Senior Project Tesla Coil and Hello all / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Senior Project Tesla Coil and Hello all

Move Thread

LAN_403

YertNamreg1218

Wed Dec 17 2008, 11:27PM

Registered Member #1867 Joined: Wed Dec 17 2008, 11:09PM
Location:
Posts: 2

Hi everyone! I've lurked a little bit, but now that I have something to show I figured I'd introduce myself. My name is Trey and I'm a senior computer engineering major at Rose-Hulman Institute of Technology in Terre Haute Indiana. For our senior design project my team is building a CW audio modulated tesla coil.

So far we have built a proof of concept device that successfully played audio through an arc using a flyback transformer. The proof of concept was driven using a full bridge and controller that was based off of McCauley's PlasmaSonic. That was back in October.

We have moved forward and have a much better h-bridge (read smaller, so lower inductances). We are still driving it using a tl494, so its not operating at its resonant frequency and I'm concerned about cranking up the power and blowing stuff up.

In the future we plan to replace the tl494 with a primary feedback system so the coil always operates at the correct frequency. Then to inject the audio we are going to build a modulator power supply. The picture show our current setup. You might notice a weird cage looking object in the background. Our professors made us build a "safety cage" to help ensure none of us got killed. Its nothing too special, but it should help to protect from spurious RF emissions.

If you guys have any thoughts or suggestions I'd love to here em.

Anon01

Wed Dec 17 2008, 11:49PM

Registered Member #1589 Joined: Sun Jul 13 2008, 06:40PM
Location:
Posts: 70

Welcome Trey German, amused by your screen name. :P

Well, I like the setup, it looks a bit big for 120V mains voltage, that being said, what do you plan on using for the power supply, rating wise. Why do you not do a PWM controller since you're already using the TL494 instead of a supply modulated design?

Right now I'm doing the basic UCC based primary feedback based audio modulated SSTC by sending squared audio to the enable pin. I plan on moving the TL494 or adding the driver next to this one and switching back and forth depending on how I want it modulated.

Thanks,
Sam.

YertNamreg1218

Thu Dec 18 2008, 12:18AM

Registered Member #1867 Joined: Wed Dec 17 2008, 11:09PM
Location:
Posts: 2

Ya I purposely built it way to beefy, so that its harder to destroy by me being an idiot. After I graduate the ECE department at my school will be using it as a demo for both current and prospective students.

For our supply we are using a full wave rectifier and filter cap to generate ~170VDC. Obviously I have a variac ahead of that to control power input to the bridge. Right now we are actually using a small benchtop supply to power the bridge while we are getting everything tuned in.

I learn new things about TCs everyday, and I'm not sure thats exactly how we'll end up driving it towards the end of the school year. I'm taking a DC/DC power supply class so the reason I'm leaning towards modulating the supply is I will actually have the knowlege to completely design that on my own (professors like it when I design things).

From past experience, what is the best way to get high quality audio at the loudest volume?

Anon01

Thu Dec 18 2008, 12:28AM

Registered Member #1589 Joined: Sun Jul 13 2008, 06:40PM
Location:
Posts: 70

Well, there's a trade off. PWM modulation by far gives you the best audio, but it isn't too loud and does not give the streamer length compared to disruptive mode. Disruptive mode though gives a much louder output and gives a lot hotter longer streamer from the output in my opinion. I have not designed/interpreted the PWM driver for my SSTCs yet, though I used on a flyback array some time back.

Bottom line:
PWM:Much better audio, but not as loud. Gives a corona discharge to free air.
Disruptive: Rough quality at best, but very loud at power levels >1kW, though anything above 80VAC rectified is pretty loud due raw air displacement at audible frequencies.

I run my coils on 120V AC in though a variac, and it gives off about a foot max, sometimes it does 1' 3" with the right frequencies. I am going to upgrade mine to fullbridge and use 240V AC in. Sure, I'll have to double my primary turns, but it will increase output considerably with more V put in. My coils are overly large for their output as well. 4.25"x20" 30AWG is the winding length. :P Could handle 3kW or so with cooling of the base of the secondary, since it will pass amps to ground. @_@

Personal opinion is to go with driver side modulation than the bridge supply, but if you have a SMPS that is practically ready for modulation, go right ahead. :P

It really does depend on what you're aiming for.

Thanks,
Sam.

...

Thu Dec 18 2008, 01:09AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

The first step to getting high quality audio is to make the coil small, so it has a high Fres--which goes against the make it loud. At 500khz there is a ton of hiss, up at 750khz it becomes bearable, but really you need over 1MHz to sound even close to rivaling a speaker.

But, if you just want to make tones, and use a musical interrupter, the hiss from low frequencies does not hit you nearly as hard because you only have output when you are making sound--which will drown out any hiss. For a coil this large I would recommend a eastern voltage research's xesnoonic or the likes--which won't give great sound (don't even think of playing voices) but will be very loud and will play melodies very well.

But it is possible to get decent ''hi-fi' sound of a coil running a normal fullbridge, see my pllsstc

which is about the same as a pair of laptop speakers (although it is much louder, it will fill a classroom no problem). I will warn you, it takes some tweaking to get good, loud sound out of a pll coil--mainly setting the center frequency and uppler/lower limits of the tuning range very carefully to get a good modulation level.

Steve Ward

Wed Dec 24 2008, 05:38AM

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

The main problem with audio modulated SSTCs, in my opinion, is that tesla coils arent linear devices. Neither are loudspeakers for that matter, and thats why its such a bitch to find really good ones!

Consider that loudspeakers have been tweaked for centuries to make them fairly good voltage to displacement transducers, with low distortion and fairly decent efficiency.

With a tesla coil we first throw efficiency right out the window. Our power goes into making plasma, not sound.

Secondly, you will most likely find that the volume (or air displacement) of plasma probably isnt related linearly to bus voltage, or PWM %. So most basic audio-modulated SSTCs are pretty bad for distortion.

While you certainly could modulate the power rails, i see this as possibly extra system loss as the power is processed more than once. PWM modulation of the HF driver directly is most likely more efficient. In any case, i think it would be fantastic if someone actually tried to linearize the system to produce decent sound quality!

The main issue i see with bus modulation is from a controls perspective. You can only *increase* the bus voltage realistically (usually there is some form of output rectification on any DC converter) and you must rely on the coil to draw power to lower the bus voltage. This could prove difficult to control, particularly if you havent identified the "transfer function" for a tesla coil (at least its impedance vs voltage or something). It becomes quite the nasty problem when you factor in all of these details, and i think this is why no one has really done it "right" yet... its downright hard!

But if theres anything i learned from my senior design was that the prof's really dont care if the thing actually works right in the end. They like to see you struggle for a bit and at least gain insight into the problem, perhaps develop some solutions (even if they fail, you should document them), and then give a long-winded presentation at the end. It really didnt matter that the self-oscillating class-D amplifier i designed actually sounded *really good*, what mattered was that i went through a process and documented it. Perhaps you already realize this, or perhaps (better yet) your senior design isnt like mine :P.

Having done amplifier design and SSTC design, i have a pretty good idea of what makes either one of them work well. To make a truly high fidelity audio modulated SSTC is gonna take more than luck!

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