Author Topic: Converting an excersize bike to a generator test bed  (Read 8897 times)

0 Members and 1 Guest are viewing this topic.

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: question ???
« Reply #33 on: December 09, 2005, 12:37:32 AM »
More thinking outloud.

Can't say if that is too high. But when 20A is flowing, the losses start to add up fast. I wouldn't want to pedal any more than needed. All that work and money already, I'd sure go with a little more work for 0.3 ohms instead of 0.8 ohms.

Grab a stepper motor and do some tests for a great example of "Where did all the power go?"  Many will put 10ma into a 12V battery with a finger flick, but only put 30ma into the battery at 1100RPMs.


Go back to the formulas for number of turns.

More RPMs means less turns (and less resistance). Thats why I mentioned different sprockets may help?

The coils are not solidly mounted now? The air gap will end up smaller later? That will help at least a little too.


Talk of capacitors comes up from time to time. Most people decided they won't help for any standard configuration.  I believe BIG caps could help some stepper generators in some situations, but honestly gave up on steppers before I tried it in a realistic form. Like so what if it gained 20%, from 50ma to 60ma... oh boy. :/

« Last Edit: December 09, 2005, 12:37:32 AM by ghurd »
www.ghurd.info<<<-----Information on my Controller

willib

  • SuperHero Member
  • ******
  • Posts: 2414
  • Country: us
impressive differences
« Reply #34 on: December 09, 2005, 08:19:41 AM »
air gap is 0.468 and the coil width is 0.390 that leaves .039" per side..

you are right , a triple wound coil isnt THAT  much harder, and the payoff is much greater.

i've did some calculating :

with  six single wound coils per phase  @ 500 rpm , lol bear with me,with one ohm load


http://www.otherpower.com/images/scimages/2965/rpm_per_volt.jpg


there would be 759 W going to losses in the coils & only 253 w going to the load. (126W/coil)


with double wound coils 438 W is lost in the coils BUT 587 W goes to the load, one ohm .& (73W/coil)


Now with the triple wound coils only 156W is dissipated in the coils per phase and 521W goes to the load. (26W/coil)

quite impressive differences..

« Last Edit: December 09, 2005, 08:19:41 AM by willib »
Carpe Ventum (Seize the Wind)

willib

  • SuperHero Member
  • ******
  • Posts: 2414
  • Country: us
Re: question ???
« Reply #35 on: December 09, 2005, 08:23:32 AM »
looking forward to hearing further results.
« Last Edit: December 09, 2005, 08:23:32 AM by willib »
Carpe Ventum (Seize the Wind)

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: impressive differences
« Reply #36 on: December 09, 2005, 10:05:04 AM »
Now try to built a tiny alt.

Like 2.14 gazillion turns, in a 0.3 air gap, and legs 0.3 wide.

I may give up on my tiny stuff.


For Peter, in the US 2.14 gazillion equals about 100 less than infinity. ;)

« Last Edit: December 09, 2005, 10:05:04 AM by ghurd »
www.ghurd.info<<<-----Information on my Controller

willib

  • SuperHero Member
  • ******
  • Posts: 2414
  • Country: us
coil production
« Reply #37 on: December 10, 2005, 02:22:04 PM »
Looks like i should get going on coil production, i've decided to go with the triple wound coils , 0.05ohms /coil and 0.3 ohms /phase..but i havnt tried 4 strands yet. lol
« Last Edit: December 10, 2005, 02:22:04 PM by willib »
Carpe Ventum (Seize the Wind)

dinges

  • Hero Member
  • *****
  • Posts: 1294
  • Country: nl
Re: question ???
« Reply #38 on: December 10, 2005, 02:51:13 PM »
You're right about the steppers. With my small ones (ex 5 1/4" floppydrives) I can't get more than 22mA, no matter how fast I turn. Voltage goes way up, though.


The reason here is not resistance, but inductance: as you turn faster, the frequency of your AC increases. With higher frequency, the coils inside the stepper create more 'inductive reactance', a bit comparable to resistance, in that it limits your current.

Xl=2*pi*f*L


where Xl=inductive reactance (ohm)

f=frequency (Hz)

L=induction (Henry)


With axial flux generators without a core, your induction L is many times less than in a stepper, with its metal core, so reactive inductance is much less of an issue, whereas resistance isn't.


Willi, you may further reduce your (effective) resistance by wiring the coils in delta, as opposed to star; this reduces your internal resistance of the 3 coils by a factor 3. Made the calc yesterday evening, I wrongly assumed the difference was only 1.73;


My next projects will probably be made with two wires per coil, to reduce resistance. Another lesson learned from experience, when just listening to other people's advice could have had the same effect... I think that with more wires, you'll quickly get diminishing returns (you're reducing your coil resistance, but your other wiring & connecting resistances stay the same, unless you take measures for these too). At those high amps, a few tenths of an ohm quickly become losses of several volts, as I found out with the bike genny. Again, experience confirming what mr. Ohm had predicted. Another few lessons learned.


Good luck,


Peter,

The Netherlands.

« Last Edit: December 10, 2005, 02:51:13 PM by dinges »
“Research is what I'm doing when I don't know what I'm doing.” (W. von Braun)

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: question ???
« Reply #39 on: December 10, 2005, 03:36:36 PM »
Yes, but I think steppers show the resistance problems.


Old (1984?) Apple 5.25" drives have a dandy pancake stepper.

« Last Edit: December 10, 2005, 03:36:36 PM by ghurd »
www.ghurd.info<<<-----Information on my Controller