous page (15), down about five posts, second picture...
I'll fix it once everyone 'gets it'.
Sorry for that...

Bill

Is it the third photo down on page 15?

[wcorey]

Is it the third photo down on page 15?

You got it. Did I do it right?


Bill

[Bevel bob]

My 60 watt NC alternator with simple stator coil un-earthing and linking alteration produces over 14volts under load and keeps my battery charged even with lights on , It would be nice if someone could test this settup to see what is actually availiable .I think this is what most of us need.(14 pages of gobbledygook and i'm none the wiser), I do try to keep the revs over 3k as it is better for the big end, a 35 watt HL is fitted,but but 50w would be better,I expect that would leave me stranded in the backwoods though.

[DewCatTea-Bob]

____ Apparently my previous post was snagged before I got it completely done.

For one thing, an alternator can't be 'designed' to produce 12 or 6 volts, when it's RPM can't be expected to stay steady.

No, but it does need to at least reach the intended voltage it's obviously designed to operate for/within, at an rpm that also suits the purpose. Just happens in this case it's probably close enough for either application.
You can't tell me (I know, you're not) that the operating voltage range isn't a consideration when designing an alternator for 6 or 12 vdc.

____ Well Yes & No... 'Yes', the alt.designer must be in consideration of all related voltages, but 'No', he can't do much to directly control the particular 'voltage' output-value of the alternator itself, cuz such a std.type alternator (made to run at varying RPMs) doesn't really produce a 'voltage' (or 'amperage') of any particular value/amount , all on it's own... as alternators merely produce electrical-POWER that's measured in 'wattage', and it's actually the external electrical-components (in circuit with the alternator) which are responsible for what cause (variable!)- voltage & current to be derived from the 'wattage' output of the alternator. _ (Which is just one reason why I suggest that people should think in terms of 'wattage' [rather than just voltage, or current].)

____ Now I may be jumping-over some posts at this point, cuz it seems this thread is creating new pages faster than I can keep-up and also it seems Bill has stopped noting the "round" number of of his various posted testing results, (I think I last recall "round 4"), so I'm lost as to whether he's up to round-7 or what,, and with my PC having trouble quickly loading thread-pages with many large pix, it's not easy to figure-out on my own where we're at or hat all has been done, so I can keep-up record of everything.
__ So next (for now).....

...some of which is waiting for testing to help confirm of deny.

What particular testing is this referring to? If it's on my end, I'm ready for another round...

____ As I had indicated hopes of testing each alt.winding alone, and then both connected in series, both ways. _ All four tests done with 'half-wave' rectification, (this time).
However, you then requested schematic-diagrams for each, and I have yet to draw-up & post the two series examples.
I'll get to those by tomorrow.

(Still probably more stuff to insert into this posting (?).