Sketch For Bending Transducer (Prototype already built)

[LineArray]

Hi there,

this bending transducer works much like the well known
devices from MBL. But instead of having the shape of a melon,
enclosing a volume totally, this device is open and can be
coupled into an enclosure.

There are 2 curved planar panels radiating.
With proper dimensions and wise choice of material,
fullrange operation and applicable dynamics should be
possible ...

What do You think guys ?


Cheers

[LineArray]

Front View with magnets removed ...

[Ed Holland]

Nice idea LineArray

One other idea that you might consider: rather than the diaphragms being rectangular (as viewed from the front) - they could be trapezoidal i.e. different width from top to bottom. In this way it might benefit behaviour as a full range driver.

Keep thinking - you are definitely on an interesting track here.

Cheers,

Ed

[LineArray]

Thank you Ed,

i thought of trapazoidal shape, but i think the mode
of movement of the "voice coils" will suffer from that ...

Do You think of that because you want to avoid resonances ?


Regards Oliver

[Ed Holland]

Actually I was considering frequency response, plus the size & mass of the source, thinking that you might achieve a graded transition from high frequency to low frequency as the panel dimension increases.

However, I note your concern as to the behaviour of the voice coil if the two edges are not parallel - the whole assembly would experience additional forces in the vertical direction as well as the intended squeezing.

Ed

[LineArray]

Hi Ed,

i built a prototype 3 Years ago - veeery sloppy - just to
test the idea. Diaphragm and VC were very heavy.

Resonant frequency was subsonic, you could watch the decay
with your eyes when you tipped the "voice coil" open
circuited.

I was astonished that highs produced were "OK". I think
upper frequency limit is limited mainly by the width of the
diaphragm and the sonic speed of the material.

Distance from VC to VC limits frequency range.
My Proto had only one VC, the other end was fixed.

From my gut feeling i would say, that due to the sonic speed
of diaphragm material chosen, a quarter wavelength
(with longitudinal dispersion along the diaphragm) is
the maximum width (path from VC to VC) of the diaphragm.


With AL (c around 6000 m/s) lambda is 6000[m/s]/20000[1/s]
= 0,3 m. So max width would be 7,5 cm.

(...The 'Planot' device e.g. MUST have torsional vibration ...)

With two VC ( excitation from both sides) maybe width can be
doubled. If the upper frequency limit additionally is compromised
i consider a width around 20cm as practical limit for fullrange
operation, when Aluminium is used.

However beryllium is faster around 12000 m/s ...

Total mass seems not to limit upper frequency limit, but
overall efficiency, like things are with conventional dynamic
speakers.

Although trapezoidal shape may be interesting for other
puposes, it seems not necessary for reaching higher
frequencies IMO ...

What i intend is forced oscillation of the 1. bending mode
over the e n t i r e audio range. I hope, that with proper
dimensioning and damping, this is achievable ...

Cheers,

[ABJensen]

It is the principle in MBL isnt it.

[APi]

I have been thinking this a lot lately:

1) Radiation pattern is symmetric -> Good power response.
2) Magnets are not blocking sound propagation in the air.
3) Magnetic flux can be high.
4) If the tube is very long you can actually use it as a combined "speaker box and reflex pipe" which can be tuned to maximize bass untill the cut-off frequency. But I would use a bit stuffings inside?

One cone material that could be used is thin FR4 printed wired board which has relatively high sound velocity and easier to use than exotic metals. You can get FR4 from 100um thick and can etch voice coil on it.

[jzagaja]

Does it have roots in US patent 3686446, explored by Paddock later on?

[jzagaja]

So basically best diaphragm material is hardened aluminium (>8000 m/s) - 2 layers and between viscoelastic material - Latex? EE-Power Technology CO. LTD offers nice 1,8-3um aluminium foils.