@PW:
Six volts input to the ClassE driver board, about 1.3 amps, system resonating at around 330 kHz or so (varies according to environment and load, but the feedback loop keeps the thing in resonance). My primary is too closely coupled; I can get the power draw down quite a bit I think by looser coupling. Once it's working properly I'll push the input voltage up to rectified line.....

Quote from: TinselKoala on June 24, 2012, 11:56:23 AM
@PW:
Six volts input to the ClassE driver board, about 1.3 amps, system resonating at around 330 kHz or so (varies according to environment and load, but the feedback loop keeps the thing in resonance). My primary is too closely coupled; I can get the power draw down quite a bit I think by looser coupling. Once it's working properly I'll push the input voltage up to rectified line.....

TK,

I would think you would want a nice clean square wave driving the gate, unless you are attempting to operate in a "linear" mode, which it sounds like you are not.  If there is anything close to a sine/triangle at the gate (slow rise/rise fall waveform), I would think MOSFET dissipation would be a bit high, and the overall efficiency low.

I wonder if a driver/comparator between the loop and gate would not be adviseable.

PW


 

Quote from: picowatt on June 24, 2012, 12:20:12 PM

I wonder if a driver/comparator between the loop and gate would not be adviseable.

PW

It's a good idea, and some autoresonator SSTCs that I've seen just use a cmos inverter or Schmitt trigger chip with simply a short little antenna on one of the gate inputs to pick up the feedback signal, that's then cleaned up and squared off by passing through a couple more stages then passed to the mosfet drivers. A square drive will give the best dI/dt and the best induced voltage in the secondary, and you know I don't want to be in the linear region. But a good ClassE performance will have the mosfet switching at or near zero voltage and so will minimize the losses in the mosfet itself, even if the rise and fall times are a bit longer, I think. So far this one is staying cool, even at 2 amps draw so I think I'm getting close. Removing a turn from the primary and loosening up the coupling a bit might get me right there.

I think the person that designed this circuit has gone on to bigger and better systems like PLL control. I like this one for its bare simplicity. If the ClassE waveform can be gotten precise, then there will still be plenty of voltage for corona effects when I boost the input and the mosfet should survive long runtimes. I'd like to be able to run it with good corona from a battery supply.

Quote from: TinselKoala on June 24, 2012, 11:24:48 AM
@Picowatt: You asked about Groundloop's H-bridge circuit. There have been several; some years ago he and Gotoluc and I were playing with some of his circuits and effects produced by resonating coil/cap arrangements. The one used in my "resonance effects for everyone" videos on YT is a proper mosfet H-bridge using actual mosfet driver chips and works very well at usable frequencies (tabletop sized 1/4 wave resonators). I don't have that unit available to me right now, it's at the other undisclosed location, unfortunately. The schematic for that one is on an ancient thread here somewhere; perhaps Groundloop is reading and remembers it and will post it; I don't have it in this computer at all.

The H-bridge circuit that blew the Darlington so spectacularly a couple days ago is something different... and I can't really recall who sent it to me. It could have been Groundloop or Gotoluc or another person with whom I was corresponding at that time. I am really sorry, I apologise for not recalling the person who was so kind as to send me the printed circuit board and a pre-programmed PIC for it. This one is optoisolated, driving the Darlingtons directly from the opto outputs, and uses the plc and a crystal oscillator for pulse shape control, and is clocked by an external FG input. I finally got around to building it up. I delayed for so long because the specified Darlingtons are the pricey MJH11022. But now I've just built it to play around with using TIP122s. I also don't have the schematic for this one, but the output side is dead simple: just pulldown resistors and the direct output from the optoisolators driving the Darlington H-Bridge, no caps or inductors, and it tops out at a couple hundred kHz... too slow for small TC use unfortunately.

I wish I could remember who sent this to me, I've learned a lot from it and I'd like to thank them again for sending it along. If it was GL, Thanks! And if it was someone else... please let me know, and forgive me for being so tardy in building up the board. I've moved locations several times since it was sent to me...

And finally, the Class E sstc that I'm tuning uses a single mosfet as an RF amplifier and is more similar to Ainslie's circuit than to anything else -- even in its use of feedback oscillations. Just as when I was working with the COP>17 circuit, I find that it's better and more interesting to switch mosfets PROPERLY and use them to make really INTERESTING effects. I started developing the TinselKoil circuitry and resonator when I got bored with Ainslie's BS the last time.

TK,

Yes, I remember this circuit. I made it back in 2008, I believe. And I did send out some few PCB's and PIC16F84A mcu's
and I did send you one also. I did search my backup and could not find the design files or the mcu firmware anymore.
But I did find a couple of images. I'm sure I did post the design and firmware on this forum back then. If I remember correct,
then the design was an attempt to build a high voltage h-bridge.

GL.

Quote from: Rosemary Ainslie on June 24, 2012, 09:22:37 AM

My data is freely available to any one in the whole wide world.  It is all of it transferable from my flash drive.  Send me your flash drive - I'll send you our data. No way can I make the information available otherwise.  There's just WAY too much of it to include it all in this or ANY forum.

Rosie Pose

Rosemary your such a damn liar and whats that brown smelly stuff all over your face ?  Do you need some scissors to cut a hole in your knickers so you can see again ??

So now Rosemary explain how does this "MAGICAL" file of my entire COP>17 device testing and evaluation "EXIST" ??           https://skydrive.live.com/redir?resid=6B7817C40BB20460!120   

Oh Ya  ..... look at the "copyright" note on the file named "Glens Tests - 001" with a US Copyright Office Registration Number TXu 1-804-221 ( some of these images might also appear in a Scribd document ) 

FTC