n-c alternator modifications: discussion and testing

[Bevel bob]

Managed to find a heavier wattage bulb 40/45 watt, and found that the voltage at the battery still rises slowly to 14 v at about 4k revs, the battery is a bit down so i think the controler (podronics) is alowing this. Don't think a higher wattage bulb will balance (load/power) with the back lanes riding 3--4k revs .

[wcorey]

I'm wondering how close the magnets are to the core compared to the originals.

It ended up being about .035" gap on the 8 pole setup, with the original 18 coil 3ph stator it was .040". The stock 6 pole Ducati alt is pretty tight at .020", noticeably more fussy to set up on the test fixture.
The force of the magnets seems fairly close between the two, the ones in the newer rotors are even stronger and the magnets even smaller still.



The 3 sec, 3phase mod results have given me the confidence to do what I've wanted to for some time (but didn't want to jump in too fast and mess things up permanently), which is to modify one of my newer stators.
The zx14 alt was the most obvious choice for a few reasons, it's the biggest overkill, has a rotor with almost identical weight to the ali 6 pole, has very heavy coil wiring and has more space between the coils making it easier to work with. The rotor is 20 pole, the stator had 15 coils in a delta configuration and I lobotomized 9 of them, leaving 3 pairs of coils to get a third of the original output (which was rated at (I think) 450w).
The output results were very close to the 3 section 3phase mod with the smaller loads but kept going up as I increased them. The highest I tried was with the pair of caps, 55w headlight and 1 ohm resistor,
4310 rpm---15.56a---13.23vdc---205.85w! Seems the 450w rating of the original alt was maybe a bit conservative...

It stayed fairly cool at about 10 degrees above ambient, a huge improvement over it's former heat output which was substantial even at no load and with a moderate load something like 80 or 100 over ambient. It did get up to 20 over when I had it putting out 200w but that's not really a valid comparison as it will never go there in actual use.

Looks a little unlikely but somehow gets the job done.





Other than that I can do a similar mod to a couple other 12 pole, 3ph alts I already have fitted to the 450, looks like my testing has more or less culminated, now to go back and sort out what's been done. I still would like to find what's the best 'easy' mod to the stock 4 coil stator for those who don't want to get as deep as we have here. The info may already be here...

If anyone wants any of my test setup pics on this thread (I can't imagine why) you may want to make copies as I think they're superfluous at his point and just make it more difficult to sort through the data so I'm going to go back and clean most of them out.

Bill

[ecurbruce]

Bill,
Are you able to (may not want to, I don't know) switch the pullies on the test motors for slower RPMs? Looks like the smaller pully is on the fixture motor. I think I saw where you wondered about statistics at slower speeds...

Bruce.

[MotoMike]

Very cool Bill. are the pics of the altered stator above one that would fit in the 450? Looks great. did you have to saw off the core nubs?

Mike

[wcorey]

Are you able to (may not want to, I don't know) switch the pullies on the test motors for slower RPMs? Looks like the smaller pully is on the fixture motor. I think I saw where you wondered about statistics at slower speeds...

I have lots of pulley's I can use so that's not the issue. Earlier on when the output of the stock modded alt was barely putting out enough power at 3450 rpm to break even, I didn't see the need to go any slower. But now that some of the setups are producing enough it is making me curious to see what the lower rpm performance is so I may try a 1725 rpm drive motor with cone pulley's on both motors. I still want the higher rpm's and don't want to keep swapping pulley's to go back and forth.

...are the pics of the altered stator above one that would fit in the 450? Looks great. did you have to saw off the core nubs?

Basically if it works on the test fixture, it works in a single. All of the alts I've been playing around with have been set up to work on the 450, though the zx14 rotor is very close to the max diameter that will fit. Rotor width has been the biggest limiting factor in finding workable solutions but I think would be less of an issue on n-c motors because the smaller stator mounting plate fits inside most of the rotors.

I didn't have to saw off all the extra coil cores to make it work but both you and my physicist friend indicated that eddy currents would still produce heat from them even without the windings. That and the rotor magnets would still 'pull' on them, taking more out of the motor output to run it.





Bill

[wcorey]

Bruce, I added a step pulley to the the driven (fixture) motor and also dug up a 1725 rpm motor with a small pulley. With the 1725 drive motor, I can get down to <EDIT> 997 rpm and with the 3450 rpm drive motor down to 2200 rpm and up from there in a number of increments on both. I'm assuming you want numbers for some particular setup of the stock 6 pole rotor/3 section stator? So now's your chance...

I happened to have set up at the time the single phase 12 pole Denso alt. and could get 117w at <EDIT> 997 rpm but 'only' 11.9vdc, however at 56w it did 13.9vdc so should easily charge the bat with the headlight on even very near idle. I'm figuring it's so much more efficient at that rpm than the Ducati alt because it has a 12 pole rotor and 12 coil stator...


Bill

[ecurbruce]

Bill,
When I read your post, I just have to laugh out loud in delight, as you once again exceed all expectations...
Where else can one get this kind of data, though it's not always come easy, I think it's been well worth the journey.

The denso single phase alternator test is impressive, at only 616 RPM. What kind of load did you run it with?

Now that you mention it, and since you're there, it would be good to know the break even point of the three section ducati stator into the shindengen FH 008 ( as on page 37) with a load of the 55 watt headlight, 2 caps., and 5 ohm coil, reducing the speed until it's no longer enough output to charge a battery. Not that there has ever been any expectation of your research, just much appreciation!

Bruce.

