Ribbon using silver leaf alufoil

[Lgrau]

Hi,

I just started a project using silverleaf alu. The aim is to go fairly deep, around 1KHz, at best with linear frequence responce and low harmonic distortion.

I expect to keep the ribbon in plane with the front and not in the middle of the Neodym magnets. This is in order to keep diffraction low.

The foil I got is around 0.6um think, on a roll 15mm wide and with the lenght of 16cm.

I've made some test with pure alu, but it's really fragile, and the internal damping is most likely not good enough to go < 2KHz.

Measurement of pure alu ribbon.
nearfield.jpg

I've now tried to glue the ribbon to cotton sheet, with quite good results.

Measurements and pictures of the ribbon.
bomuld lim.jpg

bomuld i rockwool full.jpg

bomuld i rockwool full bag.jpg

bomuld i rockwool full closeup.jpg

Any comments to the placement of the ribbon, not in the middle of magnet field is this a hopeless approach?
Since I don't expect the movement to be more than maximum +-0.2mm I assume the magnet-field is not changing that much.

[bear]

Interesting approach.

The cotton adds mass, but structural strength. I think that shows up with ur F3 point dropping?

Have you determined your sensitivity?
Impedance?

It may be that the "silverleaf" is too thin (won't handle enough power) and also is very soft compared with the alloys usually used for ribbons...

As long as the field is essentially constant, the response will be linear with amplitude, where you may get into trouble is with higher levels. A bit depends on where the ribbon really is placed...

_-_-bear

[gootee]

You have magnets with the N and S poles on the largest sides. I assume you have one row with N facing out and one with S facing out. Imagine the magnetic field lines coming out of the N faces, perpendicular to them, then bending to be parallel to the face, going across/through the ribbon and toward the S magnets, and finally bending down into the S faces.

If the field is parallel to the ribbon, everywhere there is ribbon, then current through the ribbon will result only in forces that are perpendicular to the ribbon. But if the field has already started to bend, or the ribbon is too far away and the field always curves and never becomes only parallel to it, then there will be forces that are not perpendicular to the ribbon. And there could be forces that are not uniform, on different parts of the ribbon.

You should be able to apply a DC voltage (probably through an appropriate resistor), so you could directly observe whether or not the ribbon deforms, or has forces other than perpendicular ones on it. If the ribbon is all in the parallel portion of the field, it should simply move forward or backward, and all by the same amount so that it stays flat. If the ribbon is too wide for the field's parallel portion, you might see the edges of the ribbon trying to move differently than the center. Or the whole ribbon might try to move sideways, for example.

[Lgrau]

Originally Posted by bear

Interesting approach.

The cotton adds mass, but structural strength. I think that shows up with ur F3 point dropping?

Forgot to mention my mic is dropping heavily above 14KHz, I'm quite sure -10db at 20K is linear response from measurements on other commercial speakers. (I should get a calibrated mic.)

The mass is still in reasonable range aprox 3.2mg for the ribbon + aprox 60mg for the cotton.

Originally Posted by bear

Have you determined your sensitivity?
Impedance?

Not yet, but I think I'll user more powerfull magnets later. at least if sensibility is an issue.

The resistance for 16cm is aprox 1.2ohm, I plan to use 3 in series to get a reasonable load. (and better lower frequency perfomance)

Originally Posted by bear

It may be that the "silverleaf" is too thin (won't handle enough power) and also is very soft compared with the alloys usually used for ribbons...

As long as the field is essentially constant, the response will be linear with amplitude, where you may get into trouble is with higher levels. A bit depends on where the ribbon really is placed...

_-_-bear

Yes need to perform longer test to check stability.

It's very soft, and doesn't require much damping to remove all ringing, thats the good part, beside low mass and fairly high dc resistance.

[Lgrau]

Regarding applying dc: I'm not sure if it will bring any info, since I'm quite sure it will bend nomatter how the force is since the tension from the mounting will let the side's move less than the middle.

And finally it's also not planned that the movement will be as high as when DC is applied.

Which distortion will it give moving out of the linear magnetic field?

[gootee]

Originally Posted by Lgrau

Regarding applying dc: I'm not sure if it will bring any info, since I'm quite sure it will bend nomatter how the force is since the tension from the mounting will let the side's move less than the middle.

Just trying to provide a method to potentially help you easily experiment with ribbon placement and magnet configuration. You could try the same things with different distances from the magnets' plane, or with more magnets, different magnet layouts or stackings, etc , or whatever, and keep track of and compare the results, and you might eventually find the way to make it the most linear and sensitive.

I would try it and see. The edges might curl up or down, moving more or less than the center, for example, which might mean the field starts to curve down toward the magnets before it gets past the edges of the ribbon. (Or maybe that would tend to crumple or stretch the edges. I can't remember right now.)

And finally it's also not planned that the movement will be as high as when DC is applied.

The DC level could be adjusted to give any displacement, from zero up to the limit of the power supply. So it doesn't have to be a lot. On the other hand, you could also check to see how close to the magnets would be OK, in case it's possible to get the ribbon closer to the magnets, where there should be more parallel field for it.

Which distortion will it give moving out of the linear magnetic field?

Non-linear, I suspect.