n-c alternator modifications: discussion and testing

[DewCatTea-Bob]

Maybe I'm missing something here but hasn't getting to the break even point for a 12v system been more or less the point of this entire thread? That question is so obvious I didn't think it needed to be asked, it's the reason I rejected the stock alternator to begin with and started playing with newer ones....

I have to say, as an external viewer, It is no longer clear to me what the objective is here in this thread, and it seems it may not be consistent amongst you all internally. Is it worth putting down three to five bullet points of what you all think you are trying to do? Let's face it, 34 pages, the thread has morphed, what are you trying to do? I think it is at least worth re-stating - if for no other reason than to let the rest of us know if we should keep reading it!!!

Kev

____ I now guess it's a good-thing that Bill mistook my unfinished sentence as he did, cuz now it's brought to light that at least some of the following readers could stand some orientation (since many posts within this thread have not been properly re-titled, thus defaulting to the "6volt or 12 volt?" original thread-title.)
__ It's true that this thread has increased it's magnification over this certain expanded detailed aspect of the topic of charging-system type, but that's just how things normally go as threads progress.
Currently (and for the majority of pages), this thread is concentrating on how the stock stator could have it's stock individual coil-windings reconnected so as to obtain the maximum power-output which those windings are capable of.
So other than for mere general-interest, our main readers ought be those of us who really care to enjoy the employment of more modern high-output head-lighting, and thus wish to learn of the best method suitable for themselves, to go about obtaining the alt.power needed to run their chosen headlight.
__ Of most significant revelation thus far, it's been realized that the 'common-method' of simply un-grounding the pair of internal stator-winding lead-ends, and full-wave rectifying the (then disconcerted) output of the (yellow) pair of external alt.stator wire-leads, does NOT 'double' the power-output nor make available the stock alternator's maximum-power, as very many have assumed !!
__ I'm in agreement, that since this particular aspect of this thread has become so intensely specific and has also carried-on to become the vast-majority of this thread,, it should all therefore be restarted -(rather moved-over) into another more suitably named thread. _ But not until this particular aspect has been wound-down. _ And at this time, I'm considering that all these related posts (concerning this same aspect), should then be moved-over to the past-expected thread (yet to be created), which was to be topic-titled something like: "Alternator-testing & Experimentation Performed by: wcorey " .


Concerted-Cheers,
-Bob

[DewCatTea-Bob]

Anyway, in this instance, I used this variance to my advantage along with an additional half ohm resistor to get values from .5 to 2ohms in more or less tenth ohm steps. I applied them to ('my version' of) the 'grand series, three section, full wave, series/parallel' setup at both 3450 and 6k rpm. I had more than a difficult time as the results were very close and with the small consistency issues I have with this particular configuration, the ducks didn't always form a row. I went back and repeated many of the tests until I was satisfied that at least the average of the results made sense and fit in with the others. I'm not going to bother posting a detailed breakdown as I didn't organize it very well and was just jotting down numbers as they came up and I bounced around all over the place. I did the tests in random fashion as I obtained the various resistive values in no particular order and used them when I found them. I do have the big multi-tap 1ohm but don't have the adjustable center tap hardware, if I have to do more of this I'll rig something up with that.

What I can say is that at 3450 rpm, the wattage outputs were mostly around 57-60 at 1.3 to 2ohms and did drop off consistently below 1.3. Same for 6krpm but wattage was 133-137. There was such a small percentage of difference between them that with the margin of error I'm dealing with I feel the difference is functionally indeterminable (and likely inconsequential anyway).

____ Wow, you did get into much more detailed testing than I had hoped for !
With your last/previous test-report, the max.power-output was only pinned-down to (the neighborhood of) 1-ohm. _ And that had left the question of whether greater power (than 68w) was available at slightly above or slightly below 1.0-ohms.
We didn't really need to know how-much more power might be available, but did need to know for sure if the max.power's 'peak' was either ABOVE or BELOW that 1-ohm level.
__ I believe I had used past results to figure that the max.power-peak ought be near 1.3-ohms, and if that's what your extended detailed-testing has narrowed it down to, then were pretty-much all-set & on-track to make a conclusion.
____ What's of confusing-issue to me at the moment, is you now report "57 to 60w" @ 1.3-ohms, whereas before you got 68w @ 1-ohm. _ This progression would allow for the possibility of max.power to be found below 1.0-ohms.
We don't want the max.power to be produced below 1-ohm because intended max.system-loads are expected to be between 1 & 2 ohms, (if I'm not mistaking, as no-one has yet reported actually measuring their load-system's max.power-consumption).
Anyhow, it seems that you're likely not doing your last tests on the exact-same test-circuit example (which got the 68w)...
So to clarify what test is of most importance,, we want all three stator-sections -(I & II & III) to be arranged in parallel, (that is to say, each section separately full-wave rectified, and with all their related pos & neg outputs combined and fed into a single load).