[wcorey]

I screwed up the rpm calculation on the denso 12 pole testing, it was 997 rpm not 616, must have transposed the pulley diameters or something. Still pretty good numbers though...
The load I used to get the 56w at 13.9vdc was the 55w headlight with the pair of caps, I think the 117w was with a 1 ohm res but don't remember exactly.

I set up an ensemble to more or less emulate a 'real' bike with the 'bat eliminator' caps, a 55w headlight, tail light, and functioning ignition (hei module, magnetic pickup, 5ohm Dyna coil and plug, spun with an air die grinder) and a 24ohm resistor load 'guesstimated' to cover the use of intermittent incidentals like horn, brake light and turn indicators. (Speaking of brake lights, on a typical 12v (1056?) dual filament tail light/brake light bulb, the brake light is 21w, fairly significant when you're sitting in traffic idling with a foot or finger on the brake.)

The total draw from all of that was still only 66w at 12.4vdc, the headlight is obviously the 800lb gorilla of electrical load here. I ran the 6 pole, '3 section' modded alt with combined outputs into two inputs of the FH008 R/R at <EDIT> 2690 rpm to arrive at that, 5.34a--12.41vdc--66.26w.

The next pulley combo's/rpm ratios up or down ( about 4-600 rpm difference) went right up to the 13's or down into the 11's. So average breakeven looks to be around the upper 2000's. I did the same test using only four of the six stator coils but forgot to write down the numbers so I'll do that one again at some point, the results are pretty predictable though.



One thing that's really been annoying me is that I can't get the 6 coil Ducati stator to work with the 12 pole Denso rotor, seems to me if it worked would provide double the power output at a given rpm. Everything (properly configured) points to it being good to go but I get very low numbers, like 4vdc where it gets over 12 with the 6 pole rotor under otherwise similar circumstances. I think as of now I officially give up on that one...


Bill

[Jordan]

I like the mad scientist look of your test bench.
For my own experiments with do it yourself electronic ignition, I needed a snail cam shaped rotor to trigger a magnetic pickup reluctor. I found one from a motorcycle rear axle, chain adjuster worked perfectly.

Jordan

[DewCatTea-Bob]

____ Bill, (to set the mood) - you've been very resourceful and have also been doing good useful work, I'm sure it will always be very much appreciated !!

By: wcorey ...
" I set up an ensemble to more or less emulate a 'real' bike with the 'bat eliminator' caps, "

____ Capacitors (IN-PLACE of a battery) offer system-behavior very much different than that with a std.battery connected-to the system !! _ For one thing, the caps do not act like a load, as a battery does.
While your tests with the caps included allow for straight-forward type test-measuring and thus obtain all the desired factual data of what power is produced & consumed (by replacing the fairly less straight-forward 'wild-card' effect which a battery would have),, it ought to be especially noted that real-world outcome-results would be significantly different with a real battery included in the test-system !


" and functioning ignition (hei module, magnetic pickup, 5ohm Dyna coil and plug, spun with an air die grinder) "

____ Did you note any difference in current-draw depending on the spun RPM ?
__ Once it has been established what the current-draw is for the 'working' ignition, it could then be simply emulated with just a resistor, in place of that elaborate setup.


" and a 24ohm resistor load 'guesstimated' to cover the use of intermittent incidentals like horn, brake light and turn indicators. "

____ I'm thinking 24-ohms is possibly too high of a resistance-amount... It's my opinion that the 'average-resistance' of all those type of temp.loads ought to be figured-in some way (to equate city-driving), for when testing with the caps, (without a battery included).
It may be fair to figure-in an added resistance-amount which equates what a battery-draws averagely, cuz when a battery is included, those temp.loads have virtually no overall-effect (whether used or not [in normal fashion] ).


" Speaking of brake lights, on a typical 12v (1056?) dual filament tail light/brake light bulb, the brake light is 21w, "

____ I've brought-up this subject before (in another thread), I'm sure that's actually 21 'CP', which is more like 15-watts.
I believe I recall ya could get t.bulbs with either 3/18-CP or 4/21-CP, which is about 2 to 3w for tail, and 14 to 16w for brake lights.
I don't have the reasoning for why t.bulbs are rated in 'candle-power', instead of watts.


" One thing that's really been annoying me is that I can't get the 6 coil Ducati stator to work with the 12 pole Denso rotor, "

____ Bill, how soon you forget ! _ (Or maybe never understood before, in the first place.) ...
We have already gone-through this issue before (within the other thread where you first started covering this non-stock alternator testing stuff).
I had back-then explained that 12 rotor-poles can't work because the 6 stator-poles require OPPOSING polar-poles at EVERY-OTHER stator-pole/core-finger,, where-as a 12-pole rotor must align either all North or all South polar-poles at ALL of the 6 stator-poles, all at once !
(You'd need an 18-pole rotor, in order to (once again) get back to the opposing polar-poles aligned at every-other stator-pole, as is the original case with the 6-pole rotor !)
__ To better get a grasp of what's happening with the two aligning sets of poles -(stator & rotor), it helps to look at a clock-face. ...
If you imagine that the original 6 rotor-poles are (starting with North at 12-o'clock) - 12-N; 2-S; 4-N; 6-S; 8-N; & 10-S ,, then a 12-pole rotor must be 12-N; 1-S; 2-N; 3-S; 4-N; 5-S; 6-N; 7-S; 8-N; 9-S; 10-N; 11-S . _ So as can now easily be noted,, with the 12-pole rotor, ALL of the 6 stator-poles are all at once aligned with a North polar-pole, (during that particular rotor location instant), and the very-next pole-alignment will of course be ALL South !
__ So then, did you ever find yourself any suitable 18-pole rotors ?


Hopeful-Cheers,
-Bob