Plasma speaker

Le Quack said:

Wouldn't it be more beneficial to make it a step UP transformer for the audio transformer?

Click to expand...

There are at least two problems I see with that.

Firstly, the output of an amplifier is already several volts if not several tens of volts. If you step that UP, you get way too much variation, and you'll end up killing the arc or even sending the driver into negative which will probably blow it up.

Secondly, the low voltage windings are much thicker than the high voltage windings. The high voltage windings will waste a lot of power from the main DC power supply

Le Quack said:

And ZVS drivers aren't necessarily the best for plasma speakers

Click to expand...

Probably not. But if you've already got a ZVS driver (and many of us do) then it's a simple addition. Realistically, plasma speakers are pretty terrible speakers anyway and it's just for fun.

That's interesting, never thought modulation could be so simple!

As far as blowing MOSFETs go I think I've found a sweet spot. When being used as a plasma speaker it can stay on for over 10 minutes, barely gets warm and it's quite loud.
I just finished optimizing the pre-amp, will solder it up and post schematics as soon as I'm done. Oh, I also forgot to mention that the pre-amp picks up a radio signal if you just attach a long wire to the input - it's kinda funny actually


Here's a musical instrument I cooked up with it - I call it The Arcsynth:

It uses an Arduino to sense touch (capacitance, actually) on a cardboard with tin foil keys.

It's not perfect (yet), which kinda suits my lousy playing XD Might make next version bigger to make it easier to play..

The keyboard has one octave (12 keys) plus 4 buttons (lower all keys by one semitone, increase all keys by one semitone, record and play). The video doesn't show but double tapping play will make it loop whatever was recorded. It also doesn't show that there's an option to record to permanent memory, which means the song will still be there if you shut it down.

You truly have created a unique instrument! Great work! Thanks for posting the progress.

Every ones builds are great to see!
I would love to undertake a like build when I get some time from work and I think everything needed for information is here in this thread.:beer:

I'd love to see your preamp schematic, I'm in need of a good preamp.

Cyparagon:
I think Le Quack was thinking you were feeding the high impedance audio signal from a headphone jack in to the 120V winding, at least that is what I had thought you meant as well. The output of a headphone jack is usually less than 200mVp-p, so feeding that in to a stepdown transformer would probably not modulate the signal of something like a ZVS driver very well. If you're using an audio amplifier with a low impedance output you might be seeing as much as 24Vp-p from even a 12V low power car amp, which through a 9:1 stepdown would be at least a couple of volts at high current modulation.

I was toying around in my head with the idea of using a transformer to modulate the DC Bus voltage on my SSTC for audio modulation... but then I realized I would need a 1500KVA transformer! Driving a 1:1 modulation transformer with a 5 to 25Vp-p signal would be ideal for most SSTC work, but finding one with the KVA rating needed would be very difficult. I am thinking about making one from two MOTs with the secondaries in parallel as this would give me the needed isolation voltage and KVA rating. I'd still need an amplifier/pre-amp to feed audio to it though (and a second MOT as I only have one). This would be pretty ideal modulation though as you could vary the quescent bus voltage for audio control and the TC would operate in true CW manner.

Modulating the power rail directly like that requires a bit of power - probably a lot more than a headphone jack can supply.

Thats one reason i like the 4046 PLL , when running a half/full bridge for a flyback or sstc its very easy to modulated the output with a audio signal from pritty much any ipod/cd player , raido Ect . And the sound out is very good

Here's the schematics for the whole thing. This is probabably nowhere near the ideal circuit but it works pretty well for me and is *very* stable (can leave it playing music for over 10min without heating the MOSFET much).


As requested here are some pictures.. swear I'll never ever solder "on air" again, I'd rather leave a breadboard inside the box next time!

Oh god, please don't solder on air. That's just asking for shorts!

What may be a neat idea however is to do the same thing, except trim the wires neatly, and then maybe pot the circuit in epoxy or something similar, kinda like this; (well if I could find a picture of what I was trying to find anyway...)

lawl spаghetti.

I think they have the emitter and collector backwards on that PNP (MPSA92).

It's called "Point to Point Construction" and it is a nightmare, haha. Excellent work though!

Your preamp is basically a linear single ended amp, which is great for low power stuff like you're doing with modulating a 555 via pin 5. For some reason I thought you were using a dual rail voltage preamp based on opamps or even a push/pull bipolar topology. Sometimes the simplest solutions evade us.. Again, nice work!!!

Thanks, Sigurthr!
The preamp idea was from Alex1M6@YouTube, I just tweaked the resistor values until it sounded best. Seems like that kind of amp is very common (pun intended).

Glad to know they have names for the ugly stuff as well XD Never gonna do that again, seriously
Couldn't really visualize the epoxied circuit

Thanks for catching that, Cyparagon, it's fixed now.

I threw one together today from spare parts/boards.

Class-D Amplifier; Audio Modulated Flyback "Plasma Speaker" & trad Speaker - YouTube


http://youtu.be/96VuQuqY_ss


I mostly wanted to test my audio modulation board and repurpose an experimental mosfet driver board, so don't expect much. Due to the use of a beefy TO-263 MOSFET and single ended flyback topology on a LOPT with a 10turn primary the transistor runs cool indefinitely. Since I wanted to test the audio modulation this setup is configured mostly as a Class-D amplifier, and not as a plasma speaker. The flyback transformer was used mostly because I needed a large gapped core and it seemed handy.

Up!

What about using a square wave to drive the reset pin of the 555? Wouldn't that work better for synthesized notes than driving pin 5? It wouldn't affect frequency/duty cycle and would probably provide a nice non-CW output..

I think I'm gonna try this with the arduino output!

I'm sorry. But that plasma speaker you soldered is just too funny! Good job on making it work though!

Thanks and feel free to laugh