Proximity of permanent magnet increases toroids inductance?

[void109]

While working on a project I noticed something wonky and I don't understand how or why it is happening.

Typically if I have an inductor, the presence of a permanent magnet will reduce the inductance according to its distance - however I have started using a type of nanocrystalline core that is showing a marked increase in inductance when at a certain position from a permanent magnets pole.

I can't rationalize this - could someone help explain this? I'd be most appreciative, it's been bothering me for some time!

I shot a quick video to demonstrate what I'm talking about:



Thanks!

 

[harveyab]

Here is a patent that may give a small insight for you:
http://www.faqs.org/patents/app/20090146772
Hope it helps a little.

 

[harveyab]

My last post, I think was quite off the mark. Sorry.

After watching the video, another thought may be resonance. You mentioned that the inductance meter was set to 1kHz. And you were careful to set the height with the guitar picks. You also pointed out how very sensitive it is to the exact position in the center of the magnet. Inductance meters are definitely not "fool proof". I suspect you have set up some kind of resonance that is tricking your meter. Does your meter have another frequency setting? Does it show the same rise with a peak at the same distance and position? You might try building up a simple oscillator with the toroid and various capacitors to see what the resonance frequencies are. This may be a better way to determine the inductance also, as a double check. I think that the cores themselves have a resonance frequency (without regard to the capacitor). You could also hang a scope on with your inductance meter to see if anything "interesting" is happening while you are measuring. You might also try sweeping the frequency with a function generator and a scope to see if there are any resonance peaks.

Hope this helps, if I am not too late with this response.
Harvey

 

[void109]

Thanks for the response - its been awhile between posts, however if you see this here's some additional info:

Since seeing that behavior I've been able to reproduce the effect with various toroids (always of the tape-wound variety), also when checking the inductance I've used both 1Khz and 120Hz. I have also used long test leads to keep the meter away from the field.

Also noted that after hitting an increased inductance point, I can move the toroid away from the magnet in the direction of the magnetic field and the increase tends to hold, until a certain point and it drops away rapidly. Upon approach, you have to get close enough past a certain critical threshold (closer than the drop off point) to see the increase again. Its like a rubber band that attaches at the right spot, and as you move it away it stretches until it snaps away.

I haven't done any further testing because I don't even know if this is anomalous and worth looking in to, since I'm still in the early stages of my electrodynamics self education. :)

If this does sound impossible or wrong, I'll do further tests.

 

[AJ Bentley]

the increase tends to hold, until a certain point and it drops away rapidly. Upon approach, you have to get close enough past a certain critical threshold (closer than the drop off point) to see the increase again.

That sounds suspiciously like hysteresis.

I haven't checked on exactly how commercial inductance meters work but I would guess they are pretty unsophisticated and simply look at the impedance for a fixed frequency.

If you are disturbing the magnetisation of the toroid core, you may case it to partially saturate, which could distort the waveform sufficiently to give anomalous readings. Could be up or down.
I would expect to see hysteresis effects.

You would have to make careful measurements of resonant frequency using fixed capacitors with a sig. gen. and a scope to get accurate inductance measurements at each distance.

IMHO It's worth looking into for the insight you'll gain into the behaviour of inductance generally.

 

[harveyab]

Your description,

the increase tends to hold, until a certain point and it drops away rapidly. Upon approach, you have to get close enough past a certain critical threshold (closer than the drop off point) to see the increase again.

This definitely sounds like resonance to me. That is the overall circuit is going into oscillation, and as you move away, the loop gain drops below one (or the phase shifts too far) and the oscillation dies out. You have to get closer than that point before the oscillation will "self start" again.
This is not core hysteresis, as that is an effect that is present on each cycle of the excitation frequency. What you are seeing is related to oscillation around resonance. I have seen very similar effects in my own experimental setups.
Harvey

 

[AJ Bentley]

This definitely sounds like resonance to me.

You could well be right about that - I guess the self-capacitance of a toroid might be enough to get into oscillation at quite a low frequency (120Hz !)

In which case it's a p. poor inductance meter that doesn't account for the possibility.

Just goes to show that you shouldn't believe everything you read (on a meter )