A Student's Attempt at the Plasmasonic Headphones.
Jul 23, 2012 at 1:59 AM Thread Starter Post #1 of 57

akiroz

Previously known as b1o2r3i4s5.
Joined
Feb 8, 2012
Posts
14,356
Likes
302
Before I begin, I shall introduce myself as a high school student who's really interested in both electronics and audio. Experts out there, please feel free to correct anything incorrect information mentioned here. Also if anyone have any questions or suggestions, feel free to ask. (I'm a Head-fi regular)
 
Alright, Some of you have probably heard of the plasma headphones. It's been discussed here and there around the web (even here on Head-Fi) and there once was even a commercially available plasma headphone. (Source) The Plasmasonic1. 
 
The Plasmasonic1 works by controlling ions created by a corona discharge using a variable electric field as opposed to plasma speakers (Tweeters) that use plasma arcs.
 
After looking through the designs for the Plasmasonic1 it seems like a relatively easy build but I see no attempts at this device... There may be quite a few health hazards but I see people doing even more dangerous things online (e.g. Tesla coils, rail guns Lasers... etc.) So I've decided to take on this project, my goal is to replicate the Plasmasonic1 using info provided on their site.
 
This project will probably be on-going for a few months to around 1 year because I have school work and IB exams coming up in 10 months. So I will keep this thread updated as I slowly progress in this project.
 
I have actually already begun the project before writing this post. According the the Plasmasonic1 page, the system requires ±3KV @ 500µA. Since the current required here is not very high, I've designed a HV PSU Based on the Cascade voltage multiplier.
This PSU Consists of a step-down transformer to convert 220Vrms Mains to 110Vrms and 2x 20-stage Cascades to reach the required ±3KV. Each cascade stage consists of a 400V 33µF electrolytic capacitor and a 1N4007 diode. 
 
Here's some photos of my PSU under construction:
 

 
That's all I have for now, I'll update this post again after completing the HV PSU.
 
UPDATE 25/7/2012:
I have just completed the HV PSU along with a 1/100 Potential Divider to measure high voltages.
I a]have also gotten my hands on some Metal Gauze to make the Electrode Grids and "Safety net" Grids
Photos of Completed HV PSU:
 

 
Photo of Potential Divider and Metal Gauze:
 

 
UPDATE 26/7/2012:
- Calibrated the potential divider using a low voltage reference.
- HV PSU Output voltages slowly reaches ±2.9x KV over few minutes after power on.
- Corona Discharge test:
   - Used multimeter probes connected to PSU Outputs, tip of probes spaced 1cm apart
   - Observations: No light observed, even without any external light source. Heard "Hissing" sounds.
 
UPDATE 27/7/2012:
- Bought some HV Wiring (25KV rated) and some banana plugs to make Interconnect.
- Decided to use Acrylic glass to make transducer body for easy manufacturing.
- Project On-Hold until my school opens. (So I can "Hijack" the workshop's laser cutter 
tongue.gif

- Some Pics of HV wiring and interconnect:
 

 
Further Update 27/7/2012:
Just before I put this project on-hold, I decided to do one last discharge test with a round piece of metal gauze and one of my mom's sewing needles. This time, I got some Great results!!! I'm finally able to see the discharge glow at the tip of the needle and less hissing sound!! This is looking great got my headphones.
Pic of Test setup:
 

 
Unfortunately I can't take a picture of the discharge glow because my camera won't detect such low levels of light. It actually took my eyes a while to adjust to the dark before I can even see it myself.
 
UPDATE 28/7/2012:
Alright.... I really wanted to put this on-hold and do some work but I really wanted to see some more progress and results so I just designed the Amplifier for this plasma headphone project.!!
 
The Input stage is based on the BB: OPA2134 for a gain stage and balance converter (pretty standard stuff) then the Output stage is based on the Toshiba: 2SC4686A High Voltage NPN Transistor (each 1200v max.) configured in a 4-stage Cascode with voltage ladder. The quiescent current (Iq) for each one of these cascodes is 120µA, 4 total (2 per ch, 2ch). 4 stages were chosen because I decided to play it safe and have just 700V across each transistor.
 
