Stator rewinding theory question

[Sidecar Bob]

I have wanted to try rewinding a stator for years and It is time to replace the one in my CX650E based winter machine so I figure this is the time to do it. Because they use the same armature and the wire is visibly larger I have run GL1000/1100 stators in my GL500 and then my CX650E based machine for a lot of years and had good results but what difference it makes is just a guess. If I am winding one myself I would like to do things right and end up with a bit higher available current but not so much that it kills the Zener in the reg/rec.

So my question is:
The information I have indicates that the standard G8 type stator used in GL500 and CX/GL650 uses 16 AWG wire and 24 turns per pole. I understand that increasing the wire size should increase the current capacity but how much bigger should I go? Also, do I need more turns, less turns or the same as stock?

[socrace]

Doing better than oem isn't easy.
It's like the famous saying about the 3 laws of Nature: 1.you can't win 2. you can't break even 3. you can't change the rules.
One possibility, though, is to buy more expensive wire that's same gauge as oem but with wire diameter slightly smaller due to thinner insulation. Then you can rewind a turn or two more in the same space and increase low rpm charging ability. Try to wind all 3 phases as identically as possible, to avoid losses due to unbalanced currents.
Regulators don't use a zener like the old British bikes, they short the windings down to less than a volt, so power (voltage x current) is very low.

[Sidecar Bob]

The drawing below is from the GL1000 FSM. See that diode with the ends of the perpendicular line bent over? That's a Zener. I'm not 100% sure that the reg/rec in my 650 has a zener but I would be very surprised if it doesn't.
If you don't already know, a zener diode has a rated "threshold voltage" below which it does not conduct and above which it becomes a short circuit.


The information I have about rewinding a CX stator says to use 1.25mm wire and 16 AWG is 1.29mm. After I posted that the original had 16 AWG wire I was told on the CX500 forum that it the original wire is actually 1mm (18 AWG = 1.02) so the 16 AWG is already bigger than the original. I am told that it is the maximum size for 24 turns.

[socrace]

I'm an electrical engineer (probably should have stated that) with experience in power systems.
The advice I was trying to give you was that you don't have to worry about overpowering an SCR regulator with new windings. Because of the way they work, relatively little power is taken from the windings. The schematic you posted is an SCR regulator, not a zener regulator. Zener regulators were obsolete from about 1970 on.
Hand winding a stator is a major undertaking and I respect your ambition in taking it on.

[Sidecar Bob]

I used to know a lot more about some parts of electronics than I remember now (use it or loose it) but I see a zener and an SCR in there. If I am interpreting it correctly the zener triggers the SCR which becomes the actual short (or near short). The end result is still a load of current being dumped to ground and that has to cause heat. And the higher the voltage being regulated the higher the current that has to be dumped and thus the greater the heat.

But we're only talking about a small difference so it probably wouldn't overheat the regulator anyway.

[Sidecar Bob]

I was just thinking - your idea about getting a couple more turns per pole is intriguing. From what I remember of transformer theory from school (over 40 years ago), starting with only 24 turns per pole, one extra turn should increase the output voltage by a bit over 4%. I think that means that if the stock alternator starts charging the battery at 2500RPM (just an example pulled from my head) one with a single extra turn per pole should start charging at about 2400 RPM and 2 extra turns at about 2300 RPM. For someone who spends a lot of time in city traffic that could be useful. But I spend most of the time at 80+ Km/h so it probably wouldn't make a significant difference to me.

Someone on cx500forum suggested that increasing from 1mm wire to 1.25mm wire would let the stator "run cooler and have roughly a 10 % higher output so should be giving 275 watts" (original spec is 252W). What do you think?

BTW: This pic shows the original stator (left) next to a GL1100 stator (right). After I took that I added longer wires and the appropriate grommet to the 'Wing stator and installed it in the 650 for 7 years until a winding decided to short to ground.

Stator Comparison.JPG

[backyardbob]

A few years back I rewound the stator in a 1200 with great success. I used 16g wire (from our local motor rewind shop).
about 20 years ago I used the schematic from my honda manual (you could buy them from a honda dealer then) to reconstruct my regulator. But one of the sensing resistors (on the right in the diagram) I replaced with a variable resistor which allowed me to adjust the voltage . The first one lasted eighteen years. Presently testing another same design.
I just guessed at the value of the capacitors, got lucky. I would like it if anybody could advise on their value.
I agree that the exact number of turns on each stator core is not important, It normally puts out 50 volts or so ac, so a volt or two is not important.

[Goboy]

The big news in electronics currently is oxygen free speaker wire, it claims to have higher conductivity and corrosion resistance. I don't know if you can get 16 ga. strand and it would be pricey. Probably worth looking into.

[Sidecar Bob]

Yeah, and speakers are such a critical application that they need every bit of help they can get....

[CYBORG]

[Sidecar Bob]

Clicked Submit before I was done typing

I spent 12 years designing speakers and I remember when the original Monster Cables came out. They were made out of what appeared to be multiple strands of magnet wire braided together and they claimed that they had all sorts of beneficial qualities but what they actually turned out to have was a large internal capacitance. The result was that they had low impedance at very high frequencies and they blew the outputs of enough amps that they removed them from the market.

BTW: I still recommend good old 16 gauge lamp cord for speakers. And if you just need a couple of short pieces get a couple of extension cords from the $ store and cut the ends off.

[backyardbob]

a critical element of stator winding wire is temperature resistance. around 195 degrees or so. I doubt if speaker wire will meet this requirement.

[Sidecar Bob]

I dug out the failed stator today. First I measured the resistances again to see how far off my memory was. One pair read 0.5 ohms and each of the others about 6 ohms.

Next I unwound it and counted the turns on several poles. There were 27 turns per pole.

With the insulation removed I measured 1.22mm, with the insulation 1.25mm.

So the stock 252W stator used in the GL500 & CX/GL650 has 24 turns per pole of 1.0mm wire and the stock 300W GL1000/1100 stator has 27 turns per pole of 1.25mm wire.

I guess that explains how they got 50 more watts from it.

If I expected to have to drive to work every day in the future &/or I really wanted to get absolutely every watt possible out of it I would probably consider one of Randakk's stators but I think I will rewind my own just so I can say I did.

[Whiskerfish]

Looking forward to seeing how this works out for you. Not something I would have ever considered myself.

[Sidecar Bob]

I picked up my wire today. I found a local place that does motor rewinding and he wrapped a pound of #16 around a corrugated box and sold it to me for $10. I asked him about what to coat the armature with and he said to just clean up the original stuff and touch it up with epoxy so the corners don't nick the new wire.

I also asked him about what to pot it with. He said they dip the rewound armatures in a varnish tank, then bake them in an oven. He thinks regular varnish would do and I could bake it in a toaster oven but he would dip & bake it for me for $25. I am considering going that way because it would be a lot less mess & bother.