Electromagnet power transfer question.

[gyulasun]

If I understand it right the flyback shouldn't effect the over idea of my design but could be adapted to improve over all functionality?

Right,  it does not affect anything but helps increase overall efficiency by regaining some part of the input power you already furnished in,  this way the input power consumption can be reduced.  For this early time in the development, no need for it.

Gyula

[gyulasun]

Is there certain shapes of magnetic cores that are more efficient? For example, do magnetic fields travel better through a curve or a sharp 90 degree angle or both equally? 

I am not aware of any such problem this may cause I do think it can be disregarded. Flux follows its guide by all means, be it curved or rectangular.

In config4, I just want to double check that by simple turning the wire in opposite direction on C and D will combine the two poles to produce one pulse in the output coil? 

Not sure on this question but assume you mean the turning sense of windings on either C or D coils can be made to get the sum of the two induced voltages, like you connect two 12V batteries to get a single 24V voltage source?

I would assume it would be better to have the core be as small as possible to reduce resistance? Would I potentially run into problems if the input [primary] coil is right next to [but divided] the output [secondary] coil, or should I given them a little space? 

If you mean if it is a problem for the input and output coils to magnetically couple into each other due to being very close to each other, then letting a certain space is better.  What do you mean by "to reduce resictance"  wrt the core size / smallness?? What resistance?

Gyula

[Nali2001]

Right,  it does not affect anything but helps increase overall efficiency by regaining some part of the input power you already furnished in,  this way the input power consumption can be reduced.  For this early time in the development, no need for it.

Gyula

Hi Gyula,
Well I have to disagree with you on this one.
Capturing the 'flyback' is not at all without consequences.
In this video you see a core with a coil being powered with a 50%duty dc square wave.
http://www.krystyna.nl/Machine/ClampCore.wmv

(Also note that the used circuit has no diode on the fet to protect if from the flyback spikes.)
But anyway, you can see that without the flyback capture the core is experiencing 'soft attraction'. Since the core can "relax" or loose its magnetism between the 50% off cycles of the square wave. Well, then in the video the diode and bulb-load is attached and the light you see is from the captured flyback. But now there is a change in the core to core attraction. You can now see that the cores are dead solid attracted which means that although the coil is only on for 50% of the cycle the cores are experiencing a 100% full 'on' magnetic field. What is happening is that the capture diode and bulb is preventing the core to 'relax' or in other words prevent to core to naturally get back into a near zero magnetic level in between each dc pulse. Well you understand that this effect is only really working in a close looped core condition. But other then that you might say 'so what' when the core is full magnetic all the time.. But one must understand that this now eliminates the core to be used as a dc pulse transformer since there is now (almost) no change in field strength anymore. And the shown systems above are just that.., dc transformers. I have found that flyback capturing is not always a plus. Since in close looped single polarity pulse situations it just "locks up" the core. Which is not what you want in these meg like systems... You want the core to self reset its level of magnetism in between each dc pulse. Since it is obviously not wanted that the core still has 90% of the remaining magnetism from the last pulse, since then there will only be 10% field change at max.

(Tim, these things are for example what I mean when I say 'there is more to it' and 'not that simple')

Regards,
Steven

[nwman]

Steven,

I know its all over my head and really complex. In-spite of doing back to school for a few years to learn this specific tech all I can do is push in a direction and learn everything I can as I go along. I also wish not to spend any money testing ideas that I can simply learn the results and ideas from others [that are willing to share]. Thanks by the way.

I just wanted to get something on here that I heard from a guy that works at the University in my town. I'm not sure of his credentials but he works with people at the U that are working on things like this. He was trying to tell me something about how a large factor that gets over looked is the gases that are around the magnets while they run can effect the performance. Talking about "a lot to it"! I just had to nod my head and say "ok... sure". I have never heard of this.

Tim

[gyulasun]

Hi Steven,

Thank you for bringing this up,  unfortunately I forgot to consider the cores are closed in Config4 of course and the created current by the flyback pulse makes a flux stream that magnetizes the core in the off input periodes with a magnetic polarity that adds to flux made by the on time part periode of the input current, hence a continuous attraction.  Sorry for this, in my mind I concentrated on flyback current capture of a 'normal' pulse motor where the cores are mainly open, not closed.

I think if you modify the flyback capture by first charging up a capacitor through the diode by the flyback pulse, then you discharge the capacitor through a second switch to the lamp and this discharge would be made under the off time of the input square wave, (when the first switch is off) then you could reduce the full magnetization of the core by the flyback current during the off time  because in this case the magnetizing current will be the charging current going into the capacitor only ( which is an exponentially decreasing current so its energy content is also decreasing, unlike to the full 'normal' flyback current).  I have not tried this yet, this is theory at the moment, and I do not feel a very good solution for the problem though.

Thanks for the excellent video too.

Regards,  Gyula