" I applied them to ('my version' of) the 'grand series, three section, full wave, series/parallel' setup at both 3450 and 6k rpm. "

__ This would seem to not be of the same tested-circuit,, cuz for one thing, the term: "grand series" was only applicable for describing the 'straight-series' type re-connection of ALL TWELVE individual coil-windings into a straight stringed-together series-circuit. _ Thus 'grand' is no-longer, once that establish string has been broken-down.
Also, while the term "parallel" seems properly associated with the desired circuit for the three stator-sections, the term "series" seems out of place (without explanation for it).
So more than just a chosen set of terms needs to be given, in order to properly realize exactly what circuit-scheme has been tested. _ So if a schematic-diagram (of the particular tested circuit) is needed, (since you certainly already spend enough time doing the testing), then please ask for anyone to draw-up your test-scheme for you. _ Because to insure clarity, a circuit-scheme diagram/drawing is really actually needed to be attached, each time a different circuit-scheme has become the post-topic.


Grateful-Cheers,
-Bob

[DewCatTea-Bob]

Its pretty clear that the often carried out modification of simple unearthing and series connection of a 60w NC stator does not give the best output possible however i for one would not want to carry out anything much more dificult or complicate the wiring with switched outputs or additional rectifiers,I suspect that most owners would think the same,Just hopeing that something practical and usefull to the average guy will come from this.Perhaps several options with estimated outputs and simple instructions ---we can but hope.

____ For the likes of yourself Bob, there has already been a best conclusion reached.
That is to leave the alternator just as Ducati had it ! _ And go-ahead and (slightly) increase your system-load or switch-over to a 12v.system.
I've found no related deficiency-problem with replacing the stock 6v/30w headlight-bulb with a 6v/45w sealed-beam.
And for conversion over-to 12-volts, simply replacing the stock regulator/black-box with a pair of power-diodes, and running lights-on at all times, keeps a 12v.battery balanced well enough (and the battery regulates the juice to the load-system).

[wcorey]

__ This would seem to not be of the same tested-circuit,, cuz for one thing, the term: "grand series" was only applicable for describing the 'straight-series' type re-connection of ALL TWELVE individual coil-windings into a straight stringed-together series-circuit. _

What to call it then, two-thirds-less-than-grand-series? Each of the three sections still have the a series arrangement between the two spools as well as what was formerly two separate windings on each spool also now connected in series.

____ What's of confusing-issue to me at the moment, is you now report "57 to 60w" @ 1.3-ohms, whereas before you got 68w @ 1-ohm. _ This progression would allow for the possibility of max.power to be found below 1.0-ohms.

Yes, you're right, I'll have to get to the bottom of that, must have f'ed something up. Looks like I'll have to make a tap for that adj. 1ohm and do a test series for real...

Bill

[ecurbruce]

Bob says"____ Bruce, did you ever get it figured-out WHICH of Bill's two posted "series/parallel" version-depictions, must be the right-one which correctly corresponds to your inquiry presented him ?"

I don't think there was a "right one" at that point, I was just trying to catch up, and needed a visual aide. Turned out well, though, I got more than I asked for!

[DewCatTea-Bob]

" What to call it then, two-thirds-less-than-grand-series? "

____ If you're indeed still working with the same parallel setup as before, it ought be referred to as 'Three sections in parallel'. _ (If ALL 12 coil-windings were to be connected in parallel, then it would obviously be called 'Grand parallel' .)


" Each of the three sections still have the a series arrangement between the two spools as well as what was formerly two separate windings on each spool also now connected in series. "

____ Indeed! _ However, I'm afraid that 'series/series/parallel' would be too too confusing.


" Yes, you're right, I'll have to get to the bottom of that, must have f'ed something up. Looks like I'll have to make a tap for that adj. 1ohm and do a test series for real... "

____ So I now gather that we are indeed still testing the same tested-circuit -(sections I & II & III) in parallel, (despite the new/lower result-figures).
__ I'm sorry that you seem to have run into another complication Bill ! _ I hope you find a simple/easy-fix for whatever the cause, and we then get back (at least !) that 68-watts !
__ Now that I'm sure of what the tested-circuit is, I'll begin a scheme-drawing for it.


Hopeful-Cheers,
-Bob

[wcorey]

____ So I now gather that we are indeed still testing the same tested-circuit -(sections I & II & III) in parallel, (despite the new/lower result-figures).
__ I'm sorry that you seem to have run into another complication Bill ! _ I hope you find a simple/easy-fix for whatever the cause, and we then get back (at least !) that 68-watts !

