NXT Exciter On Mylar Diaphragm?

[borispm]

Hi everyone,

I am playing with the idea of trying to fit the largest NXT exciter that Parts-Express carries to a mylar sheet instead of foam board or so. Any idea on that? Will it work or not? I am thinking of getting something with wave propagation like Quad ESL's main panel does. I know it's a bit different from NXT's recommendation on stiff panels, but I'm hoping for accidental discoveries.

Regards,
Boris

[LineArray]

Hello Boris,

a stretched thin foil will reproduce sound if excited.

Problem with the "Dayton" style exciter sold at PE is
they are designed to operate as "seismic mass" without
being attached to a frame themselves.

They use their "sticky spider legs" to attach to the
membrane itself to hold their own mass.

If you do that to a thin foil, you add very much mass
in relation to the membrane, which lowers the resonance
of the stretched diaphragm.

The "legs" will also influence wave propagation
- i expect kind of 'hampering' - and its weight will
pull on the upper area of the foil.

Attaching the exciter with its magnet to a frame
to hold it's own weight and attaching only the
moving part of the exciter to the membrane
is the better way when using foil.

Kind Regards

[borispm]

I am actually inclined to using a "Podium Sound" design, the only difference is a mylar sheet is used instead of the honeycomb, and only one exciter will be used.

[mondogenerator]

ive often though about the validity of doing this myself, and since i havent i couldnt criticise. i would really love to try this with a large piece of PVDF(?) but the obvious expense stopped me. I feel a little wary of the exciters, since many require a 'solid' diaphragm, but if you fixed the rear of the exciter, alllowing only the motor and mylar to move, then im sure youd be getting somewhere. mylar tension, and the concentricity of the exciter, would then be about the only 'tweaks', or problems. i.e. film resonance dependant on the tension, and carefull tensioning of the film to retain VC concentricity. im assuming the exciters yours talking about are 'dynamic' and not 'piezo'.

i might still have a go myself, 2x1 foot ellipse, with the boggo standard brass piezo mono morph fixed to rear battening, and bonded to a tensioned mylar film radiator. with enough thought i think a decent baffle mid-HF driver could be made. maybe no ESL but possibly better or at least cheaper alternative, minus the 'doubling' of a more basic amateur ESL attempt.

[borispm]

In fact I'm not so convinced by NXT's total chaos theory. I'm more inclined to the concentric propagation in QUAD's ESL. Maybe the edges of the mylar membrane can be damped with open cell foam "sponges". Maybe normal food wrap may work?

[borispm]

My goal is to get fairly even response from 150Hz-15kHz with an A4 sized diaphragm. Is that too high an expectation?

[sreten]

Hi,

It simply won't work, you'd be better off stripping down a normal driver.
they work best, unsurprisingly in the way intended. I had a pair of
decent Missions + sub, Wharfedale still do a possible package with
far larger satellites, called loudpanel, two drivers per panel, the
panels do 160Hz upwards.

WHARFEDALE LoudPanel Speakers BNIB on eBay (end time 21-Dec-10 21:24:21 GMT)

Best you can hope for , with my Missions, is fairly neutral speech quality.
Surprisingly good though when used for AV rather than hifi.

rgds, sreten.

[borispm]

Hi Sreten,

Is it the the Mission M-Cube that you have?
My idea is a bit different from the NXT panel, instead of allowing resonance, I am actually trying to damp the edges of the panel so that it behaves like a whirlpool in an infinitely large body. I hope by reducing the resonance rather than allowing it to occur, I may have a bit more low end but with even lower efficiency. Efficiency will be even lower than the NXT panel, but with my current idea I may end up with a virtual point source that is quite a distant away from the actual speaker.

Right, tearing a motor from a normal driver should work even better!

[theAnonymous1]

I think I might give this a try. I don't have high hopes for it, but I will be on vacation for 2 weeks and it will give me something to do. It will be the final exciter experiment I do before tossing them in the trash.

I remember coming across a diagram on this forum that showed how to stretch mylar for an ESL using only tape and a flat surface. It basically just showed the sequence that the different edges and corners should be stretched and taped to get even tension. Does anyone know where I can find this?

[borispm]

Hi the anonymous one,

Good to know that you're going to give it a try. I do not have high hopes for that too, I guess the frequency response will be similar to those 1" to 2" drivers, maybe slightly better. But I'm actually expecting far better stereo imaging than the NXT design as the foam at the edge can absorb and reduce resonance, and hence stopping the "total chaos" of the panel. Instead, it will be a well controlled vibration like a whirlpool in calm water. I am looking forward to trying mylar sheets with different thickness and amount of damping. As far as my memory can recall, there is a psycho-acoustic theory suggesting the amount of delay and decay of the delayed signal can affect the listener's perceived source of sound, and this may results in different amount of "ambience" or "soundstage" or whatever you call it.

Regards,
Boris