How to reliably reset a square loop ferrite ring?

Theory says that B will be 0 when (+/-) Hc is applied, but for some reason it just doesn't work. When I turn off the current, B jumps back to Br.

What I want to do: take a toroidal ring from an old computer (Soviet era) and demagnetize it, which should make the winding exhibit a relatively high inductance. Now I want to detect if a current high enough to permanently magnetize it had flown through another winding, which will manifest itself as low inductance of the measurement coil. Kind of a latch to be used in a resettable fuse.

The cores work well in pulse experiments, but it is notoriously hard to bring them to the linear region.

Best regards, Piotr

Reply to
Piotr Wyderski
Loading thread data ...

Sounds like bringing a schmitt trigger to a linear region :>

piglet

Reply to
piglet

Sort of, but being done on a daily basis in every PC power supply. :-)

But these cores are not perfect, the slope has about 3..5% of linearity region. The problem is it doesn't want to stay there.

Best regards, Piotr

Reply to
Piotr Wyderski

I seem to remember a "magnetic washing machine" which started off with an alternating current higher enough to saturate the material, and gradually reduced the amplitude to get the residual flux down to a relatively low level.

I can't remember the context. Google finds a reference in paleomagnetism research on old rocks, but I've never been anywhere near anything like that.

formatting link

--
Bill Sloman, Sydney
Reply to
bill.sloman

that's a classic demag strategy.

NT

Reply to
tabbypurr

And with wide applications, every colour CRT TV set used it to clean the mask at startup. But is was made of soft iron. Is this technique appropriate for square loopers?

Best regards, Piotr

Reply to
Piotr Wyderski

Whose theory? ;-)

I'm not aware that there's any comprehensive theory of magnetism. It's a horrendously complicated interaction, and literally anything is possible within a material. The stuff could be computronium for all we know. (Computers have been built with the stuff! Though not purely magnetic stuff, but wires and such too.)

I once played with some square permalloy cores. I had a square wave generator, the core in series (just a single winding, 20 turns or something), in series with an electrolytic capacitor (10uF), and then a load resistor (100 ohms or so).

Normally, a saturable inductor should be demagnetized by this, but it always walked off to one side. A small amount of bias could push it back (say, 10k from +V), but it would always drift back to one side.

By 'side', I mean the output is normally ~0V across the resistor, but as it shifts towards saturation, some fraction of the input is let through. For a square wave input, the output is a lower duty cycle square wave, the duty cycle being determined by the flux remaining until saturation. So for positive bias, wider positive pulses are transmitted, or for negative bias, wider negative pulses are transmitted.

The amount of flux walking was much stronger than any leakage through the capacitor, so it wasn't that.

It's probably not unreasonable to have a material property like spontaneous magnetization, but there's no obvious reason for that to occur in this material.

Also, a lot of ferromagnetic materials have a butterfly curve, rather than a simple hysteresis loop (flat in the middle -- low initial permeability). Maybe it's different from one side or the other so it has a magnetic rectifying effect...

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

The initial magnetization curve is called the virgin curve, and not in vain. Square loop material is developed to resist the demagnetization.

You might get some results heating the ring to above the Curie temperature.

--

-TV
Reply to
Tauno Voipio

I was stunned to see how sharp the effect is in ferrites. This particular material has its Curie region between about 105 and 110 degrees centigrade. Would make a good thermal protection device.

But heating is the last resort, I'd like to do it all-electronic, so asked whether there are some ancient tricks, as today nobody seems to care about weird magnetics.

Best regards, Piotr

Reply to
Piotr Wyderski

Good point. The more I experiment with magnetics the lesser the results correspond to the bookish knowledge. It seems that most stuff are rough approximations applicable only to the linear region.

[snip] So it means more experiments. :-)

Speaking of which, I've just managed to drill a hole through a ferrite toroidal core (6x3x3mm, my hole is below 1mm) and I am ready to experiment with transfluxors. :-)

Not bad for the very first attempt to drill in ceramic:

formatting link

Best regards, Piotr

Reply to
Piotr Wyderski

Depends on your choice of books. ;)

Ferromagnetic materials are very complicated, even crystalline ones, mostly due to the quantum interactions. Classically, ferromagnetism doesn't exist

--it's the exchange energy that makes the spin-aligned state energetically favourable on short length scales.