This round of low ohm testing has been a head scratcher on many levels, It would take up too much time to explain it all, if I even could. Things like getting a 1.1ohm reading on one resistor and .7 on another and putting them in series and reading .9, then getting repeatable test results that would be in line with it being 1.5 ohm. I can't read the resistance in real time while the test is running but before/after readings go up and down as (I assume) the resistor heats and cools but the volt/current output readings stay steady nonetheless. And that's just a couple of many various odd scenarios I'm seeing.
To counter the problems, I'm doing many redundant tests and more or less using the average set of results. It does appear that 1ohm is about the center of peak power, I got 67.6w on one test but some are closer to 66w, seems like splitting hairs to me but what do I know. Outcome of the test readings look bell curve-like, seem to take a dive below .8 and above 1.3.

Three sections in parallel, each independently full wave rectified, outputs combined going into one load.
3450 rpm

.5ohm-----53.2w
.6ohm-----61.1w
.8ohm-----65.4w
.9ohm-----66.2w
1ohm------67.1w 8.11a, 8.28vdc
1.1ohm---65.7
1.2ohm---64.9
1.3ohm---64.4
1.4ohm---61.0
1.5ohm---59.9
1.6ohm---56.3
1.7ohm---58.6
1.8ohm---56.1
1.9ohm---55.3 5.36a, 10.32vdc



Bill

[DewCatTea-Bob]

" It does appear that 1ohm is about the center of peak power, I got 67.6w on one test but some are closer to 66w, seems like splitting hairs to me but what do I know. "

____ Well if we were simply looking for merely the max.power, that small difference would indeed be "splitting hairs". _ And it appears that you've done more tests than need be done to determine the peak of the working-impedance but, it's understandable that you actually had to do so, so as to make sure that things are properly adding-up for a consistent dependable trustworthy result.


" Outcome of the test readings look bell curve-like, seem to take a dive below .8 and above 1.3. "

____ That seems as it should be, to confirm that the actual working-impedance is not below 1-ohm, (and looking to be only slightly over that).
While it might be okay for an intended system-load to have a total-resistance down to that .8 drop-off point, it would be quite undesirable for the load-system's total-load to ever dip below that point (for any extended time).
__ Anyone considering this best/top power-production arrangement, ought to check the load-resistance total of their intended load-system, to be sure that it's not much lower than 1-ohm.


" Three sections in parallel, each independently full wave rectified, outputs combined going into one load.
3450 rpm "

____ Just what we need !


" .5ohm-----53.2w
.6ohm-----61.1w
.8ohm-----65.4w
.9ohm-----66.2w
1ohm------67.1w 8.11a, 8.28vdc
1.1ohm---65.7
1.2ohm---64.9
1.3ohm---64.4
1.4ohm---61.0 "

____ These above figures indicate that the resulting working-impedance is certainly not too low for a high-consumption load-system like Bruce plans, (which as this is, really is what he actually needs).
I'm sure it's mere coincidence that the peak working-impedance seems to be at 1-ohm (actually indicated to be just under 1.0-ohms), as the 1-ohm results from the three sections being in parallel, (suggesting that each section has an independent working-impedance of 3-ohms, alone).
__ These figures indicate that the tested stator arrangement is pretty-much good for 67-watts down to 1-ohm, meaning that if the intended system-load's total-resistance became as low as 1-ohm, the alternator could then handle the total-load with max.efficient power-transfer of it's full 67-watts. _ And if the system-load's total-load became merely .7-ohm, then there would be a drop in power-transfer efficiency and therefore a power loss (of about 4-watts, meaning that slightly higher RPMs will be needed to get back to the very-same relationship to the 'break-even' point).
The 8.3-volts indicates that at this 3450RPM, the breakeven-point has not yet been reached for a 12v.system with a load as great as just 1-ohm. _ So to break-even at a RPM that's as low (or lower), then the system-load has to be actually less demanding than that of only 1-ohm.
Some basic-math would indicate that a 2.2-ohm total-load would allow breaking-even near the 3450.
So it seems the good news for Bill's project, is that his load-system ought tally to be effectively near 2.5-ohms (+/-.1), so he could expect to break-even between 3100 & 3300 RPM. - (According to my fuzzy-math.)


" 1.5ohm---59.9
1.6ohm---56.3
1.7ohm---58.6
1.8ohm---56.1
1.9ohm---55.3 5.36a, 10.32vdc "

____ Now of clarifying interest concerning all the result-figures obtained with load-resistance values higher than the 1-ohm peak working-impedance...
The somewhat seemingly magic 1-ohm for maximum available power, is merely indicative that a MAXIMUM of 67-watts is AVAILABLE at 3450-RPM. _ So for load-systems with total-loads which consume less power and thus equate to ohm-values (likely) between 1 & 3 ohms, this merely means that somewhat less than the maximum 67-watts will be actually needed. _ In other words, concerning ohm-value figures like those listed just above which are consuming less power than at the 1-ohm level, the related indicated wattage-values are merely what's being consumed, and is NOT only what's available.
As the testing has determined that as much as 67-watts are available at all times (during the 3450-RPM), and all those lesser wattage-figures are merely the amount (of that 67w) being consumed. _ So for instance, where we see that the 1.5-ohm load consumes only 59.9-watts, that means that there's then 7-watts left to spare !
Furthermore, it should also be realized that it's only important to insure that the system-load's resistance-value does not drop (significantly) below that of the working-impedance of the power-circuit, but on the other-hand it's perfectly-find if the system-load's value is way lower than the working-impedance ! _ (Their respective values only need to be 'matched' when seeking MAX.power-transfer.)


