Is It Possible To Shape An Electric/Plasma Arc?

[Lancelot59]

I was watching the visualizer while listening to a song in winamp, and this particular one has two speakers that shoot sine waves. I was watching it and I thought to my self: "it would be great if I reproduce that in real life".

After thinking about it my thoughts went to back to that arc speaker I asked about last week. How could one create shapes such as a sine wave with an electrical arc? I'm thinking an array of magnets on opposing sides of the arc. Then you just need to control the polarity and presto. It's a proven method for controlling electron beams, but I'm wondering how effective it would be on such a large arc and at creating such complex shapes.

Although I wonder if it's possible with the correct signal to shape the arc on it's own. It can already be made to 'resonate' and create sound. Or is that particular approach stretching it too much? I can see it being near impossible to get the wave amplitude to be large enough for the naked eye to notice.

 

[Wetmelon]

Even if the rest is possible, I would be surprised if you could get a strong enough magnet to effectively bend the wave with a frequency high enough to make it look like a wave rather than a slightly bent line...

 

[Lancelot59]

Really? How strong are we talking here?

Would running the arc inside a container filled with a gas that ionizes easier than nitrogen/oxygen such as argon be helpful? I made a strobe light in high school that takes only 10000V to ionize the gas. The arc travels right though the centre of the tube around the curve, but that's just following the path of least resistance.

Would an environment like that make it easier to bend the arc? Also what kind of control could I expect to have over the arc? I'm guessing getting the fine control needed to create things like sine waves with short periods or an audio waveform would be very tricky.

 

[denni89627]

I don't think your speaker would sound too hot in an argon tube.

 

[Lancelot59]

This isn't for a speaker, I know the speaker arc needs to be open. I'm talking about just making fun shapes with an arc and some electro-magnets.

 

[Averagesupernova]

A magnet will deflect an arc. The longer the arc, the easier it should be to deflect it midway between the electrodes. If you are serious about this, start with the a tesla coil which is the easiest way I know of to make long electric arcs.

 

[Lancelot59]

Well what I need to do is form complex shapes line sine waves. Would an array of opposing electromagnets do it? The reason I think a tesla coil would be bad is because of it's proximity to the magnets that would bend the arc.

 

[Hologram0110]

You should be able to a achieve something that looks like a sinewave with an array of alternating magnets. If you want the sinewave to move, you could do be able to do it with a 1D array with all the current in the electromagnets following a triangular wave pattern.

Where I think you will run into trouble is the aspect ratio of your device. If you have an array of magnets with alternating polarization the farther you are from them the more uniform the field appears. This will make it very difficult (but not impossible) to produce sine wave of significant amplitude.

The best way I can think to make this work would be to construct an array of strong electromagnets controlled by a pc or other micro controller but you will need to position them as close as possible to the arc. Might also help if you make the electro-magnets oval instead of circular to produce a distorted field to make it look more like a sine wave.

 

[Lancelot59]

Yeah, that's one of the things I was wondering, is how to narrow the field so I can have more magnets closer together to get more control over the arc.

Could I get finer control by having two arrays of opposing magnets, and having the magnets directly across from each other be of opposing polarity? Or would that just distort the field more?

Third, how does the oval shape stretch the field out more?

 

[Hologram0110]

Having an array of magnets on both sides the way you describe would help, as long as you get them aligned properly (though I'm assuming you wouldn't want this because it would obstruct the view, which is why your making this?). This would basically double your magnetic field assuming you make everything symmetric.

In the static case:
The width of the magnets would determine the length of your sine wave (2 magnet widths = 1 wave length).

The amplitude of the sine wave is more complicated. The electrons moving in the arc will feel the force due to the magnetic field and their velocity according to F= qv X b. Therefore the distance they travel in the y direction (with the sin wave propagating along x) will depend on how long they are in the magnetic field (The relationship is actually more complicated since the magnetic field is non-uniform). The amplitude is also limited by the physical size of the magnetic field, which is controlled by the size of the coil.

