PIEZO NXT type panel

[jzagaja]

As for the third strategy I'm curious what will happen if you take such a foam:

http://apama.pl/APAMA%20FOTOS/PANEL%...T7.5.200-1.JPG

spray with a stiff resin and excite transversely and longitudinally.

[LineArray]

Hi,

for me the link does not work, could you post a picture please ?

Maybe there was a misunderstanding concerning my former
post:

I do not think that it is very useful to increase modal density
at low frequencies by searching for materials only.

I'd like to do it by design ...

But i am curious concerning that foam ...

[jzagaja]

The idea is as follows - generate longitudinal waves like in yours plus transverse and if they are shorter than panel dimension then radiation like above fc occurs. Below bending waves are excited on pyramids - "wide coincidence region". Radiation will be efficient but the question is if f0 are not higher than in a plain membrane.

[LineArray]

Originally posted by jzagaja
... Below bending waves are excited on pyramids ...

Can you please explain ...

"Below" means below concidence frequency or below the
range where (bending) wavelength is in the range of panel dimension ?

What do you mean with "excited on pyramids" ?

What kind of motion do you intend for a single pyramid e.g. ?

[jzagaja]

I mean the torsion motion on the edges, radiation below coincidence. You can check it with your transducer just compare curved membrane with a flat one. Curved will be more efficient. Bending waves forms around exciter and edges so more "knees" you have more the longitudinal waves are transformed into bending transverse waves.

[LineArray]

Originally posted by jzagaja
... just compare curved membrane with a flat one. Curved will be more efficient ...

Do you mean the large scale curvature, like in my tranducer or
do you mean the "small scale" curvature introduced by
pyramids (crumbles, ripples) on a structured surface ?

My transducer has both, but the "small scale" curvature (ripples)
is small against bending wave length ( at least up to the mid
frequency range).

[LineArray]

What about "fractal folded" strucure ...

Of cause the structure below is not fractal enough

By that you introduce

- larger surface
- length of possible propagation paths increases and is
randomly "smeared" over neighboured paths
- stiffness, mass and speed of bending waves will vary regionally
and is path dependent, thereby varying along each single path.

I think this is what i have already to some extent in my
transducer, although it is not really "fractal".

[LineArray]

more like this ...

http://darwin.bth.rwth-aachen.de/opu...html/img12.gif

[jzagaja]

Any change in curvature, bending stiffness generate bending waves. Please have a look again (page 14+):

http://www.wikipatents.com/gb/2400264.html

[LineArray]

That is understood, but i thought we were discussing

- how to increase modal density by structure (not overall size)
- some idea of "broadband" coincidence

???

Your foam with pyramids was an attempt to go into that
direction. I just tried to understand your idea, while finding it
interesting ...

I am currently thinking about a different approach to
increase modal density in the lower frequency range of a
DML, which does not necessarily imply curvature of any scale.