Something new I'd like to see / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Something new I'd like to see

1 2 next

Move Thread

LAN_403

Dr Hankenstein

Fri Nov 02 2012, 06:42AM

Registered Member #1642 Joined: Sun Aug 17 2008, 11:36PM
Location: Black Canyon City
Posts: 96

All right, folks, the title says it all. I,d like to see some real innovation out there; No more arm-chairing! I showed you how to high level plate AM modulate a VTTC @

; now I would like to see the same thing done with a solid state MOSFET coil. I think and agree that part of the "scariness" is the amount of heavy metal (iron) it takes to do it with tubes.

The concept is very simple. In the case of the tube model shown, I just took a standard 833 Tesla Coil driven with MOT's, used full wave rectification and alot of filtering (if you don't filter' you'll get 60~ hum (50~ if that's what you have), then I hooked up a 1KW modulator in series with the coils B+ line and the plate; that's all.

The beauty about high level modulation is described as follows: Assume the coil is running at 1KW input. At 100% modulation you'll produce 4KW peaks! You can't do this with any other type of modulation including FM, Class E, etc. You'll notice the flames really jump on modulation peaks.

So here's how to do it on the cheap:

For a 600 watt MOSFET Tesla Coil, use a mixer and a big car stereo bass amp around 400 watts will do. You'll want about 60% audio power for 100% modulation. Feed this signal backwards into a big filament xfmr on the LV side. I'm assuming a 6 or 12volt to 240volt/250VA xfmr would be a good start. Place the 240 volt winding in series with the coil's DC supply and the bridge buss; you're done. You're ready to produce high level AM modulation with your coil.

Another method would be to place a suitable choke in series with the coil's DC supply line and the bridge, and use the reversed filament xfmr in series with a capacitor and a choke (check out Heising Modulation for more details) Disadvantage here is a lot more iron required for this, without much gain.

This is a similar concept to the well known QCW coil, only you're using a xfmr as the power control and not a pass transistor. This concept was actually developed in the early 1930's and is still in use today, so it must be good.

I know a person who just completed this vary same thing a few months ago with a Kenwood 520 transceiver running a pair of 6146's in the final and man, did it ever improve his signal! That's what gave me the idea that it can very likely be done with a MOSFET Tesla Coil as well.

Hope you like my idea enough to get out of your arm-chair and make it happen; keep us all posted!

regards,
Hank

Steve Conner

Fri Nov 02 2012, 08:15AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Well... One reason for getting into solid-state coiling is that you can get surplus IGBT bricks cheaper than pole pigs. Especially if you don't own a truck and have to pay someone to move your pigs around.

Solid-state coilers have a kind of "anti-iron" attitude and wouldn't use a modulation transformer if they could avoid it.

There are plenty of ways to high-level modulate a SSTC without transformers. The simplest way is just by feeding it halfwave rectified, unfiltered mains, which bumps the peak-to-average power ratio up nicely.

The QCW is a more elaborate version of this, using a Class-D high level modulator (not technically a pass transistor, nothing running in linear mode could stand the power!) to produce any modulation waveform you like.

The only thing I've not seen done with a SSTC is a Class-H modulator using two supply rails, a low voltage one for the carrier and a high voltage for the peaks. I've seen it done in a ham transmitter at classeradio.com.

Personally, I'm waiting for a Harris 3DX-50 to show up cheap on Ebay, so I can hook it up to a Tesla resonator.

Kizmo

Fri Nov 02 2012, 08:23AM

Registered Member #599 Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 624

Steve Conner wrote ...

The QCW is a more elaborate version of this, using a Class-D high level modulator (not technically a pass transistor, nothing running in linear mode could stand the power!) to produce any modulation waveform you like.

Dont forget steves tesla gun which uses phase shifted h-bridge for this. Bus voltage is fixed and modulation is done by phase shifting one side of the bridge which does essentially same thing. No extra stuff needed except little bit more advanced gate drive and controller (and good igbts since one side of the h-bridge is always hard switching at turn off)

Steve Conner

Fri Nov 02 2012, 09:42AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Yes, even more real innovation!

I read a paper about a power supply with a phase-shifted bridge using IGBTs for the soft-switching side, and GaN FETs for the hard turnoff side. It's used in a 30kW Class-D audio amp that fits in a 1U rack. :O

Dr Hankenstein

Fri Nov 02 2012, 03:29PM

Registered Member #1642 Joined: Sun Aug 17 2008, 11:36PM
Location: Black Canyon City
Posts: 96

Good point, Steve.

