Class E SSTC thread / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Class E SSTC thread

1 2 3 4 next last

Move Thread

LAN_403

Marko

Tue Jul 04 2006, 10:40PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

I've just done some 'higher power' tests on my small class E coil. It's running at 24V and 30-40 watts, have to measure this yet. I can get 6,5cm strikes (almost 3 inches) from this, tank voltage gets up to about 100V peak for now.
Interesting thing is that arc, after connecting, immediately wanders away, and coil continues to break out from topload on it's own for a while (completely in CW). It even pwned my old mini SGTC. Coil won't break out from topload if nothing is near, but I guess I need just a littel more voltage for that.

I like this shot, and there is also a video of this I can upload if anyone is interested.
This isn't a good stunt for a very flammable celluloid ball topload so I kept runs limited.

Now i'm willing to put it back into it's old wooden case.

I want it to be portable and power it from an external 12..15V source, and I don't mind dragging big mains transformers around or within the coil.

I always wanted to have a power supply inside the coil, wich would be on a relatively small board (not more than about 4*8cm) and provide voltages required for class E stage.
It would need to produce about 100 watts, and nice thing would be if I could regulate it somehow (0-150volts)

mazzili ZVS would be very simple, good and neat, requiring lower voltage mosfets and being highly-efficient.
Bad thing would be requirement of additional circuit for regulation/modulation.

Other things I could think off are boost converter, requiring big IGBT being more lossy and possibly having or maybe some type of deadtime-regulated brdge using TL494, wich I haven't explored yet.

I'd also like your ideas here.

I also imagined I could use the power supply for audiomodulation (altough PLL does for lower powers, this would be of higher quality and suit narrowband class E driver more)
There are also some more problems like filter capacitor slowing modulation down, wich I need to study more.

One more thing I'm interested is that I now want to interrupt the coil, but I've recently heard that this leads to problems with this type PLL circuit (Conner's first circuit).

I wondered if I could use audiomod or inhibit input on 4046 chip itself, or do something else to shorten duty cycles without de-locking PLL.

rgs,
marko

ragnar

Wed Jul 05 2006, 01:21PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

I'd love to see a movie, uploadie, uploadie! =)

Failing that, you can email it to me, hehe

have you got any waveform pics? :)

Marko

Wed Jul 05 2006, 02:17PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

OK, here it is

I was very suprised to see the coil break out from it's topload in CW at such low power input.

I updated my post in

after I got my camera working again with waveforms and stuff.

It looks like almost perfect ZCS operation. On low power run I got 3.3 Vin tank voltage as predicted, and on 24V it hits 100V (some 4 Vin ).
Maybe I need to adjust coupling a bit more.

It's 500ns div (freq about 1,4Mhz), gate voltage is 10Vdiv, DS voltage (sine) is 50V div.
ClassE stage supply is 16V.

ragnar

Thu Jul 06 2006, 02:56AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

For the power supply, if you're feeling bored and inventive, you might consider building a big solid-state marxgen / charge pump. =P

i.e. charge caps in parallel, switch FETs to put them in series through your load... I'm gonna get flamed for this one. ^^

ragnar

Thu Jul 06 2006, 09:53AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

Hehe, I admire your oscilloscope

I realised that I'd forgotten how elegant class-E was, so I just made a little lashup today with a single isodriver of mine (substituting a diode for the shunt capacitor).

Will write more soon =)

WaveRider

Thu Jul 06 2006, 01:09PM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

Goodonya BP and Firkragg... I've been lurking a bit and your work is looking great!!!

(I'll try to get back to class-E stuff this autumn...too much other work at the moment!...)

Marko

Thu Jul 06 2006, 01:12PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Looks good bp, I hope you'l give more specs soon.
I'd say this is very good if input voltage is 12..15V.

Class E is somewhat frustrating because you must always do everything to suit output amplifier, you have to reduce coupling even if higher one makes longer sparks, tune it for streame/no streamer and etc.

Big plus is posibillity of running cheap MOSFETs on very high frequencies.

All that isolation you have is probably unnecessary for class E (since there is only one fet).
I don't know if I should include a small GDT for isolation in my circuit..

Currently I was trying to make a transformer for power supply, but I had bg trouble getting it apart.
I cooked out a hell of small ATX power transformer but ther was too much epoxy.
I guess I'l have to use the transformer in it's current state (means ZVS).

I'l have to add a 'SS variac' separately

For now I can trygiving the coil more power, I can double a bigger 24V transformer to 60V (i'm a bit scared t blow it up now, and especially for gate driver wich I have only limited supply)

Also BP, are you running your gates a bit sine? Or that's some other signal?

Guessing upper is DS voltage your class E looks pretty good.

For the power supply, if you're feeling bored and inventive, you might consider building a big solid-state marxgen / charge pump. =P

i.e. charge caps in parallel, switch FETs to put them in series through your load... I'm gonna get flamed for this one. ^^

A solid-state powered Cockroft Walton is how very efficient solid state step-up converters are made. But it's not practical for my app.
Marx would be possible but it would require a lot more switching devices..

ragnar

Thu Jul 06 2006, 01:51PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

hehe, I've seen identical Hung Chang and Labtech 3502 oscilloscopes before.... Hung Chang was there first, right?

Tomorrow I'll be clearing up some mess, doubling my gate-drive power, possibly doubling my frequency, and then doubling the voltage swing by making a two-ended class-E drive.. And it'll probably result in quadrupled silicon deaths. =D

I daren't push these little IRF830s any harder... ^^

hmm, I wonder if that little spike in my DS waveform corresponds to the same miller spike in my GS waveform.

Marko

Thu Jul 06 2006, 02:09PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

I see you are going to sleep now

And I realized about your scope before...

I built a double ended-driven coil too but spark wasnt't actually 2x big. But it can indeed help you to put more power into resonator. What input voltage/power is he currently running at?

EDIT: I've testedmy current ZVS driver circuit on ATX transformer as it was, but I was unable to get desired voltage output. (ofcourse).

I could get 200V AC minimum, and it is pretty stable, when loaded with 40W bulkb it drops for just few volts. To bad I need some 120V or less as maximum at full power.

Epoxy is too hard and I could never get the core apart without destroying it

I'l have to find something else around...

,

liquidFire

Thu Jul 06 2006, 10:19PM

Registered Member #479 Joined: Thu Jul 06 2006, 07:06PM
Location:
Posts: 3

Firkragg wrote ...

I want it to be portable and power it from an external 12..15V source, and I don't mind dragging big mains transformers around or within the coil.
I always wanted to have a power supply inside the coil, wich would be on a relatively small board (not more than about 4*8cm) and provide voltages required for class E stage.
It would need to produce about 100 watts, and nice thing would be if I could regulate it somehow (0-150volts)

I've done a similar thing last winter, it's a rather small class-E coil with on board power supply, all powered by a 12V battery. I thought about wearing that thing as a hat, but I think there will be a grounding problem then...
The PCB is 100 x 120mm and contains all needed parts, including the secondary.
It's a very simple boost converter that powers the class E amp and it's not running very stable, but good enough for a tesla coil.
Picture 1
Picture 2
Schematic
Video using some audio modulation

1 2 3 4 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob