Plasma speaker/singing arc using a Flyback / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Plasma speaker/singing arc using a Flyback

1 2 3 4 next

Move Thread

LAN_403

Dr. Dark Current

Mon Sept 10 2007, 05:23PM

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

Hi all, I've been recently experimenting with the singing/musical arc using a flyback.
At first I tried single FET driven with pulse width modulation, but the distortion was unacceptable. Then I tried a bootstrap-style halfbridge (ala Steve's brick driver), now the sound quality was a bit better and louder, but still with some distortion. Next to try is classic halfbridge with dead-time modulation, I'll try if higher powers result in better sound.

If everything fails I'll probably try linear modulation with a big transistor on a large heatsink, lots of wasted heat but I want good sound at decent volume

(and build a demo unit too...)

Has anyone experimented with a singing arc before? If so, what circuit worked best for you?

J.M.

Edit> do you think a saturable reactor is a good idea?

ragnar

Mon Sept 10 2007, 10:25PM

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

A single MOSFET in series with the centre-tap supply works well. Make sure you play around with the bias plenty -- you'll need to float it juuust under 4V to get good sound.

Steve Ward

Mon Sept 10 2007, 11:49PM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

From my experience with designing amplifiers (including class D amps) is that some sort of negative feedback to the input of the amplifier (from its output) will be necessary to really achieve the best possible sound. I think one problem is that the arc producing device is likely not a linear system, likely in the manner that it will cause some form of clipping. So im not entirely sure what the best approach is here, but i would start with simple voltage feedback from the output of the driver and through a LPF. If the flybacks sound output is linear then this will work well, otherwise you might need some method of compensating for this.

Also, in my opinion, making sound from a flyback arc will never really sound "good" partially because the carrier frequency is only ~2x the highest audio band frequencies. I need to rig up a SSTC to the class D amp im building and see how that sounds

Avi

Tue Sept 11 2007, 01:30PM

Registered Member #580 Joined: Mon Mar 12 2007, 03:17PM
Location: Melbourne, Australia
Posts: 410

i have managed to get quite good results by using a 555 driver and placing the audio signal via a 10nf capacitor to pin 5 of the 555.

Dr. Dark Current

Tue Sept 11 2007, 07:29PM

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

Steve Ward wrote ...

Also, in my opinion, making sound from a flyback arc will never really sound "good" partially because the carrier frequency is only ~2x the highest audio band frequencies. I need to rig up a SSTC to the class D amp im building and see how that sounds

hmm, I thought that if you sample audio at x hz, you can reproduce all frequencies all the way up to x/2, so setting the freq. to 40kHz should be plenty... am I right here?

Dalus

Tue Sept 11 2007, 07:41PM

Registered Member #639 Joined: Wed Apr 11 2007, 09:09PM
Location: The Netherlands, Herkenbosch
Posts: 512

Steve Ward wrote ...

From my experience with designing amplifiers (including class D amps) is that some sort of negative feedback to the input of the amplifier (from its output) will be necessary to really achieve the best possible sound. I think one problem is that the arc producing device is likely not a linear system, likely in the manner that it will cause some form of clipping. So im not entirely sure what the best approach is here, but i would start with simple voltage feedback from the output of the driver and through a LPF. If the flybacks sound output is linear then this will work well, otherwise you might need some method of compensating for this.

Also, in my opinion, making sound from a flyback arc will never really sound "good" partially because the carrier frequency is only ~2x the highest audio band frequencies. I need to rig up a SSTC to the class D amp im building and see how that sounds

Why don't you build a tesla coil driver as a class D amp. You could use the feedback signal from the coil as frequency to modulate using a comparator as in a standard class D amp.

Bjørn

Tue Sept 11 2007, 08:27PM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

hmm, I thought that if you sample audio at x hz, you can reproduce all frequencies all the way up to x/2, so setting the freq. to 40kHz should be plenty... am I right here?

You are right, but you also need a very good control of the amplitude to get good quality sound.

With a simple driver you det all sorts of distortions and the control of the amplitude is not good. Increasing the frequency usually helps a lot. The distortions gets pushed into ultrasonics and you can average several cycles to get the correct amplitude.

Dr. Dark Current

Wed Sept 12 2007, 04:59PM

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

I tried the half-bridge with dead time modulation, I dont know why but the sound was absolutely unlistenable. I'm probably going with linear modulation...

Dr. Dark Current

Sat Sept 15 2007, 06:58AM

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

Some more findings: more power does not necessarily mean better sound. Actually, I was surprised how easily flybacks saturate. Going from 10 primary turns to 15 turned the sound from nearly unlistenable to pretty good.
When we're drawing those 5"+ arcs from the flyback, the core is actually heavily saturating, but we don't even notice it, it just gets hot. When I run flybacks like this I use something like 7V/turn (50kHz), but with the singing arc I get clean sound under 3V/turn. This also means you can't output much power so the sound level will be low. I think that's just the best sound you can get from a flyback...
I'll try different flybacks but I think it will not be much better.

As to the driver circuit, the bootstrap half bridge made the clearest sound in the end, better than linear modulation with bipolar transistor.

Marko

Sat Sept 15 2007, 04:01PM

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

Edit> do you think a saturable reactor is a good idea?

That's exactly what I intended to use in order to tune or even modulate a royer oscillator; but when I googled a bit it appears that they aren't very linear. And when you are hard switching anyway it's a little of benefit, deadtime modulation is much more grateful.

I'm although considering it for tuning, if it ever comes to that.

I would do such a circuit simply like a class D amp, using higher-than-needed frequency (like few hundred kHz) and don't push the power (volume) too high. You don't expect great output from it so you don't need big mosfets either. Things like IRF730 are fast and more than powerful enough.

Even if transformer doesn't saturate you may blow it; don't expect the circuit to produce great amplitudes at low frequencies. It's more like a tweeter.

1 2 3 4 next

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