Pedal Powered PMA

[valterra]

Introduction

This will be my diary for a pedal-powered PMA I have been contemplating.  Feel free to learn from my failures and victories.  And most of all, please comment extensively so I can avoid making any major (read: expensive) mistakes.

[dinges]

It's an interesting project, I've finished one like it myself.

http://www.fieldlines.com/story/2005/12/10/163646/91

http://www.anotherpower.com/gallery/exercise-bike-generator

Good luck!

[valterra]

I've got MiniGen on the way, so I can see a working PMA up close and personal.  Thanks Ed.

Looks to me (not having had a chance to actually handle one yet) that this is designed so that the magnets rotate around the coil.  This is totally different than the dual-rotor designs I've see and considered using for my project where the magnets pass by the coil.

Question:  Are there dis/advantages to this design?

[valterra]

I have learned on this site that voltage is determined as the product of the number of poles and turns, the rpm, and the strength and size of the magnets.

I've also read that current is determined by the size of the wire, but that's as much as I know.

Since a coil of larger with few turns would take up the same physical space as a coil of small wire with many turns, it seems that the wattage rating of a particular PMA is relatively fixed.

That is, since Watts = Volts x Amps, if you increase Voltage while reducing Amperage, your wattage level stays about the same.  The only variable in the equation (once the device is built) is RPM.

If resistance in the wire is a fixed value, then wattage (voltage and amperage) would increase as the RPMs do. (duh)  But where is the base value of the Amperage determined?

If I design a PMA that produces, for example, 1.5 volts per revolution per second, what kind of formula can I use to determine the base amperage?  In other words, how many A or mA per revolution per second?

Your comments welcome, as always!

[valterra]

I have seen these!  They are posts that encouraged me to try this.  My design (more in a later post) will hopefully use the sprocket from a multi-speed bike, so I can get that back wheel turning very quickly!

[Ungrounded Lightning Rod]

I've also read that current is determined by the size of the wire, but that's as much as I know.

In a battery-charging application with a rectified alternator, neglecting capacitive and inductive effects (which are minor compared to resistance in the designs and at the frequencies we're using), current is determined by:

 - the generated voltage

 - minus the sum of the battery charge voltage and the diode voltage drops (two diodes in series)

 - divided by the resistance.

Most of the resistance is the resistance of the wires in the coils.  (The other major components are the resistance of the wiring from the mill to the battery - which is enough to matter a lot - and the internal resistance of the battery itself - which is small unless you've just been charging or discharging heavily.)

Current is limited in two ways:

 - The amount you can drive at a given RPM is directly limited by the resistances.

 - The amount you WANT to drive is limited by the heating of the wire.  (You don't want to burn up your mill, so you set it to furl before it is driving so much current that it fries.)

More turns of thicker wire, higher voltage, higher resistance, and lower current.  Fewer turns of thinner wire, lower voltage, lower resistance, higher current.  Figuring the curent you'll get at a given RPM is complicated slightly by the fixed voltage of the battery.

What matters for current limiting to avoid burnout is the current density in the metal of the wire.  So you can get the same amount of ENERGY (at the alternator output) from a given amount of copper in the alternator, whether it's wound as a lot of turns of thin wire (for high voltage and low current) or a few of thick wire (for low voltage and high current).

(Losses in the transmission wiring go up with the square of the current times the resistance, so you need four times as much copper in the wires to get the same percentage losses at twice the current and half the voltage  This means you'd prefer to go with high voltage and low current to keep expenses low.  But you have to make this tradeoff by selecting your battery bank voltage.)

[valterra]

Okay,  so even though resistance is higher, a high voltage, low current design is more efficient.

That jives with the explanation I always heard about why power line voltage is so high.

Problem is, there aren't  any 110v batteries commonly available.  So the 12 (or 24 or 48) volts of our batteries forces us to use a higher current, lower voltagw.

All conversion losses aside, that means a 120v 1A line would need to be converted to 12v 10A in order to be stored.  (I know charging voltage is a little more than 12v, but for the sake of discussion let's  keep it simple.)