Now that I think about it in more detail, the oval coil suggestion is very complicated and I would have to do a lot of math to figure out if is a good idea or not. Stick to circular unless you want to run the numbers.

You could create a narrow tank out of a highly insulating material like glass that would contain the gas in a fixed (narrow) region so that the arc must remain close to the coils. You would need to make sure that the glass wouldn't provide a path of least resistance however.

Also you need to make sure that your magnetic fields are not strong enough to cause circular motion of the electrons.

 

[Lancelot59]

If the magnets are directly above/below the 'tank' then they'd be out of the way. Symmetry would be key to making this work cleanly.

I was thinking of having a much denser set of magnets, so that I can create things like audio waveforms. The object of the ionized gas is to make the arc straighter, but also longer. So for the same power I get a much longer arc. That should make it easier for me to manipulate.

What sort of calculations would I need to run to create an effective oval coil? I've only got first year going on here, and I haven't gotten to physics 2 yet. But they're just formulas, so I'm sure I could figure out how to run the numbers.

For the material of the tank, glass would be a nice solution. Just need to make sure I get some properly tempered glass so if it shatters it doesn't gouge everything. Failing glass I could always use acrylic or another plastic.

Isolating the tank from ground wouldn't be too hard I think, and I've never heard of plastic or glass conducting. It makes sense that the arc would want to follow the path of least resistance through the gas to the anode.

When you say circular motion, do you mean the electrons themselves could start spinning? Or the arc will start to follow the magnetic field down to the north magnetic poles of the control magnets?

 

[Hologram0110]

I don't think that you can place the magnets above and below, review your magnetic theory, but the magnetic force is perpendicular to the velocity and the magnetic field lines. IE the magnetic poles must be perpendicular to the plane containing your sine wave.

If you want them above and below you would need to use electric fields. Again I should stress that unless you get very close to the magnets the effect won't work. You also typically need a significant current and/or number of loops to produce a strong magnetic field.

With regards to them just being formulas... This is true, but that doesn't mean they are nice to work with. In this case your dealing with vector calculus that would normally be solved numerically. Calculating the magnetic field intensity vector produced from a coil at a position off axis is a non-trivial task.

For one of my undergraduate project I tried to make a maglev train (without much success might I add). I tried using this program that may or may not help you with what your doing...
http://www.vizimag.com/ There are lots of other programs designed to deal with magnetic fields too.

 

[Lancelot59]

Right, I remember now...I this could pose an issue.

I wonder if there is a way to make the magnets stronger...perhaps use a superconductor? Or dunk the coils in a really cold solution?

I haven't done vector calc yet, so I guess solving those would be a bit tricky.

I'll take a look at that program. So would this project fall under the 'cool but totally impractical' category?

 

[Hologram0110]

Superconductors are very frequently used to produce strong electromagnets. There are some downsides though, superconductors currently require some sort of cryogenics of at least liquid nitrogen temperatures (or lower depending on the type of super conductor). They are also pretty expensive. Also when using super conductors to make strong magnets you typically also use a high current since you don't have joule heating. The magnet is a large inductor and therefore resist changing current quickly which can sometimes damage electronics if you aren't careful.

Liquid cooling of normal conductors can also work however you run the risk of shorting out your electrical system. Depending on currents this may not even be sufficient to keep the wires at the center of the coil from melting.

I would describe this as very possible, but certainly not easy. I wished I had thought of it for my 4th year undergraduate engineering physics project it would have been perfect since we could even use the physics department equipment... It would certainly be impressive for a first year to pull it off, especially if you made a control system for it.

 

[Lancelot59]

Well I do have this habit of making things tricky for my level work to some degree. Mostly because I obsessively beat it over the head till it decides to work.

A modelling program would certainly help with the complex magnetics involved with this. So I'll take a look at what you posted when I get home. The superconductors needed for this on my budget is just wishful thinking, as is the system needed to contain the liquid nitrogen. :/

I have an idea for a system to control the electromagnet's strength, but not the polarity yet. I'm thinking an IGBT.