Acoustic magnetic generator.

[verpies]

In response to a private message I'm posting the diagram below that illustrates what I mean by a "magnetic circuit".
I'm posting it here because attachments are not allowed in private messages.

This diagram merely illustrates the electric analogy of a magnetic circuit, it is not a suggestion to try this circuit with the Villari effect.  The electric analogy is not complete because it does not show the reluctance of empty spaces between core segments and the PM.
Not that it's bad for it but I have not thought about reluctance modulations of this circuit in the context of magnetoacoustics.  Maybe it would be useful with it and maybe it would not - I'm too tired to think about it now.

In any case, without knowing precise locations of the nodes and antinodes, the chances of doing something useful with it are very small.

[itsu]

After days fiddling around with the setup i had, no noteworthy things happened.
I decided to switch over to another setup like being seen in the video (piezo flat on a glass plate, 1 ferrite rod glued on it).

Also here nothing strange found in the lower frequency regions, but when upping the frequency to around 55KHz
there started to be some sparks and something started to glow at the bottom of the piezo were it touched the glass.
I think it must be some residual flux which i had used to solder the connections to the piezo earlier.

Some movement was seen as the construction was sliding across the plate now and then pointing to some
vibrations causing also some audible tones again around 4-6KHz.

I did had some volts (3/4) on the pickup coil.

Nothing spectacular, but at last something was happening.

Video here:  https://www.youtube.com/watch?v=XpQsvv-8XCM&feature=youtu.be


Regards itsu

[verpies]

That arcing is worrisome.  I never experienced it but then I had never driven my piezos above 300V.   My IMT outputs maximum 210V_RMS but it delivers 5x more current than your IMT at that voltage.  The IMT has back-forth primary winding above the secondary (both spanning the whole circumference of a ferrite toroid) and it attenuates 3dB at 3.8MHz.
I like that you stood the device on a table, because the table constitutes a mechanical counterpoise for the other side of the piezo.

That the whole contraption moves on a glass table must be caused by some standing waves in it at acoustic resonance and its multiples.  Too bad, that we cannot see them. 
When acoustic reflections come back and the standing wave builds up, it squeezes the piezo turning it into a HV generator (that voltage can be high enough to cause arcing).  Note, that some cigarette and gas lighters use hammer-struck-piezos to generate HV sparks for gas ignition (see here).

Your thick rod and two return flux rods are very close to each other now, and your magnetic circuit becomes more parallel than before because the space between these rods is small enough to "allow" significant magnetic flux to take a shortcut outside of the ferrite.
If you read my previous message, you'll note that parallel magnetic circuits stand a better chance to have large flux variations on one of its branches, than series circuits ...especially if magnetic circuit nodes correspond to acoustic nodes and antinodes.

Strangely, an open magnetic circuit is more parallel than tightly looped circuits (this is most likely the reason why this guy got so lucky   ...this is not to be misconstrued that open circuits produce the largest flux variations and EMF ∝ dΦ/dt.

[itsu]

When acoustic reflections come back and the standing wave builds up, it squeezes the piezo turning it into a HV generator (that voltage can be high enough to cause arcing). 

Hmmm,  should i not have seen that HV on my yellow probe signal?

Strangely, an open magnetic circuit is more parallel than tightly looped circuits (this is most likely the reason why this guy got so lucky  

I tried that setup lateron, see picture, and it produces similar results as the earlier setup, but then when i applied some pressure
on the magnets (probably to much on one side) the piezo shattered.

Looking at the damaged piezo, i can see where the glow came from earlier, it seems the edge where the back side "flips over" to the front site
is "eaten away" by this burning.  The back is now isolated from this front patch.

These piezo's are very bad to solder as the conductive material does not want to be soldered very well.

Anyway, still one to go, so i will revert to "clamping" the leads to the piezo as i have done before, only problem is that i have to use
the edge of the glass plate to accommodate the clamps.

Regards Itsu

[verpies]

Hmmm,  should i not have seen that HV on my yellow probe signal?

Yes it should, unless the piezos has generated that HV locally at the edge.

Looking at the damaged piezo, i can see where the glow came from earlier, it seems the edge where the back side "flips over" to the front site is "eaten away" by this burning.

Acoustically heated glass can become conductive.  Are there permanent melt marks on the glass table?

These piezo's are very bad to solder as the conductive material does not want to be soldered very well.

This is a well know problem with piezos.  See this video.
Also, wires that you soldered to the piezo are too thick.  I use sub 0.5mm flexible stranded wire for a 5cm length and join/solder them to thicker wires for longer distances.