Electrostatic AMT?

[Keith Taylor]

Very good thinking Kenpeter. If I understand you correctly you are alluding to something I had thought of in a half baked way. A drawing would be very usefull here. For purposes of explanation the "pleat" could look like a capital H. Maybe the vertical parts are rigid stators with a very floppy membrane running diagonally across the opening. When the horizontal element of the H is in the centre this position is created by the lower part of the membrane "sticking" to one stator and the upper part to the other. Thus we are exploiting something that Electrostatics is good at--generating force with small spacings. We also consign membrane tension to the rubbish bin.

This idea needs a lot of thinking about as the motion is quite unlike that encountered in 99% of transducers. David mentioned adverse effects of zig zag pleats in an EM AMT. I could be wrong, but tend to view the pleats as having a rigid part (The conductive foil) and a floppy part (The folds). Perhaps, under drive, the folds alternately "balloon" out and pull taught to preserve parallel pleat motion.

Keith PS Borat, I have to disagree with your aversion to pleats. Ignoring motor means, pleating confers advantages in getting an improved impedance match to the air and can result in an acoustically smaller transducer with directivity benefits. Your idea would work well in an AMT!

[Keith Taylor]

In the drawing I am attempting to explain kenpeter's idea. All aspects of practical realisation have been neglected. When we consider the crackling noises associated with unwinding sticky tape from a spool we may have a serious problem. I think kenpeter was envisioning a hand over of the sticking area from the bottom of the pleats to the top. This would defeat AMT motion and leave you with something like depicted on the right.

Keith

[DavidJanszen]

Originally Posted by Keith Taylor

In the drawing I am attempting to explain kenpeter's idea.

Wow, Keith, very astute. I did not understand, until I saw your sketch.

What an interesting idea, kenpeter. Surely not one that lends itself to easy realization, but puts forth a quite intriguing motor mechanism. On the plus side, in its ideal form, it would apply the absolute maximum possible electrostatic force. I won't begin to list the challenges.

About the AMT impedance match, I feel I am repeating myself ad nauseam, but the membrane in an AMT experiences a higher impedance air load than in a planar. I believe this is disadvantageous when driving electrostatically, where the extra loading is likely to more than offset the increase in driven area. ESL's are already as well matched as possible to the air load.

[kenpeter]

Don't know that it could actually kink over hard 90 that way?
Was thinkin something halfway between pic on the left and
right. Or perhaps an S shape rolling down the middle of this
crazy audio railgun...

Course, any slack (obviously some is necessary) is gonna
crack like a whip if this rolling wave hits XMAX either end.

Picture crew of the Argo shouting in unison "Prepare to fire
the wave motion tweeter!"

[DavidJanszen]

Originally Posted by kenpeter

Don't know that it could actually kink over hard 90 that way?
Was thinkin something halfway between pic on the left and
right. Or perhaps an S shape rolling down the middle of this
crazy audio railgun...

Course, any slack (obviously some is necessary) is gonna
crack like a whip if this rolling wave hits XMAX either end.

Picture crew of the Argo shouting in unison "Prepare to fire
the wave motion tweeter!"

Okay, what if the goal were to normally drive cells to their limits, but use an array of cells, all of different depths? Each would take a different amount of time to reach its limit. When there is a very high frequency input, for instance, a majority of the shallow cells would bottom out, but not the deeper ones. At lower frequencies, the deeper cells would "take over" when the shallower ones have bottomed out. The aggregate mechanical response might produce sound.

Further, if the shallow cells were arranged at the center of the array, and cells became progressively deeper with increasing radius from the center (circular transducer) or distance from one edge or the center-line (line transducer) , this might create a holy grail variable aperture that would provide uniform dispersion at all frequencies.

Of course, I think the physics of the basic motor would preclude the development of any real force, because even if unwanted membrane-to-membrane charge transfer and induction problems could be solved, and materiel fatigue were not a problem, and the thing could be fabricated, and the air load were not an issue, still the area where the force would be applied is only in the near vicinity of the line where the membranes meet, but hey, I still love the idea.

A first pass apparatus might be variant on the Leyden jar, with a strip of polarized membrane suspended between two stators, attached at the bottom of one stator and the top of the other. If the part that crosses between the two strips can be made to climb up and down the stators, that would at least be a cool science demo.

[mige0]

Originally Posted by DavidJanszen

ESL's are already as well matched as possible to the air load.

Hardly ! And to repeat myself - learn the basics !

- are you sure you are *the* Janszen ?

Michael

[Few]

Posted by DavidJanszen: "ESL's are already as well matched as possible to the air load."
Hardly ! - are you sure you are *the* Janszen ?
Michael

OK, so please illuminate us concerning the nature of the impedance mismatch between the ESL diaphragm and its air load.

[kenpeter]

You makin it too complicated. But if you are determined to go there...
I don't see how havin middle bottom out early prevents the deeper
outer portions from transducing treble. Dispersion worse, not better?

Now, you could run an array of cells with a hemispherical DC offset.
that would make the wavefront appear to come from a dome... Or
have cells that simply stick out further in the middle? But that might
be tougher to manufacture than fakin a dome shape with DC...

I'm really more concerned how to make XMAX a less traumatic event?
How does one absorb the energy of the whiplash without shredding
the ends of the ribbon in the process? Or is transfer of energy to air
good enough all by itself to provide that damping?

Might be a moot point? I think XMAX here is arbitrary. We can make
stators and ribbon as lengthy as air friction losses will allow. Its not
frequency limited by depth in the manner of an AMT, where you are
fighting to sum phases from distributed depths within the same pleat.

[DavidJanszen]

Originally Posted by mige0

Hardly ! And to repeat myself - learn the basics !

- are you sure you are *the* Janszen ?

Michael

Son of.

Of course, I would suggest a similar directive regarding the basics to you.

I will review the validity of your contention that the air impedance is inconsequential to an ESL, and report back. For now, as I have already said, I believe that compounding the air load, such as with pleats, will decrease the amplitude of membrane motion for a given amount of force, not that this is the only attenuation mechanism at work in an AMT.

In the meantime, please don't let this stop you from backing up your assersion with an expplanation of some sort.

[DavidJanszen]

Originally Posted by kenpeter

You makin it too complicated.

You think so?

Anyway, I think it's fun to think about damage from an ES rail gun whipping to its limits, but as a real concern, now might be too soon. The initial challenge with any new transduction or motor scheme is getting it to move at all. I think if you can get the Leyden jar thing or something like it to work, though, you might be on to something.