How do arc furnaces sustain themselves without significant resistance?

[Upsilon]

I'm highly interested in building my own small arc furnace for hobby use. Thus, I want to understand the nature of electrical arcs before proceeding to have maximum control over my design. So far, I know that electrical arcs form when the dielectric strength of the material (in this case, a gas) is overcome. Once this barrier is overcome, the gas becomes locally ionized, allowing electricity to flow with relative ease.

What I don't understand is how the arc sustains itself. From what I understand, this plasma formed has very low electrical resistance. Without significant resistance to generate heat, how is the plasma sustained? I also want to know how one would control the intensity of the arc and thus the temperature of the furnace, but I feel that the answer to the previous question will lead into this.

 

[Baluncore]

What material will you be melting in the furnace ? You may need a shielding gas to prevent oxidation.

What electrode will you be using in the furnace ? Carbon might be used.

Will you be using AC or DC in the furnace ? Where will your power supply come from ?
How much power do you need ? How many kg or tonnes of melt ?

Are you aware that the arc formed will radiate sufficient UV to burn your skin and damage your eyes ?

An arc welder employs series inductance to keep the arc running. It regulates the current with either a series resistor and or an adjustable choke transformer. The choke limits the current by saturation of the partially withdrawn core. You could use an arc welder transformer as your current limited current source.

 

[anorlunda]

Baluncore's advice is practical. But to satisfy curiosity about arcs and their control, the article linked below is very good. Arc lamps were commonly used 1802-1900.

https://en.m.wikipedia.org/wiki/Arc_lamp

 

[Upsilon]

What material will you be melting in the furnace ? You may need a shielding gas to prevent oxidation.

I'll be using it for several different purposes. If I work with an oxidation-sensitive material, then I will operate the furnace in an inert environment, such as argon gas. This will affect the behavior of the arc though, no? Since argon has about a fifth of the dielectric strength of air.

What electrode will you be using in the furnace ? Carbon might be used.

For operation in the presence of oxygen, I will use graphite rods. However I may use tungsten wire when operating in an inert environment just because of its greater workability.

Will you be using AC or DC in the furnace ? Where will your power supply come from ?
How much power do you need ? How many kg or tonnes of melt ?

My furnace will be quite small; the chamber is going to be a 4" tall silicon carbide crucible. I'm hopeful to be able to achieve 2500°C at maximum (by this I mean the temperature of the target material, NOT the temperature of the arc. This will largely depend on insulation), but I could settle for less. I won't be dealing with very much material at a time, so I believe I should be able to use the standard 50 amp power recepticle in my garage. I would need to build a transformer, but that is pretty straightforward. In addition I shouldn't need to run it for more than a few minutes at a time.

Are you aware that the arc formed will radiate sufficient UV to burn your skin and damage your eyes ?

Yes. The crucible that will be used will also be encased in a brick housing with a cover, which should contain the bulk of the radiation. I will also use eye protection (like a welding mask) while using it.

An arc welder employs series inductance to keep the arc running. It regulates the current with either a series resistor and or an adjustable choke transformer. The choke limits the current by saturation of the partially withdrawn core. You could use an arc welder transformer as your current limited current source.

I'm not sure I'm following entirely. Limiting the current of the circuit allows the arc to sustain itself? This is what I didn't understand in the first place - since the ionized gas is very electrically conductive, where does the heat energy required to sustain the plasma come from?

 

[anorlunda]

You must also be careful to not introduce harmonic noise into the power grid, or else the electric company will cut you off. Arc furnaces are notorious electric noise sources. For that reason, Baluncore's suggestion of using a pre-engineered electric welding machine is very good advice.You need a higher voltage to create an arc than to sustain it. I don't think you can apply E=IR to a plasma arc. Resistance is a linear concept, but arcs are highly nonlinear.

 

[Upsilon]

You must also be careful to not introduce harmonic noise into the power grid, or else the electric company will cut you off. Arc furnaces are notorious electric noise sources. For that reason, Baluncore's suggestion of using a pre-engineered electric welding machine is very good advice.

What exactly is harmonic noise, and how do arc welders/furnaces cause it? Would the power company really care that much about it running for only a few minutes on rare occasions? I'm not talking about going into the metal ingot casting business or anything.

I apologize in advance for anything else I don't understand; I have limited electronics knowledge but am willing to learn.

 

[Baluncore]

I question the configuration of your arc furnace. It will probably depend on the resistivity of the material being heated.
Is the material being heated by a current flowing through the material with a short arc at the electrode contact ?
Or, is the radiation from an arc being used to indirectly melt a resistive charge, without the charge being part of the circuit ?

An arc is usually started by electrode contact or a very short gas ionisation breakdown. Once an arc is struck, the ionisation continues, an inductor dynamically adjusts the arc voltage to keep the current flowing; as V = L * di/dt. The current is usually limited by the magnetic circuit of the transformer.

You might consider a cleaner way of heating an electrically conductive material by induced circulating electric currents, as in the induction furnace.
https://en.wikipedia.org/wiki/Induction_furnace

 

[anorlunda]

What exactly is harmonic noise, and how do arc welders/furnaces cause it? Would the power company really care that much about it running for only a few minutes on rare occasions? I'm not talking about going into the metal ingot casting business or anything.

