Passing both DC and pulses through a toroid?

"This equation shows that the amount of energy lost in the material in one cycle of the applied field is proportional to the area inside the hysteresis loop. Since the energy lost in each cycle is constant, hysteresis power losses increase proportionally with frequency.[11] The final equation for the hysteresis power loss is[12]"

https://en.wikipedia.org/wiki/Magnetic_core#Hysteresis_losses

That agrees with what I said. Yes of course in a typical scenario you'd evaluate frequency in the context of a chosen inductor/transformer in which case frequency also impacts magnetizing current. But if that's held constant hyteretic losses are linear with respect to frequency.



Yes speed of light is a time constant...not one that matters for our discussion.

the wiki author overlooks the empirical & scientific fact that the shape and size of a BH loop varies with freq for all core materials - except air ...

the speed of light is the upper bound for EM wave behaviour - so entirely relevant ... else some of your postulates would be dangerously close to being correct ...

If that's what you're sticking with time to cite a source.

I assume you're mixing hysteresis losses with other AC losses...which is what I wanted to move on and discuss.

empirical & scientific fact that the shape and size of a BH loop varies with freq for all core materials

Click to expand...

I do not know about the empirical facts, but this is governed by dielectric (and magnetic) relaxation time constants. For conventional magnetic materials, that will depend on the domain size and hence on the processing steps.

If you want a high mu, you shall require large domains and that will decrease the frequency limit. Smaller domains will give lower mu and higher frequency limit. For all practical purposes, these frequency limits are duly described in the data sheets.

But within that limit, the shape and size of the B-H loop will stay practically constant (independent of frequency). Else they will be useless for any practical purpose.

Only magnetic component in air is oxygen (nitrogen is diamagnetic) but the concentration is so low that it can considered same as vacuum for all practical purposes.

Empirical evidences are rarely useful unless we consider a pure crystalline material.

I have difficulties to relate the recent discussion to the original question. It's about behavior of a soft magnetic core with DC and AC magnetization. Although we know that a magnetic core has losses and hysteretic H to B relation, you abstract from these parameters for a first order solution.

The standard assumption is that a BH loop is nearly constant over frequency - but there are very many materials data sheets - especially of square loop materials - that show shape and size changes with frequency - these plots are hard to generate accurately without lab grade measuring equip and good knowledge of what you are doing.

For example std formulae for losses for Supermendur include freq term at 50Hz of F^1.050 rising to F^1.270 at 400Hz in the hysteresis part which points straight away to increasing hys losses with freq.

Yes there are "other losses" if you want to discuss these please open another thread.

Also look up domain resonance and skin depth of mag materials at frequency - where just like copper ( or any elec conductor ) where there is an electrical skin depth for current ( dependent on conductivity and mu of the material ) there is a flux density skin depth for magnetic flux in a core related to mu and conductivity too - which can severely alter ( i.e. raise ) losses at higher frequencies.

The width wise expansion of the BH loop is another issue - I will see if I can find an easy to understand reference.

- - - Updated - - -

Here is a reference paper that offers some insight for standard cores at DC to 5kHz - change of BH loop with freq;

www.google.com/url?sa=t&rct=j&q=&es...92a56dc8.pdf&usg=AOvVaw2d83xDLVU5URRIy-NmTrIa

Here is a reference paper that offers some insight for standard cores at DC to 5kHz - change of BH loop with freq;

https://www.google.com/url?sa=t&rct=j...U5URRIy-NmTrIa

Click to expand...

The paper you quoted says exactly the same thing I was trying to convey.

Permalloy is one of the high mu materials known: the magnetic domains are very large in size. I do not know whether it is common to use it as a core material in a practical transformer.

You can damage it by hammering it- just bending ti randomly will reduce the mu and break the domains.

The relaxation effect was first described by peter Debye- the Langevin equation was known before: see https://en.wikipedia.org/wiki/Langevin_equation, particularly the section on Generic Langevin equation.

Supermendur is another similar example.

In brief: large domains take longer time to move and align; they simply cannot respond well at high frequency and that is the main cause for losses.

In brief: these ideas are not applicable for finely powdered and sintered ferrites widely used in transformer cores today.

A close reading of that paper shows:



i.e. ferrites show this effect, tape wound amourphous cores also show this effect - in actuality all mag materials show this effect ...

i.e. ferrites show this effect, tape wound amourphous cores also show this effect - in actuality all mag materials show this effect ...

Click to expand...

The author mention the word ferrite in the paper only two times.

He gives his own paper as reference. So I had to dig more.

This is a very old paper and I have to put a screenshot:

More importantly, the results he presents are simulations and none are actual experimental measurements.

It is the same old relaxation dynamics. That is what I understand.

- - - Updated - - -

I have difficulties to relate the recent discussion to the original question. It's about behavior of a soft magnetic core with DC and AC magnetization.

Click to expand...

The question was badly asked: if I understand correctly, it can be recast as follows:

What will be the output of a transformer, assume no core saturation, if a 5V rectangular single pulse, riding on a DC offset of 5V, is applied to the primary? (1:1 primary:secondary)

Please excuse me if I have not understood the question.

My understanding is that the DC offset shifts the operating point on the B-H curve; if you assume no saturation, you will get a simple 5V pulse with the DC offset removed.