Thane Heins BI-TOROID TRANSFORMER

[broli]

Broli,

This seems a difficult one in Femm.
Below my interpretation, but it will not run, because Femm does not allow the touching of two circles. It such case it can not build up a mesh network of points where it calculates the flux.

Would a similar setup with 2 rectangular cores work?

Perhaps someone can predraw it in autocad so it can be imported in FEMM. The reason why I picked a toroid in vizimag is because I can draw a simple toroidal core all around it. With a rectangular core I can't do this, perhaps in FEMM you can. Or if someone else could make it in autocad so it can be imported in FEMM.

The discovery is that if you do have a coil all around the core that it seems to take more current to neutralize the primary than a conventional transformer tries to do.

[SchubertReijiMaigo]

Thank teslaalset, generally after a good night, we have better mind ^^

The goal of any transformer is to power heavy load, you don't use a 3 KVA transfomer to power a 40 watts light bulb, in any transfomer you have some reactive current made by the primary to the main line. Reactive current don't consumme energy but add some intensity and heat the wire (that's what electrical company bills the reactive current for the industriels). Both method is good high Z or high R and saturation... The main deal here is to create the magnetic diode. I have even thinked to add 2 small (0.1 or even 0.05 mm) symetricaly on the outer toroid, yes you sacrify some inductance value but you have perfect linear B/H curve and you can carrying large current. The permeability is stable the whole sinus and virtually any load, the magnetic diode is perfect !!!
With FEMM you cannot simulate BEMF you can only simulate DC fields and and each coils separately, that's why when we try to play with, we have some weird flux line and comportement, like previously: the flux run outside the iron block...

[broli]

Some more images.

The air gap causes more flux (about 15 times) to be concentrated on oneside of the core. This should cause a bigger induced voltage in the big troidal coil. Then I started to raise the amperage in the big coil until the field died in the primary. The resulting amperage is seen in the last image. Ofcourse this simulation doesn't tell us what the generated voltage is. If the induced voltage would go down by a factor of 7 compared to the input voltage we end up were we started.

[teslaalset]

Thank teslaalset, generally after a good night, we have better mind ^^

The goal of any transformer is to power heavy load, you don't use a 3 KVA transfomer to power a 40 watts light bulb, in any transfomer you have some reactive current made by the primary to the main line. Reactive current don't consumme energy but add some intensity and heat the wire (that's what electrical company bills the reactive current for the industriels). Both method is good high Z or high R and saturation... The main deal here is to create the magnetic diode. I have even thinked to add 2 small (0.1 or even 0.05 mm) symetricaly on the outer toroid, yes you sacrify some inductance value but you have perfect linear B/H curve and you can carrying large current. The permeability is stable the whole sinus and virtually any load, the magnetic diode is perfect !!!
With FEMM you cannot simulate BEMF you can only simulate DC fields and and each coils separately, that's why when we try to play with, we have some weird flux line and comportement, like previously: the flux run outside the iron block...

SchubertReijiMaigo,

I did some simulations in FEMM with tiny air gaps, but FEMM doesn't like very small gaps, due to the method of using mesh network.
If e.g. you use a 0.05 mm gap, it simply will not run, only when you have a very powerfull PC.
One way to get around this is to define a core material in the FEMM materials library that has a non-linear curve of a gaped core.
The other problem is that FEMM doesn't cope with full cycle B-H curves.
I am trying to find a way to get around that as well.

Did you encounter the same issues with FEMM?
If you have a work around for these issues, please let me know.

[SchubertReijiMaigo]

Sorry for the two consecutive post,

@ Broli I have the same problem FEMM can't modelize correctly the toroid...
I don't know how to use autocad unfortunately, but I post also a "Easy winding industrial design" of the BTT. This BTT have two demi cores and a two demi I at the center. The I can be gapped or not, in my version, he have a gap of 0.2mm. The machine can wind more easly and make industrial production avalable of the BTT. Enjoy !!!