Mostly Permanent Magnet Motor with minimal Input Power

[gotoluc]

if you used two cube magnets on top of each other to double their length this way you would also achieve a double distance between the coil and the side bars may show the effect if any.

If you increase the air gap (distance) between the outer surface of the coil and outer core, the result will be a drop in pull force, even though you double the magnets. As the outer cores and coil can do as much work as the inner core and coil. So the closer to zero air gap will give the best results.

I assume that the improvement in coil inductance (when you use a side bar to bridge the outer like poles, say South poles) comes from removing the "fringe" South pole flux from the coil area and direct them into the outside bar. I say this because seemingly you indeed close the magnetic circuit by using an iron side-bar but the magnets to be bridged being like poles they cannot close to any other pole in the bar because at the other end of the bar the same like pole enters the bar from the other end magnet.

The increase in the coils inductance comes from adding iron core close the the surface area of the coil. As you should know, the closer the copper wire to the iron the more the inductance gain.
I cannot close the loop between the inner and outer core (like a transformer) as this would short out the north and south flux and the coil would barely move. I know this because I tried it. I thought it would boost the inductance big time, and it did but the coil also barely moved.

So I assume that the increase in pulling force does come from the increased coil inductance and this inductance increase comes about by applying more flux from the permanent magnets, shifting the core's working point towards a higher permeability area of the B/H curve. Of course as you also mentioned in the video, there is an upper limit in increasing the flux by using stronger and stronger magnets: this upper limit is set by core saturation.

You may think that the extra pull force is from an increase in Inductance but that is not correct. I'm sure there is a small increase caused by the inductance boost but the most of the extra pull force is from the outer coils opposite pole working on the outer core and magnet pole just like the inner coils pole working on the inner core and magnet pole. I can prove this by making the coil move the same just using outer cores and magnets with no inner core and magnets. The only thing it's just the opposite pole.
I don't know if you follow me. I've been trying to tell about this before but people are not picking up on it.
A coil produces the opposite pole half way through its thickness. Just like a one inch thick magnet will have 1/2 inch of North and 1/2 inch of South.  Well, a coil that's wound one inch thick will also do the same.
I have experimented with this some years back and I though this was known.  Is it not known this way?
Do you understand what I'm trying to explain?

The possible ultimate core and magnet shape for your setup would be to use a cylinder  core (say a ferrit rod) in the middle (on which also a cylinder coil would slide) and use a ring magnet at both core ends with its ID matched to core's OD and then use a soft iron tube (with certain wall thickness) outside instead of the bars to connect the ring magnets' outer like poles. No stray flux would escape from such setup.

I don't know about that. How can a cylinder work better the what I just demonstrated?... also, doesn't a ferrite loose inductance when you apply magnet flux?

Thank you Gyula for always asking important questions and sharing your knowledge.

Luc

[gotoluc]

Hi gotoluc!

Thanks for sharing your very valuable experiments.

I see very brillant your use of bacefm as an alternative drive synchronized with the moving of the coil.

You can use imho this as a very interesting piston motor; a "motor" indeed. But true you can drive a generator and try to find the limit of efficiency of the system.

But before to do so, could you scientifically and rigorously measuring the power output of it? Could you add a pouly with a baril, a wire, a mass, and measure the time your motor will pull the mass up to 1 meter? This giving us the effective mechanical output power to compare to the electrical power. A graphic of the evolution of the mechanical power in function of the electrical power input would be a very must; imho.

Cheer, Khwartz.

Thanks for your positive comment.

The coil weight is 100 grams. I can place it vertically (against gravity), power it and calculate the amount of mm it moved up in x amount of time using x amount of watt.

Would that work?

Or much easier for me is a capacitive (joule) discharge and tell you how many mm it moved up.

Luc

[gyulasun]

Hi Luc,

If you increase the air gap (distance) between the outer surface of the coil and outer core, the result will be a drop in pull force, even though you double the magnets. As the outer cores and coil can do as much work as the inner core and coil. So the closer to zero air gap will give the best results. 

Okay, I accept that and although it is irrevelant now, I think you have tested how the coil inductance changes when you increase the air gap and found the inductance increases in a lesser extent compared to a previously smaller air gap? I mean you measure 867 mH for the coil with the single magnets at the ends (when there is no bridging bar between the outer like poles) and then you apply the bridge and get 942 mH.
Then you double the air gap for the bridge by using double magnets (i.e. doubling their thickness) and you measure also an increase in inductance with respect to the no bridge 867 mH value but now this measured value would not reach the 942 mH value received for the previous smaller air gap case, so this is what I figured.
Because you measured the pulling forces for the different cases, obviously the size of the air gap has a much stronger influence on the force than the increase in inductance does.

I don't know if you follow me. I've been trying to tell about this before but people are not picking up on it.