I apologize in advance for anything else I don't understand; I have limited electronics knowledge but am willing to learn.

Harmonics are caused by nonsinusoidal wave forms. Lots of harmonics coulee ruin radio, TV, and computers in your whole neighborhood. The utility would be flooded by complaints if you are lucky. If you are unlucky, the transformers near your house could explode and you would be liable.

Commercial arc furnaces melt scrap steel for recycling. When something big and solid like an engine block touches the electrodes, it is like a big short circuit that comes and goes as the object moves and vibrates. If not properly engineered, nearby utility substations explode. Arc furnaces are the most undesirable load I can imagine for a power grid. The best way to design them is to build them in someone else's country.

Welding machines are analogous to arc furnaces but they are designed to protect the grid from the bad effects.

 

[Upsilon]

Or, is the radiation from an arc being used to indirectly melt a resistive charge, without the charge being part of the circuit ?

This. I won't necessarily always be trying to melt electrically conductive material.

You might consider a cleaner way of heating an electrically conductive material by induced circulating electric currents, as in the induction furnace.
https://en.wikipedia.org/wiki/Induction_furnace

This is actually something else I want to do too, but as I said above I won't always be trying to melt a material that this can work with. I suppose I could heat a conductive medium with induction and conduct the heat into what I want to melt, but then I'm limited by the melting point of the medium. A medium like tungsten or other high-melting metal could be used but significant pieces of these metals are ridiculously expensive.

 

[Upsilon]

Harmonics are caused by nonsinusoidal wave forms. Lots of harmonics coulee ruin radio, TV, and computers in your whole neighborhood. The utility would be flooded by complaints if you are lucky. If you are unlucky, the transformers near your house could explode and you would be liable.

Commercial arc furnaces melt scrap steel for recycling. When something big and solid like an engine block touches the electrodes, it is like a big short circuit that comes and goes as the object moves and vibrates. If not properly engineered, nearby utility substations explode. Arc furnaces are the most undesirable load I can imagine for a power grid. The best way to design them is to build them in someone else's country.

Welding machines are analogous to arc furnaces but they are designed to protect the grid from the bad effects.

Is there any practical way to prevent this from happening? As you have mentioned commercial arc welders already do this.

Another thought I had was to use a car battery to power the arc furnace, keeping it completely off-grid. A typical car battery can theoretically supply 200A for around 15 minutes. Now, obviously continuously deep-cycling the battery will destroy it in as little as 30 cycles. Since I plan on only using it for a few minutes each time, this would be a 20-30% cycle, which would allow it to last considerably longer than deep-cycling it. Not to mention that I will seldom be using it. With all of this in mind I think the car battery could last a year easily.

Of course, I would rather have a direct connection to the power grid, since car batteries aren't exactly cheap. I have seen several videos of homemade arc furnaces, yet they have no mention of generating harmonic noise. Take this video, for example:

 

[anorlunda]

Is there any practical way to prevent this from happening? As you have mentioned commercial arc welders already do this.

Baluncore briefly outlined how welders accomplish it in post #2. You didn't seem to understand what he said. Therefore, your best option would be to simply begin with a commercial arc welder to make your oven. Some can be powered by DC. I see several AC and DC models for sale on Ebay starting as low as $61.

Yes it is a good idea to use a deep discharge battery, to keep it off the grid. However, as said several times here already, commercial arc welders protect the grid so that you do not have to use DC.

Your basic starting point should be to calculate how much power you need to melt the material. Then multiply that by 5 or so to account for the electrical and thermal inefficiencies.

 

[Upsilon]

Baluncore briefly outlined how welders accomplish it in post #2. You didn't seem to understand what he said.

Reading over it again, is this just referring to an AC step-down transformer? If so then I was already planning on using one anyway. I guess I didn't make it clear that I wasn't simply going to use the power straight from the 50A recepticle to spark an arc. I want to use a transformer to convert it to a 200A or so supply of lower voltage.

 

[Baluncore]

Welding machines based on inverter technology are now available. They use electronics to regulate the current. They are used for stick and TIG welding and sometimes called caddy welders because they are light weight movable units. They usually produce DC which may consume your electrodes faster than AC. They cost $200+ new, but are less reliable than the old adjustable transformer welders.

A transformer based arc welding machine is designed to produce about 25V at 100A. The current is not limited by the resistance of the circuit but by the adjustable geometry of the magnetic flux in the core of the transformer. Transformers produce AC but it can be rectified to DC by a diode bridge if you need that.

Bigger transformer welders run from three phase AC, or two phases of the three phase AC. Smaller units run on a single phase and neutral. Make sure you have access to the power you need for any particular machine.

There is a big screw through the transformer that positions the movable part of the core to set the current limit. Look for a welding transformer with a handle that moves the magnetic core, with a pointer that shows the core position and expected current. One example if you google images or search the auction sites is an 'EMF Transarc'. Expect to pay below $100 for a good second hand unit. The scrap value of the copper is greater. Try to get one with some heavy cables. They are cheap because they are heavy.

Three-phase arc furnaces have three electrodes above the charge. 3PH is a good way to make an arc furnace since the arc transfers progressively around the electrodes as the AC cycles. You can use three identical choke welders to reduce the 3PH voltage and to limit the power for a high current arc furnace.