Magnet Inductance Boost with Finemet nanocrystaline core

[wings]

Naudin test
http://jnaudin.free.fr/2SGen/indexen.htm#hysteresis

http://jnaudin.free.fr/2SGen/indexen.htm

http://jnaudin.free.fr/2SGen/images/inductive_conversion.pdf

[Kator01]

Hi Dave,

you are absolutely right, I just checked the BH-Curve of these finemet-core.
Now I might add some information concerning the current -reduction.

I have studied the old transductor-techniques ( 1930)  now for about one year including extensive practical testing for power-level up to 1,5 KW. I mention this so that your know that I am speaking from practical experience about the following process:
If you look at the attached pic which I extracted from the Fe SquareLoop-Core-brochure, I have added 2 red arrows which mark the level of a static pre-magnetisation - done either by approaching a magnet ( like Luc did)  or by a dc-current through an additional control-winding. Now the remaining magnetic loop oscillates between almost full inductance starting at the level marked by the arrows and low inductance in the right upper saturation-area. In such a way you have no reverse-magnetisation which takes place if you make the way around the full loop ( see the bottom--figure). Therefore you have no de-and remagnetisation-losses. You just use a smaller BH-Area of the curve. You just use a well known magnetic-amplifier-technique. I had the chance to visit a company here in germany which produces transductors ( part of a magamp ) and I was told the the losses using a normal transformer-lamination for 50 Hz for these devices is no more than 5 % of the maximum-power to be controlled. A 20 KW-transductor in the size of a small briefcase consumes 1000 W for controlling the power-out.
Using permagnet magnets just needs the kinetic energy for approaching the coil.

Regards

Kator01

[Kator01]

Hello all,

just a remark on naudins experiment : he did not magnetize the core symmetrically like Luc did. So the curves can not be compared. Secondly there is a difference pulsing the coil without a counter-permanent-magnet like in the muller-setup. Here one static and one dynamic magnetic field work against each other including dc-puls in the coil.
Third the coils in the muller-gen are solenoid-coils with both ends facing high magnetic resistance ( air) alternating to a state of very strong magnetic fieldstrengh in the airgap if the magnet fly by.
So this process is much more complicated in the muller-gen.

Kator01

[gotoluc]

Hi Luc
            I measured the increase in inductance in Square loop  cores about four
years ago when playing around with Steorns Orbo.Any way I did a quick test again
It was what I expected with the same core I used then.
26.5mH no magnetic bias 33.8mH with Magnetic Bias.With Square loop the increase
of inductance will remain until it is reset by a reverse bias.As you can see the increase is not as great as your tape wound core,but that is to be expected being made from a more exotic material.The answer to your question why is there a reduction in current when your core is biased,is because the inductance has increased therefore so has the impedance .When you bias your core as shown in your Vid, does the inductance remain high when you remove the bias magnet?

Regard Dave

Hi Dave, thanks for posting your additional comments and new test results.

I'll try again to see if I can get a small Inductance change in the FERROXCUBE.

Concerning the current drop when the core is biased. You believe there is an inductance gain! ... how interesting. Kator01 says there is no Inductance gain. I don't know what an Inductance gain would look like on the scope. So I started this topic to see if anyone has the answers. So far we have a split.

Not sure why you ask "does the inductance remain high when you remove the bias magnet?"

I think the answer is no but I show the differences in the video (with and without magnet)

Thanks for sharing

Luc

[gotoluc]

Hi Dave,

you are absolutely right, I just checked the BH-Curve of these finemet-core.
Now I might add some information concerning the current -reduction.

I have studied the old transductor-techniques ( 1930)  now for about one year including extensive practical testing for power-level up to 1,5 KW. I mention this so that your know that I am speaking from practical experience about the following process:
If you look at the attached pic which I extracted from the Fe SquareLoop-Core-brochure, I have added 2 red arrows which mark the level of a static pre-magnetisation - done either by approaching a magnet ( like Luc did)  or by a dc-current through an additional control-winding. Now the remaining magnetic loop oscillates between almost full inductance starting at the level marked by the arrows and low inductance in the right upper saturation-area. In such a way you have no reverse-magnetisation which takes place if you make the way around the full loop ( see the bottom--figure). Therefore you have no de-and remagnetisation-losses. You just use a smaller BH-Area of the curve. You just use a well known magnetic-amplifier-technique. I had the chance to visit a company here in germany which produces transductors ( part of a magamp ) and I was told the the losses using a normal transformer-lamination for 50 Hz for these devices is no more than 5 % of the maximum-power to be controlled. A 20 KW-transductor in the size of a small briefcase consumes 1000 W for controlling the power-out.
Using permagnet magnets just needs the kinetic energy for approaching the coil.

Regards

Kator01

Hi Kator01,

thanks for posting this additional information.

Regarding your explanation on the current drop when the core is biased with the magnet. Unfortunately I don't have any schooling!... so I can't grasp anything that would require this knowledge. What I'm trying to tell you is, after reading your post I don't understand it any better as to why the current drops then before. If you can explain it in a layman way maybe I'll get it.

What I have learned to date is all through physical experiments, trial and error.

ADDED
One thing I would like to add is, if we compare both scope shots, one is no magnet and the other is with magnet. You say there is no increase in Inductance when the magnet is added since the flyback spike at off time is flattened. However, I would like to bring to your attention that with no magnet the current is 230uA and with the magnet it drops to 80uA which is about 75% less current used. So one could expect a change in flyback output. Another thing to consider on the core with magnet is, even though the flyback does not have a long high voltage spike after the off time but notice how much wider it is. See the off time waves I posted below. The core with magnet may have just as good of an RMS value on the off time flyback then the narrow high voltage spike. If we compare each wave after the off time by surface area (RMS) I find they look to take about the same surface. So if one is taking 75% less current and the fyback is about the same RMS value then is this not something to consider?

Thanks for sharing

Luc