Well I follow you but this time I did not revise your years ago tests on this setup, sorry. Yes, a coil will also have a Bloch wall in its middle part just like a permanent magnet does, no doubt on that.

I don't know about that. How can a cylinder work better the what I just demonstrated?... also, doesn't a ferrite loose inductance when you apply magnet flux?

Well, a cylinder coil in itself would not work better than your rectangular coil (assuming equal inductance and number of turns) but I meant also using ring magnets instead of the cube magnets. I simply thought of occupying all the space around a cylinder coil by having covered it all around by a ferromagnetic tube to replace the bars (in fact you would have an infinite number of "bars" from the generatrix of the cylinder tube), this way the opposite poles could be utilized the most from the coil.
Now I looked for radially magnetized ring magnets (i.e. their ID is say North all inside and their OD is South all around outside) which would be needed for my suggestion but unfortunately such is not manufactured...  can only assembled from arc magnets like is shown here: http://www.supermagnetman.net/product_info.php?products_id=380  or also could be assembled from say smaller cube magnets also wrapped up in an enclosure. Understand now what "cylinder" setup I mean? Of course such setup would cost more to assemble than your present one and I also prefer initial testings with a cheap solution, I just indicated an 'ultimate' setup in this respect.
Regarding a ferrite loosing inductance when influenced by (strong) outside magnets, yes it normally does, mainly when cross section area is small versus the flux strength, I mentioned ferrite as a first straigthforward thought for cylinder shaped cores...

rgds, Gyula

[gotoluc]

Hi Luc,

Okay, I accept that and although it is irrevelant now, I think you have tested how the coil inductance changes when you increase the air gap and found the inductance increases in a lesser extent compared to a previously smaller air gap? I mean you measure 867 mH for the coil with the single magnets at the ends (when there is no bridging bar between the outer like poles) and then you apply the bridge and get 942 mH.
Then you double the air gap for the bridge by using double magnets (i.e. doubling their thickness) and you measure also an increase in inductance with respect to the no bridge 867 mH value but now this measured value would not reach the 942 mH value received for the previous smaller air gap case, so this is what I figured.
Because you measured the pulling forces for the different cases, obviously the size of the air gap has a much stronger influence on the force than the increase in inductance does.

Well I follow you but this time I did not revise your years ago tests on this setup, sorry. Yes, a coil will also have a Bloch wall in its middle part just like a permanent magnet does, no doubt on that.

Well, a cylinder coil in itself would not work better than your rectangular coil (assuming equal inductance and number of turns) but I meant also using ring magnets instead of the cube magnets. I simply thought of occupying all the space around a cylinder coil by having covered it all around by a ferromagnetic tube to replace the bars (in fact you would have an infinite number of "bars" from the generatrix of the cylinder tube), this way the opposite poles could be utilized the most from the coil.
Now I looked for radially magnetized ring magnets (i.e. their ID is say North all inside and their OD is South all around outside) which would be needed for my suggestion but unfortunately such is not manufactured...  can only assembled from arc magnets like is shown here: http://www.supermagnetman.net/product_info.php?products_id=380  or also could be assembled from say smaller cube magnets also wrapped up in an enclosure. Understand now what "cylinder" setup I mean? Of course such setup would cost more to assemble than your present one and I also prefer initial testings with a cheap solution, I just indicated an 'ultimate' setup in this respect.
Regarding a ferrite loosing inductance when influenced by (strong) outside magnets, yes it normally does, mainly when cross section area is small versus the flux strength, I mentioned ferrite as a first straigthforward thought for cylinder shaped cores...

rgds, Gyula

Hi Gyula,

from what I read above you're understanding everything perfectly well.

I also now agree that if you could find a radially magnetized ring magnet then the cylinder setup would probably work best. However, we also need to keep in mind that the coil would need to be held by guides if one wants to use the mechanical movement of it. So the outside cylinder would need to be two halves with a slit of the thickness of the material holding on to the coil.

To improve on outer surface area loses what I was thinking of doing for my next super build is making the center core 1/2" to 3/4" thick but 6 inches wide. That way 90% of the outside coil surface can be covered and converted to power.

The only thing I don't know about is the ideal coil wire size, length, amount of turns and resistance (combination of all) to obtain the most pull per watts.
If you know of a way to calculate that would be of great help.

Thanks for sharing.

Luc

[Khwartz]

Thanks for your positive comment.

The coil weight is 100 grams. I can place it vertically (against gravity), power it and calculate the amount of mm it moved up in x amount of time using x amount of watt.

Would that work?

Or much easier for me is a capacitive (joule) discharge and tell you how many mm it moved up.

Luc

You're wellcome! Luc.

I have thought this possibility to put vertical the axis and work against grativity but if it very easy to realize, it is not so easy to have an enough accurate time measurement on so short period of time.

The formula any way is: P[W] = M[kg] * g[m/s2] * L[m] / T.

Cheer.