There are special cases that are easier to solve, e.g. the Ising model in 2

-D and spin glasses in dilute ferromagnets, but the general polycrystalline case doesn't lend itself to analytic solutions (or even practical numerica l ones). Theory can provide guidance, of course.

Cheers

Phil "one of those pesky physicists" Hobbs

Reply to
pcdhobbs

Did you use the copper rod awash in fine carborundum approach?

The copper eventually wears away, but the grit sticks into the soft copper and grinds away the hard ceramic. I've used it to get through glass, but it was a very long time ago.

--
Bill Sloman, Sydney
Reply to
bill.sloman

OK totally off thread... except that it's a "magnetic butterfly"

formatting link

Optical interferometry measuring the length change in a nickle rod. My daughter (then age 3-4??) colored the data, it's been taped to my door for years. You can "unfold" the butterfly and see it's a hysteresis curve.

George H.

Reply to
George Herold

Oh I should have added that the current is the current through a solenoid that surrounds the rod.

GH.

Reply to
George Herold

As inductance has no measurable effect in a DC cct, except for an internal flux density, you have to use AC or pulsed DC to get any measurable information from it.

A flux sensor will tell when Ni from two windings are cancelling, but there is no static information that tells whether the core is demagnetized in a powered circuit. The square loop material's remanence will also be unaffected. The simplest static indicator is the flux sensor.

If the control winding is used to intentionally saturate (reset)a core, the time required for the circuit to saturate in the opposite direction through a second winding (typically one turn carrying the current being measured) can be calibrated - a measurement technique used in some metering circuits.

During the flux swing, a voltage will be measurable across the control winding - the control/clamping of this voltage becomes one of the calibrating mechanisms. Current will be propotional to the Vt product. This is a relatively slow measurement method, similar to capacitive integration in voltage measurement - unsuitable for fault detection.

RL

Reply to
legg

The current measuring system that was - and may still be - used on the Lond on Underground was a pair of C-cores that were clamped around the busbars.

Once the two C-cores were clamped together hard enough to look like a torio d, the whole was driven by a multi-turn winding on one of the halves as sim ple Royer oscillator, which flipped whenever the core saturated.

The difference between the current needed to saturate the core when it was aiding the current in the bus-bar, and when it was opposing it, was a very accurate measure of the current in the busbar (and lot lot smaller, in prop ortion to the number of turns in the multi-turn winding).

My boss a George Kent claimed to have invented it a few years earlier - thi s was around 1974 - and was rather pleased with it. This was the guy who di d his apprenticeship at the Royal Radar establishment, working for Peter Ba xandall (amongst others). No slouch.

--
Bill Sloman, Sydney
Reply to
bill.sloman

I demagnetize current transformers by connecting them to a Variac. Ramp up the voltage into the saturation region, then slowly back down to zero.

A ringing oscillation should accomplish the same thing.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Now that you have that, can you use a local oscillator to saturate that area around the hole and build a mixer? Or some combination of holes in a 90* orientation to a normal flux path. I worked with a physicist years ago that was sure we could build a strong mixer by saturating part of a ferrite toroid. It has been a long time, but it seems we got a little mixing, but a lot of heat. Even had it in an oil bath. One of our efforts-- cut a slot in a toroid the same width as the height, then slide a second toroid in the slot to look like chain lengths except the toroids overlap. Might need to abrade the edges to get flat toroid sides and gap for a good fit. Then drive one toroid to saturate the gap at a 90* angle to the normal flux of the other. Mikek

PS. He always said, The Russians did it!

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
amdx

Cool, thanks!

Now, is that the same butterfly I've read about, or is that just the normal B-H curve in a different variable? (Namely, with respect to B^2, because magnetostriction is square-law.)

What's the count? Interference fringes?

Oh heck, if we're going to round out the units, then I might as well ask, what wavelength, and approximate length of wire?

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

Right It depends only on the magnitude of the B-field. (I don't know if it goes as B^2 or something else?) So it's not any "real" butterfly.

Right again. (I think each count is 1/4 wavelength of a HeNe.)

Nickel 200 alloy, it was a ~12" rod in a ~8" solenoid... so there are all sorts of end effects on the B-field. Heck I'm not even sure of the sign.. Does the nickel get longer or shorter in the B-field? It was mostly a demo to show off the interferometer.

George H.

Reply to
George Herold

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.