Plasma Question

How does a plasma cutter in an average metal shop generate the plasma? What materials are used in the nozzle (and a generation chamber?). How often do parts need to be replaced in a plasma cutter?

Regarding the operation of a plasma cutter, I'll take a stab at it! Basically the cutting head contains the electrode that a current is passed through to the work piece (metal). This arc creates an area of plasma that is controlled by shield gases and in other cases magnetic lorentz forces that constrain the plasma in a way that is suitable for cutting. Please understand this is a very rudamental description and mostly from memory, so take it for what it's worth. I can say with certainty though that in a hi-def plasma cutter, electrodes do need to be replaced on the order of 200-300 pierces. This may have improved over the years since I worked with these systems. Hope this helps! Mike

Hi,

If you pass an electric current through a gas (or vaporize a liquid or solid) then you will need a high voltage and a high current: the energy will ionise some of the atoms/molecules and a plasma is generated.

The current is bringing along its magnetic field that in turn is compressing the plasma.

So inside your (by solid walls confined) vessel there is only a small part of the volume existing as a plasma.

If you look at the curves: electrical breakdown-voltage as a function of gap and pressure, known as Paschen-law, you will see that easy ionisation is possible at low pressures of some µbar to mbar.

The higher the pressure the better the isolation-qualities of the gas, so may be a very high voltage is necessary to start the arc that is transforming the molecular or atomic gas into a plasma.

Electrodes are subject to erosion, so electrodes that are used in plasma-cutters or micro-plasma-welders are made from copper and need backside water-cooling.

Electrode-less plasma systems use high frequency excitation - often in one of the many resonances that are possible in plasmas. Lowest frequency that is often used is 13,6MHz because this is an industrial frequency and relatively easy to meet the emission requirements.

Many other frequencies are used (up to 12GHz) depending on cost and resonance.

Magnetically induced electron-cyclotron resonance is often used with the cheap 2.4 GHz microwave heaters thus requiring strong magnets.

Lower than 13MHz is possible but not very efficient as there are too long off times - the plasma is igniting at a high voltage and is extinguishing at a lower voltage. Switch-on and switch-off is within a fraction of a microsecond.

Coupling of RF-plasmas is either by plates (capacitive) or coils (inductive coupling).

Impedance matching is mandatory else a lot of power is reflected back to the generator and heating the generator and not the plasma.

So impedance matching is most often done automatically by motor-driven vacuum capacitors. (One fixed inductance lengthwise and one capacitor connected to ground at the input of this inductance, the other capacitor connected to ground at the output side of this inductance.) If the current is high there is the need of internal cooling of the inductance.

(Not possible in microwave systems so there a circulator and power absorber).

Text book: not possible to answer as there are too many totally different applications.

If you specify your needs may be there is an answer. To have a start I would recommend "Chen: Glow discharge Processes".

Photo is from our DC system (10µbar Ar 330V 1A 50cm² area of target) coating with Al and AlN (450V in N₂)

RHABE

"Lower than 13MHz is possible but not very efficient as there are too long off times - the plasma is igniting at a high voltage and is extinguishing at a lower voltage. Switch-on and switch-off is within a fraction of a microsecond.

Coupling of RF-plasmas is either by plates (capacitive) or coils (inductive coupling)."

Now this is interesting. The purpose of this inquiry is to see if additional heating of a He working fluid can be obtained through the application of a plasma generator in a flow stream. As the flow stream reciprocates within the walls of the control volume, it would be beneficial to have the plasma "on" in one direction, and "off" in the other. Kind of like the spark plug in our IC engines only arcing when we want it to.

However you were saying that below 13 MHz will cause much noise, and thus might be an issue due to RF interfering with other devices nearby, right?

How effective is a plasma at transferring its heat to the bulk of the gas, if less than 1% of the gas is in this state? Can one ratio the constituents to determine bulk gas temperature? For example if there were 1% of a 20K degree K plasma temperature and the balance was 1000 K gas, would the bulk temperature be 1190K? Or would it be higher and related to the type of plasma that is generated. The one in the photo looks to cover a large area, and likely larger than an electrical arc of the same energy. Is this true?

The idea of changing LTE has me a bit confused. I just picked up a couple of books from the local University library to educate myself more, but they are mostly concerned with 1 Bar plasmas.

Thanks again for any additional details you can provide. (Man I hope these questions make sense.)

Hi Tucson Don,

on-off control can be easily achieved by the power supply - the good ones allow switching or pulsed operation.

So there it is not necessary to use RF.

Heat transfer is really difficult and cannot be calculated in a simple manner.

There are electrons and ions and photons and hot gas.

There are electrodes or the inductive coil at some potential - either strapped by grounding or the power supply or by self biasing by catching charged particles.

Electrons do a lot of the heating in low pressure systems.

Electrons can be influenced in their pathways by magnetic fields - ions are too heavy unless big containers are used.

Two simple ways:

A. ICP (inductively coupled plasma flame spectral analysis) is used in analytical chemistry since decades so certainly there are used instruments to be misused for a trial.

B. Let your gas pass through a quartz-glass tube and pass through the same tube a microwave (from your kitchen). Add circular magnets to get a strong magnetic field slightly below 1T. This will give you the condition of electron-cyclotron-resonance. At the output -where the non-absorbed microwaves exit - a resistive absorber has to absorb the not used energy. Water, carbon, carbon-epoxi are often used or loads from industrial microwave systems. May need additional cooling.