And regarding the fear of burning it up, I am thinking it's better to go for the higher volts and do the conversion to higher current somewhere down the line.

So i'd conclude that it seems best to go for a higher starting voltage, so you can do more with it down the line.

Am I getting all of it right?

[willib]

"More turns of thicker wire, higher voltage, higher resistance, and lower current.  Fewer turns of thinner wire, lower voltage, lower resistance, higher current.  Figuring the curent you'll get at a given RPM is complicated slightly by the fixed voltage of the battery."

ULR you have to be clear when explaining things

i have no idea what you are trying to say in that first sentence ?

[valterra]

any thoughts about the magnts spinning around coil vs. passing by it thoughts a few posts ago?

[willib]

i dont understand what you mean either

maybe some punctuation will help

[valterra]

Silly - you just had to look back a few posts, like I said.  Here is the post again for those without scroll mice:

I've got MiniGen on the way, so I can see a working PMA up close and personal.  Thanks Ed.

Looks to me (not having had a chance to actually handle one yet) that this is designed so that the magnets rotate around the coil.  This is totally different than the dual-rotor designs I've see and considered using for my project, where the magnets pass by the coil.

Question:  Are there dis/advantages to this design?

In other words, a coil is shaped like a cylinder, and ring of magnets spins around the outside of it.  The varying magnetic fields move along the path of the wire in the coil.  In the other PMA designs, the fields pass over the top the coil(s).

See Here for pictures of the alternator I'm talking about.

[valterra]

Okay, I don't have a picture handy, so hopefully my explaination will work.

My PMA design basically looks like the ones on the otherpower.com projects page, except that the center hole of each rotor is bolted (with many small bolts) through the spoke holes in a bicycle hub.  So you're replacing spokes and a tire with steel discs.

I've been desparately trying to find metal discs (no way I could do it myself without totally messing it up) and also trying to figure out just how wide the hole in the middle needs to be, etc. etc. and figuratively pulling my hair out in frustration.  The design looks pretty slick, but I just don't have the parts to do it.

Then I thought - am I re-inventing the wheel here?  Literally?   Wouldn't it be possible to leave the spokes and metal wheel intact, and simply place large magents where the tube used to be?

Think about it.  The bicycle remains intact (albeit with no rear tire) and the tire is replaced with magnets.  You mount the bike on some sort of frame that holds the back wheel up in the air.  You then mount coils at appropriate places on that stationary mounting, and you have a PMA.  Seems to me that'd be a lot easier.

The biggest problem that I see is that you don't have the benefits of a dual-rotor design.  But you can put many magnets on that wheel, and it IS metal, so it essentially simulates a single-rotor design.

Am I all wet here?  I could have this done in a day with this design - no machining or power tools needed.  Maybe I should just TRY it.  Ha.  Imagine that.  :-)  

[Dennisd]

Sounds like a motor conversion, where magnets are attached to the old rotor of a used electric motor. The main problem would be getting metal behind the magnet to direct the flux outward, and a small enough air gap. However, a tenspeed bike would allow fairly high RPM, so maybe it  would work. Not that big a sacifice to try it, as the magnets could be removed and reused if the power output wasn't there. I read somewhere that a normal person can only maintain about 100 watts output anyways.  

[valterra]

So the metal wheel is actually a detriment?  I conceived this idea with the thought that a steel backing was a GOOD thing.  

Or when you mentioned directing the flux outward, ar you saying that dual-rotor design are better, but this single will work?

[valterra]

Hopefully this makes sense. Think of your basic dual rotor job, But with a sprocket on one side...  done correctly, this would replace the back wheel of a bicycle.  A chain connects to it just like a regular bike.

Am I off my rocker?  Would the chain bind because of magnets on the other side of a thick steel plate?  Would it be too hard to turn by foot?  

The reason for the idea of rear wheel replacement is that I could get high rpms with the right gears.  If you've ever turned a bike upside down and turned the pedal with your hand, you know what I mean.

[Dennisd]

Without having built one, my understanding of the theory is that steel is required behind the magnets,and a single rotor will work, but not as good as dual rotor. another concern is magnetically heating the frame of the bike?