I've built Steve Cloutier's Class E modulator described in classeradio.com and drove my 833 coil with it using a CM600 with a switching speed of 73khz on the CM600 gate. The CM600 was connected to the cathode of the tubes. Problem with this scheme I believe is actually the CM600. Just about when the modulation would start coming to life, I would be near the maximum voltage rating of the brick, close to 1200 volts.

This is really a series modulation scheme and you need to expect that 1/2 the voltage should appear across each of the devices; the modulator tube (or brick) and the actual Tesla Coil. Now using a tube of equal size of the one being driven would likely solve this problem.

Another issue with this scheme is that for a "normal" coil running at 3kv, you would need use 6kv rails for 100% modulation, 3kv of this would be across the brick; not good.

Also, if I remember correctly, the QCW coil uses a BIG chunk of iron to make it work; there's no getting away from the fact that you will need a transformer.

Besides; you're getting away from the theme of this post. The theme is to make a low budget "plate modulated" style Tesla Coil on the cheap; so get out of that arm-chair and do it!

Regards,
Hank

Dr. Dark Current

Fri Nov 02 2012, 06:26PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

If power is what you are controlling, then using a voltage amplifier to drive a resistive load will produce distortion. There is no advantage to "producing 4kW peaks with 1kW idle input power". The SSTCs usually behave close to a constant-current sink, so the distortion would be lower. Anyway, "slope detection" FM works pretty well on SSTCs. You could also use dead time or phase shift modulation, in all cases the modulator is built into the bridge. It would operate out of the optimal condition but it's all right as long as you size the components accordingly.

Dr Hankenstein

Fri Nov 02 2012, 08:48PM

Registered Member #1642 Joined: Sun Aug 17 2008, 11:36PM
Location: Black Canyon City
Posts: 96

Dr.Dark,

There seems to be a misunderstanding. I am modulating the B+ rail, NOT controlling power, 1kw is NOT the idle current; it's the full load output of a particular coil; the max you can give it, whether it be solid state or 811;s 833's, GU81's, etc. doesn't matter. When you 100% plate modulate a 1kw coil for example, you get a 4kw PEP output signal. period. You also get the same amount of "zero" downtime, so the average output is 1kw.

I'm a believer because I actually did the experiment and it's quite astonishing to see.

On another note, what I find most interesting about this post is that nearly every single reply was basically telling me how to build my own experiment; Not one single question about how I did it. And further more, I'm not convinced that any of the suggestions have even been proven.

Would it be asking too much to actually be innovative?

Kind regards,
Hank

Steve Conner

Fri Nov 02 2012, 10:16PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Well, you're high level modulating a tube RF amplifier and calling it innovative. But the technique dates from the 1920s. It's even in the first edition of the ARRL handbook which I'm reading in my armchair.

It was a cool experiment and I enjoyed the YouTube vid, but I didn't have to ask how it was done.

Any modulation scheme you can use in a radio transmitter, you can use on a Tesla coil, as it's nothing but a radio transmitter with a very poorly designed antenna. Radar pulsers and klystron modulators are fair game for inspiration too.

100% plate modulation certainly gives you 4x output power (into a resistive load, if not a streamer load) but it achieves this by a temporary doubling of the plate voltage. The stage being modulated has to be designed stronger to withstand this, so the 4x power boost isn't free. Tubes handle it without too much fuss, but say you were using MOSFETs, you might have to go from 500V rated devices to 1000V and suffer a big increase in Rds(on).

Dr Hankenstein

Sat Nov 03 2012, 12:50AM

Registered Member #1642 Joined: Sun Aug 17 2008, 11:36PM
Location: Black Canyon City
Posts: 96

Now you're talking!

Thanks for the kind words, Steve.

I'm going to go back to my arm-chair as well and relax!

Kind Regards,
Hank

Sigurthr

Sat Nov 03 2012, 10:45AM

Registered Member #4463 Joined: Wed Apr 18 2012, 08:08AM
Location: MI's Upper Peninsula
Posts: 597

I've been wanting to do this for quite some time to my 1.5KVA SSTC but I lack several necessary components: Audio Amp, modulation/filament transformer (must handle 9A across 120v winding), and about 10,000uF rated for 500V. Sadly most of this stuff isn't available locally or inexpensive.

1 2 next

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