PS: For some reason, my auto-correct does not recognize cascode as a word.....
 
UPDATE 6/8/2012:
The 2SC4686A High Voltage Transistors has finally arrived!! 20 of them!
(Surprisingly made in Japan.)
 

 
UPDATE 8/8/2012:
Drawn schematics I have so far in computer.
They can be found here:
http://www.2shared.com/document/1iqo3gtT/Amp_sch.html
http://www.2shared.com/fadmin/34098436/e28a1870/Amp_sch.pdf.html
 
Jul 23, 2012 at 1:03 PM Post #2 of 57
you're a madman! but I cant look away.. godspeed!
 
Jul 23, 2012 at 3:35 PM Post #3 of 57
Yikes!!

All I'd like to say is be extra careful with this thing!
I can imagine this achieving superb transient response, not too sure how good bass You might get though. Please do report back on how it sounds. lol
 
Jul 23, 2012 at 4:00 PM Post #4 of 57
You might consider some sort of switching supply instead of a multiplier. Or at least use smaller caps. I see no reason to have that much current available, and it will blow you across the room if you were to touch the DC supply probably killing you in the process. Also, if it were to arc in the headphones, you'd almost certainly cause permanent hearing damage, possibly severe.
 
While these are both manageable safety risks with proper planning, I see no reason to take the risks since there are much more appropriate supplies for a project like this.
 
 
Edit: Just saw the caps are only 33uF, they visually appear much smaller. Although, I think everything I said above still applies, just not quite as extensively.
 
Jul 23, 2012 at 6:41 PM Post #5 of 57
@dude
You have a point, thanks for your concerns.
However, the multiplayer circuit was designed for 3111V output with 1% Ripple and 8% Regulation @ Full load (500µA) This means the voltage will drop to 2886V when 500µA is being drawn, and these are ideal calculations, it's probably lower in real life.
Now, I haven't exactly designed a switching supply for this but I'd imagine the inductor value and duty cycle required would have a ridiculous value to step all the way up to 3KV. But if you do have a design which will provide the required ±3KV @ 500µA Please let me know and I'll consider using it instead.
Thanks!
 
Jul 23, 2012 at 9:28 PM Post #6 of 57
Quote:
 
Now, I haven't exactly designed a switching supply for this but I'd imagine the inductor value and duty cycle required would have a ridiculous value to step all the way up to 3KV. But if you do have a design which will provide the required ±3KV @ 500µA Please let me know and I'll consider using it instead.
 

I'd suggest a flyback converter if you want to consider a switching power supply. You can buy modules that will do this sort of thing for a few tens of dollars. Also, you could make your own without too much difficulty after a bit of studying in power electronics if you aren't already experienced. 4hv.org is an excellent resource to learn about that sort of thing. The magnetic component could be tiny, this is only 1.5 watts and the size of magnetics is (in general) determined by power, rather than voltage. 
 
Either purchased or DIY with a careful PCB, the power supply could be smaller than the size of one of your capacitors.
 
Jul 23, 2012 at 9:44 PM Post #7 of 57
Typically you hold over ear phones in both hands when putting them on, establishing ideal conditions for a shock across the heart when as little as 10 microamps can cause fibrillation which can prove fatal without defibrillation.
 
For what? You can get more than adequate volume and quality with 3V in IEMs.
 
w
 
Jul 23, 2012 at 11:31 PM Post #8 of 57
@dude
What a flyback converter that size with the same output capability? Not totally convinced....


@Waki
Yes, I understand this can be quite dangerous, and the results may not be very good. But the point is to prove it CAN be done!!
I've seen no evidence of this ever being done besides the site about the Plasmasonic1.
 

Users who are viewing this thread

Back
Top