Hello PaulLowran

JackH,
I think you're correct, as other people have told me the same thing in private message.  Hopefully with some lucky, cross my fingers, I can pull this off.  ...  That's pretty bad they sent you a broken core. How do they expect to run a business that way?


Rob,
Man I hope you get those cores, really. From what others have said Metglas won't ship any free samples outside the U.S.  I just replied to the guy at Metglas. He said the core I selected was not available. So I just requested the same core material Naudin used. We'll see what happens. He'll probably reply no later than tomorrow and I'll let you know.  IMHO they don't give out too many samples so I could use every bit of advantage. Just don't want them to get suspicious.
Will it work? Yes, most likely if you follow Naudin's exact design, but just as with Naudin's it will be self-running. It's very important that you follow Naudin's exact process of preparing the carbon resistor.

I have several untested designs that precisely show how to close the loop ... if I can just get a hold of an appropriate amorphous & nanocrystalline core.

Paul Lowrance

Hi Paul,
Each of the two output coils produce a sinusoidal output, that may be out of phase with each other.
If they are in phase then you could couple the outputs together and use rectification to directly drive two more MEGs or one larger MEG.
That is provided the output voltage can be reduced to a managable level.
In effect you have 3 MEGs with the output from the 4 coils of the output stage.

The only issue I can see is that the input coils of the 2 MEGs in the output stage may cause some unwanted reaction to the driver stage.

If you keep to the same size MEGs it may be possible for them all to operate at the same frequency, JLN mentioned you need to tune the setup so the output produces a good sine wave with the maximum applitude.
Who knows, with 3 stages you could have a useable output from 4 MEGs.

Another thing that could be done would be to place the output resistors (insulated) into a 1L container of water in a polystyrene box and measure the temperature rise over a period of time and calculate the energy transfered in the form of heat.


Regards

Rob

Hi Rob,

I'll try to clarify. As you know it's the net field within the core that counts. So if the permeability is 100,000 and your apply 0.1 oersted (equates to 0.1 gauss) then the net field within the core would be 100000 * 0.1 = 10000 gauss (1 T).  That's our reference. If you reverse the field then it would be -1 T.

So in all magnetic materials indeed there are always temperature fluxations caused by MCE. Normally these temperature changes are 10's of thousands of a Celsius.  ... I'll create some images and describe this process at my peswiki project and post the link in an hour of so.  This will show exactly why smaller domains equate to higher PE.

Please refer to MEMM thread for a continuation of my reply. Since this is JackH's thread I don't want to disrespect his great work in anyway by distracting or posting other material here   ->

http://www.overunity.com/index.php/topic,1565.msg14481.html#msg14481

Regards,
Paul Lowrance

Hi Jack,

I have not built a bifilar coil the size you have referred to but I did it in one layer and in smaller size some years ago.
And I did inductance measurements to find that I got at least 3 - 3.5  times as much self inductance
from the bifilar arrangement than from a non bifilar one. Let me show you a link because it confirms
pretty nicely what I found in my tinkering:
http://www.tesla-coil-builder.com/bifilar_electromagnet.htm

So the same length of wire is used on both nails. The two coils are connected in series, so that the end of (any) one coil is tied to the start of the other coil and the start of the first coil gives one connection and the end of the other coil gives the other connection, as you see in the nails' photo.

Now if you wish to do this in six layers then I think you could wind it first in two layers and figure out which endings or begginings of the second layer are to be connected to that of the first layers'.  If you could use an inductance meter to check the increase in self inductance, then you could decide how to connect the wires of the second layer to that of the first layer, and so on, always to get the increase in the self inductance (I mean by checking the connections right after the second layer is done you cannot fail later which direction you continue winding the parallel wires).

Recently I found an old patent where two coils are placed next to each other in a mirror image symmetrical relation on the same core and they are connected in parallel. Unfortunately I have had no time to check if this arrangement is any better in terms of inductance (hence flux) than a similar bifilar arrangement.  See US patent 4806834 and maybe the simplest to use this link to read it:
http://www.freepatentsonline.com/4806834.pdf

The patent includes a kind of explanation on decreasing the time constant of this arrangement and
increasing the flux if compared to a single coil or a pair of coils in series within the same coil volume.
Nevertheless, if this arrangement is able to insure as much flux as your would-be bifilar six layer coil,
then it would be easier to construct.  It is sure that the time constant of the bifilar set up I refer to above
is higher than the set up in this patent due to the series connections of all the layers (DC copper
resistances add up).

Regards
Gyula

Hello Jack,

I've never wound a bifilar coil but I remembered to these sites, maybe it can help a bit.

http://www.opensourceenergy.org/txtlstvw.aspx?LstID=1e5e11a7-7363-4eb5-891e-6a5e73ab8c46
http://www.magnetricity.com/NeoG/Bifilar.php
http://en.wikipedia.org/wiki/Bifilar_coil

There are more types of bifilar coils. I think you need a non-caduceus one.

Thanks to all,

Hello sam,