Content-Cheers,
-Bob

[Bevel bob]

Its pretty clear that the often carried out modification of simple unearthing and series connection of a 60w NC stator does not give the best output possible however i for one would not want to carry out anything much more dificult or complicate the wiring with switched outputs or additional rectifiers,I suspect that most owners would think the same,Just hopeing that something practical and usefull to the average guy will come from this.Perhaps several options with estimated outputs and simple instructions ---we can but hope.

____ For the likes of yourself Bob, there has already been a best conclusion reached.
That is to leave the alternator just as Ducati had it ! _ And go-ahead and (slightly) increase your system-load or switch-over to a 12v.system.
I've found no related deficiency-problem with replacing the stock 6v/30w headlight-bulb with a 6v/45w sealed-beam.
And for conversion over-to 12-volts, simply replacing the stock regulator/black-box with a pair of power-diodes, and running lights-on at all times, keeps a 12v.battery balanced well enough (and the battery regulates the juice to the load-system).

When my bike came to me it was rewired with a 12v podronics regulator which did not work with the Earthed coils, from the above comment i assume that to revert to a standard stator i will need to revise the wiring and dump the podronics unit, what is the likely gain in power?,sounds like i will also loose charge regulation.

[DewCatTea-Bob]

" When my bike came to me it was rewired with a 12v podronics regulator which did not work with the Earthed coils, "

____ I'm unfamiliar with the Podronics-unit, but I assume it employs a FW.bridge-rectifier, which of course would not work with any part of the stock stator-circuit grounded !
So in hind-sight, you should've simply replaced that R.R-unit with a dual-diode rect.block.


" from the above comment i assume that to revert to a standard stator i will need to revise the wiring and dump the podronics unit, "

___ Well, when you went to the trouble of un-grounding the pair of stator-winding lead-ends, did you also at that time replace the original-stock alt.cable with new wire-leads (as is known to be a good idea !) ?
If so, then hopefully you thought to add an extra wire-lead or two, to the un-grounded lead-ends, so that you could easily try other circuits without having to pull the alt.rotor again.
So you see that you would not really want to "revert to a standard stator", as that was a cheap cost-saving circuit-setup which Ducati had employed !


" what is the likely gain in power? "

____ Well we'll have to look-over all the related test-outcome result-figures but, it's been figured that leaving Ducati's stator-windings as-is, along-with the stock-like dual half-wave rectification, will produce quite-nearly 75% of the power which could be obtained by full-wave rectification of the entire stator-winding after being reconnected over-to the 'straight-series' type re-arrangement/modification.
And on top of that, the max.power-output would be made available down at a lower working-impedance, where the available max.power is then more apt to be usable ! _ Thus there's a fair possibility that a certain load-system could actually get fed MORE power than it could possibly manage to get from the more-powerful full-wave & straight-series circuit-setup !


" sounds like i will also loose charge regulation. "

____ Here's the low-down on that...
Voltage-regulation is really only needed mainly to protect lights from burn-out, and the (normal-type) battery will do that job ! _ But if a charging-system produces TOO much power (for an extended time), then the battery itself also needs to be protected,, and that's the main-purpose for a regulator-unit !
But realize this - if the load-system is kept fairly equally matched with the power-production, then the battery needs no such protection and the battery will maintain system-voltage !
All you need to do is make sure that your chosen load-system consumes slightly less than the AVERAGE power-production of your charging-system. _ And that's quite easy to accomplish with an ammeter installed on your Duke !
____ Here's a plan for you to consider... Install an ammeter (and perhaps a detailed volt-meter) on your Duke just as it is with the charging-system which you already have. _ Get used-to how-&-when the breakeven-point comes & goes, then replace your regulator with a simple bridge-block and note any metered changes.
__ You should learn sometime during the process, how to balance-out your overall-system with riding-habits and use of lights, if you aren't already doing so without trying.
You might learn that simply using a more powerful set of lights will make a regular-unit totally unnecessary.
__ Then come winter, you can then change your alternator wiring as you then wish, for the next riding-season,, and then compare your meter-readings.


Hopeful-Cheers,
-Bob