The Ossie motor

[Augen]

Hi Gyula,

I'm sorry. I forgot to mention that I was talking about electromagnets.

So here are the formulas:

F=B*B*A/(2*u) -> using fix u and max A (25mm diameter magnet -> 25mm inner diameter electromagnet) somehow we have to maximize B. Right?

??=I*N=H*l -> H=I*N/l and B=u*H -> using fix u we have to maximize H

H=I*N/l -> we have to maximize the amperturns in the possible shortest length (l). But if the inner and outer diameter is fix, by increasing N you increase l so N/l won't change. Or just a little. But by increasing N, you increase "R" so "I" will be decreased. Bottom line, by increasing N and l, you decrease H. Am I right?

EL=L*I^2/2 -> By increasing N and l (fix ID and OD) H will by less, but the store energy will be more.
So my question is: if you have for example 100m long wire, how can you make the possible strongest electromagnet from it?

Thanks, Augen.

[Jimboot]

Very busy at work atm. However I have wound the drive coil off to half. Initial tests indicate stronger effect. Current draw fluctuating a lot. More tuning of reeds & testing of rotors required. Thanks @Augen

[gyulasun]

Hi Gyula,

I'm sorry. I forgot to mention that I was talking about electromagnets.

So here are the formulas:

F=B*B*A/(2*u) -> using fix u and max A (25mm diameter magnet -> 25mm inner diameter electromagnet) somehow we have to maximize B. Right?

??=I*N=H*l -> H=I*N/l and B=u*H -> using fix u we have to maximize H

H=I*N/l -> we have to maximize the amperturns in the possible shortest length (l). But if the inner and outer diameter is fix, by increasing N you increase l so N/l won't change. Or just a little. But by increasing N, you increase "R" so "I" will be decreased. Bottom line, by increasing N and l, you decrease H. Am I right?

EL=L*I^2/2 -> By increasing N and l (fix ID and OD) H will by less, but the store energy will be more.
So my question is: if you have for example 100m long wire, how can you make the possible strongest electromagnet from it?

Thanks, Augen.

Hi Augen,

I am afraid your last question could be answered from a good 3D electromagnetic optimizer software like for instance Infolytica 3D or Maxwell 3D but their price is prohibitive, unfortunately.
Regarding your discussion of the formulas above, I notice the followings:

1) I do not think you would have to confine yourself with fixing the inner and outer diameters because then you cannot find an agreeable tradeoff. So I think you are free to choose inner diameter for the "dream" coil from 0.5*OD to 0.9*OD of the permanent magnet so that the outer diameter of the coil can range higher than the permanent magnet OD diameter, depending on the wire diameter and on the number of turns.  IF you do not choose this relationship for the coil ID, then you fully lose your freedom on selecting thicker wires, which is important to reduce copper loss. I do not think you strictly have to observe coil ID=magnet OD, ok? It is unnecessary.

2) Now if you agree with the above reasonings you can find a length for the coil which of course should be a small value so that the flux created should find the smallest distance in the air to travel. So this length in practice could be chosen as between ,say, 2mm to 10mm whatever the wire diameter and the number of layers may dictate, considering and confronting the notice on the wire diameter above at point 1.

Quote from you: "Bottom line, by increasing N and l, you decrease H. Am I right?"    You are right if you confine yourself for not letting coil ID and OD be varying.

Finally, there is the so called Brooks coil geometry which gives the maximum inductance possible for a given wire length, see here:
http://www.nessengr.com/techdata/brooks/brooks.html 

Before you say 'Heureka'  I do think we have to combine the above reasonings with the Brooks coil geometry conditions and also make a tradeoff because the condition for the Brooks geometry is not a strict or strong 'function' to  observe.
The Brooks coil calculator does not include wire diameter unfortunately. And be aware it uses radius of coil, not diameter.

I found a useful air cored multilayer coil calculator which asks for only a inductance value you wish to get and gives a tabelle of useful data on several wire gauge, ID, OD, and number of turns which all would give the same inductance value, only the coil's geometry changes. here it is:
http://www.colomar.com/Shavano/inductor_info.html  It is offered for audio crossover coils.  I mention this only to widen our general knowledge on air core coils calculations, not for further 'confusion'. 

I hope all this 'ramblings' will help you (and the others) to arrive at a "dream" coil for the job.

Gyula

[Jimboot]

http://www.youtube.com/watch?v=dkI7cMszQcI coil wound back to half. Sorry about the voice. Went to the football (aussie rules). I'm now running it on a single d cell for the last 28 hours at around 900 rpm V has dropped .03 in 24 hours. But I'm sure I can tune it better given the trace over the coils I'm seeing. Current draw at .01. V @ 1.23 .

[gyulasun]

Hi Jimboot,

If I observed correctly your present coil, you removed about half of the original windings from the bobbin diameter-wise BUT I think Augen and I were talking about coils with less lengths which does not mean reducing in the radius direction but in the coil's axle direction, reducing the coil's thickness, not its diameter.
What I tried to suggest to Augen is that to make a stronger coil you would wish to choose a coil ID of 0.6-0.7OD of the magnet, i.e. if you have OD=25mm magnet then you could make a coil that has an ID=0.6*25=15mm or around this value. And the thickness (or the length) of this coil would be any size between 4 to 8mm. Experimentation is needed here, especially for the starting ID of the coil. And comparison of the original coil to the modified coil is valid only if you use the same length of wire for the modification, ok?  This means that the modified coil will have a higher OD in the end than that of the original coil, even if you observe starting with less ID as part of the modification and also make the coil thinner.

If you really find your present modified coil shown in the video is better than the unmodified, then the improvement may also come from the higher current your coil draws due to its half as much DC resistance. And this higher current is the peak current value what you cannot really see on your analog meter that always show avarege values. (Peak current could be seen on a scope, probing the voltage drop across a 1 or 10 Ohm series resistor.)

rgds,  Gyula