Driving ESL with valves/tubes

[jfitz57]

I have seen it and read enough of it to know I won't be infringing on it.
When I said "clone" I did not mean it litterally. Much of the circuitry I'm considering hase been around since the 1940's. As a DIY'er I won't discuss legal issues.

Jim

[See Hong]

Originally Posted by Calvin

Hi,

Other very interesting devices could be the upcoming SIC power JFETs.
For example the 1200V enhancement mode or ´normally-off JFET (enhancement or normally-off sounds weird with JFETs but they call it so) of Semisouth (SJEP120R063).

jauu
Calvin

I will like to know if any one could find out how to get the parameters for this SiC JFET as I am currently undergoing a project on this device.

[See Hong]

I will like to know how and where can I download the spice model from? I have tried searching but couldnt find it online. I need the parameters for simulation which will be done using the software, Simplorer.

[markusA]

Interesting thread and DD sounds exotic, I would however like to reconnect to the original question for a moment.
Driving a hybrid ESL with a tube amp and a small ratio step-up trafo.

I've just started my own ESL project and I don't have any irl data so for now it'll have to be an academic exercise on my part.
Circlotron seems like an interesting OTL solution.
Would it be a good idea to go for a circlotron with medium-high voltage output tubes and a low turn-ratio step-up transformer?
Any candidates that spring to mind?
I guess they should at least manage 250V output for it to be interesting?

I'm just being curious...
My own circlotron that I'm currently building is much less spectacular with 6CG7 tubes.

[Calvin]

Hi,

as any amplifier circuitry the circlotron has advantages as well as disadvantages.
A great Pro is that it works fairly stable into complex loads.
Another Pro is that since its output devices are connected as followers, its output impedance is low, which is good for driving transformers and the trannies behaviour.
As a Con the burden of large signal amplitude lies not only on the output stage but also on the driver stage.
A further Con is the need for two power supplies, which needs to be floating.
Creating the symmetrical signals for the driver stages might not be trivial.

So it might be an easier alternative to use bridge-connected SE-amps with added output buffers? If You use two power supplies here it would even be possible to have a ground referenced DC-free output.

jauu
Calvin

[markusA]

I'm hoping someone will pick up the ball and run with it.
Since my panels are in the initial stages (still gathering building materials) I can't do anyting worthwile myself.
At best I can do some theoretical exercises but beeing a tube n0ob the outcome would probably be less than optimal. Trying to stay possitive...

You are spot-on conerning the cons, maybe I'll get adventurus when the panels are done? Hopefully someone else will have come up with something long before that.

[SY]

Originally Posted by jfitz57

I have seen it and read enough of it to know I won't be infringing on it.
When I said "clone" I did not mean it litterally. Much of the circuitry I'm considering hase been around since the 1940's. As a DIY'er I won't discuss legal issues.

Jim

I will. The patent is expired and you (or anyone else) is free to use it.

[jfitz57]

Originally Posted by SY

I will. The patent is expired and you (or anyone else) is free to use it.

No posts here for many moons! Have got more tools and stuff since then!
Kilovolts and Kilohertz is good!

Jim

[janneman]

Originally Posted by Calvin

[snip]So it might be an easier alternative to use bridge-connected SE-amps with added output buffers? If You use two power supplies here it would even be possible to have a ground referenced DC-free output.

jauu
Calvin

Calvin, how important is this DC-free output? If I DC couple an SRPP-type amp, with say 2kV no-signal output to the stators, is that a problem? Aren't the stators insulated anyway for the high signal levels?

jan didden

[Calvin]

Hi,

as long as both parts of the bridge remain balanced the DC-offset is of no interest signal-wise. Both stators would see equal potential, hence no potential difference. It must be ensured though that the potential difference remains zero under all circumstances, such as ageing and temperature.
Looking at the ESL-behaviour alone, a DC-Offset on both stators wouldn´t make a difference in the working of it (Diaphragm Bias Voltage should be re-thought). Looking at the ESL in a practical surrounding, the stators would exhibit a much higher voltage swing (it would aquire higher voltage potentials) with regard to earth potential. Since the stators don´t float with regard to earth potential, but are referenced via the amplifier towards earth potential. This puts much higher stress on the insulation. Besides the safety aspect, increased voltage stress leads to higher risk of flashover, speeded ageing of insulator materials and earlier breakdown.
I´m not completely sure about the effect of a constantly charged stator/insulator system, but on the surface of the insulation layer of the stator charge will accumulate, waiting on a leakage path to bleed away. This could be a any approaching object like a probing finger or the membrane. In case of the membrane touching the stator, it´ll most probabely keep attached till the charge has bleeded away. The recovery time will be high because the DC-offset of the amp constantely recharges the stator and its surface. A ground referenced stator´s recovery time may be just a fraction of a second long, depending on the insulation layers resisitivity and as such it´d be barely listenable.
But the most important deficiency remains the safety issue.
One might think about symmetrical power supply lines so that the DC-offset is at, or at least close to 0V.

jauu
Calvin