C. Not so simple but may be better: helicon plasma wave excitation.

Why do you want to stretch out your plasma? Easy in low pressure, but requiring more uniform heating in high pressure plasmas.

RHABE

Hi Rhabe,

I'm not sure I understand the question about stretching out the plasma. My desire is to see if a plasma can be created within a high pressure noble gas such as He at say 30-50 bar, and if the presence of the (small quantity of) plasma can be used to augment heat transfer within the gas. I don't want the plasma to directly impinge on any surfaces, or I'd likely overheat them.

Here's what I don't understand: the temperature of the plasma can be on the order of 10,000°K, right? And somehow, locally, thermodynamic equilibrium rules can break down in and around a plasma. Does this mean that the energy that must be put into making the plasma can create a greater net temperature gain within the bulk of the gas within which the plasma is generated, than that same amount of energy added say conductively or convectively? This seems to violate the second law, but I'm only a mechanical engineer with a bachelor's degree and not a Phd in physics, so I don't think I'm allowed to violate the laws of thermodynamics.

Thanks for your helping me understand this better. And what the heck is a Dalton?

Don

Hello,

I'll answer the easy one first. A Dalton is a unit of molecular mass, defined as 1/12th of the mass of a Carbon-12 atom. It is also called the atomic mass unit.

Classical gas thermodynamics is based on the assumption that the gas molecules behave as particles having definite position and momentum, and interacting solely by collision. The last assumption breaks down when the particles interact with electric and magnetic field gradients at both the global and local (atomic sized) scales, as they do in a plasma. (The position and momentum assumptions are based on the average molecular energy related to temperature being greater than the difference between atomic energy states and the average distance between molecules being greater than the Debye length so that Heisenberg's Uncertainty Principle has negligible effect.)

Also, since a plasma has both free electrons and ionized gas molecules to transport thermal energy, I would expect its thermal conductivity to increase compared to an ordinary gas. If a diatomic or triatomic gas is broken down into monoatomic constituents in a plasma, the specific heat of the gas will be altered since the molecular vibration contribution to specific heat is removed or decreased for the broken molecules.

The laws of entropy and energy conservation still apply, it's just that the numbers get quite a bit messier with a moderate-density plasma. For one thing, plasma does not have a static equilibrium condition while charged particle recombination is possible. At least you aren't dealing with stellar conditions, where the extreme pressure compresses the ions to within the Debye length. At this point, you must use quantum mechanics to describe the situation since the classical assumptions are no longer valid.

The short answer based on the above is that I would expect a plasma to impart thermal energy to a surrounding neutral gas with somewhat greater efficiency than simple conduction or convection through a neutral gas. I would still expect to see the usual relationship between temperature, pressure and volume. Extreme local temperatures are possible, although not inevitable. Lightning discharges are an obvious example of Joule heating of plasma. Because of the high current density involved, this heating is so rapid it generates an over-pressure shock wave that we hear as thunder.

To the best of my knowledge, controlled fusion research systems in general use magnetic confinement to keep hot plasma from hitting the walls of the experimental apparatus. This approach avoids having anything in contact with the plasma that would dissipate its thermal energy.

That's the best I can do without finding my dusty old plasma physics textbook.

Hi,

I was successful to send my yesterdays answer to Nirwana and not to this forum, so I will try once more.

"if a plasma can be created within a high pressure noble gas such as He at say 30-50 bar"

Yes certainly but may be you will need extraordinary high voltages depending on the size of the plasma.

I did work only with low pressure plasmas but I did see some high pressure plasmas and all had very small dimensions (0.1 to 1 mm typical)

"plasma can be used to augment heat transfer within the gas"

Plasma is a very efficient heater, better than anything else (?).

"the temperature of the plasma can be on the order of 10,000°K, right"

Better think about energy of an individual electron or atom or molecule and the probability distribution of energies and energy exchange by collisions, fields, excitation and emission of photons.

No violation of energy conservation or any other fundamental law, the gas has a mean temperature and the plasma has a mean temperature (better described by the energy distribution) and the electrons too.

I am not sure if it is possible to avoid any contact with walls.

You should post some more details: do you want to heat the gas or do you need it ionised or excited? Which cross-section which velocity.

Electron cyclotron resonance may be the cheapest way as only some magnets a microwave-oven an absorber and a quartz tube is needed.

RHABE

Hi RHABE,

Thanks for that catch, as all responders have been very helpful in guiding and directing what to read first in the texts I got. My goal was to see if we could scavenge some electrical output and put it back in as a plasma to superheat the gas without a second law violation. The design parameters, if this is a benefit, can be adjusted, but Re numbers on the order of low 10e5 are expected.

Thanks,

Don

Hi Tucson Don,

did you think about resistive heating?

You have the best gas to do this, for any temperature below 3000°C I would think first about a hollow tube of graphite inside a Dewar. Pass the gas inside the tube, (tightness?), pass a current to heat the tube.

So nearly no heat loss. Temperature of the tube and power to be regulated.

Variable power supply (DC or any frequency) necessary as resistance is a function of temperature.

RHABE