raymund hofmann wrote:

"Jerry Avins" schrieb:

...

Have you seen such an exciter? What holds it relatively stationary as
the class moves under its influence? Inertia? What would you expect
itsmass to be? Do you expect the frequency response depend on the
size, shape, and thickness of the glass? This is intriguing.

I have seen such a thing mentioned / described in:

http://www.tannoyna.com/commercial/pdf/wp_nxt.pdf

But I wonder if i may be able to buy some ready made moving coil exiters
somewhere.

The mass should be high enough to result in the needed forces to the
window glass without driving the voice coil too hard and thus getting
too non-linear.
Concerning the Frequency response I would have to do compensation anyway
to achieve my goal.
So the thing is mainly that it stays linear enough with the applied
power for the application.

Raymund Hofmann

I read the NXT paper on Tannoy's web site. It is clear from that paper
that achieving excellent sound quality depends greatly on the acoustic
properties of the radiating panel, and no little bit on the exciter
itself. The paper indicated that the bottom few octaves are best covered
by a more conventional woofer, crossing over at (IIRC) 3-400 Hz. You
want to go a decade below that, and I imagine they would like to know
how to do that too.

aside
I have high regard for Tannoy. I own a pair of their 15" dual-concentric
speakers in 7 ft^3 cabinets, and they are fine speakers indeed. (One of
my life regrets is that, in the 1960s, I had to move them out of the
living room to make room for the children.) On axis, their response is
flat to 30 KHz (tested in RCA Lab's anechoic chamber), but the beam
emerging from the mouth of the 12" diameter horn is so narrow that it's
easy to miss the microphone 15 feet away.

My favorite Tannoy product (as an intellectual thing) is their 48"
public-address speaker, intend to be mounted under a helicopter, and
loud enough to transmit instructions to people on the ground, overriding
the helicopter noise.
/aside

If the Soundbug produces suitable sound, by far the easiest way to make
what you want is to buy one (£20; 2 for £37) and replace the battery
with wires to a supply. To make an active one, A would buy a four- to
six-inch speaker with as heavy a magnet as feasible, and carefully
remove the cone, leaving the voice coil, dust shield (dome), and spider
intact. (I assume that the dome is not larger than the voice coil. If it
is, it will have to be carefully removed, being sure that no thing falls
or is drawn into the gap.) Make a cone of heavy aluminum foil with a
base the diameter of the voice coil and a length to project a bit beyond
the frame (or as much of the frame as you decide to leave) and glue the
cone to the voice coil. Attach 3 or four threads from the cone to the
frame to stabilize the voice coil against twisting. (The cone and its
surround provided that stability before they were removed.

Devise your own way to hold the tip of the cone (or an extension of it)
against the glass. The best way depends on too many unknowns for me to
attempt a design.

Jerry
--
Engineering is the art of making what you want from things you can get.
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯ ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