Any good suggestions for our QCW "sword sparks" mechanism research?

[SimonLiu1999]

Hi There,
I'm an undergraduate student of a University in China. This is my first time to ask questions here. I need some help for my research.

Recently, we decided to do some research on the formation of the sword sparks. We noticed that QCWDRSSTC has been quite common and many people have produced the sword sparks. I am  interested in this phenomenon but I can't find any academic literature to explain how this special sword shape forms. Now, we want to do research on our own to give a reasonable explanation. We have pretty good experiment equipment such as very high speed camera to capture the growth of arc. However, the former research is too few and I am not sophisticated in this field. So I sincerely hope you can offer me some advice on this research. You can give me your assumption or offer me some existed explanation...

I think this research is meaningful to guide us to get the best electrical arc. I will publish my experiment results if If it goes well.

Unfortunately, Coronavirus is still severe in China and we are required to study online at home. I will start this research after I can go back to the lab.

[Mads Barnkob]

You will find the most information from Steve Ward and Gao at loneoceans.com

Generally it is the ramped DC bus voltage that makes the long spark growth possible, over a very long time, maybe up to 10ms. The sword charateristic shows above 400 kHz resonant frequency.

[SimonLiu1999]

You will find the most information from Steve Ward and Gao at loneoceans.com

Generally it is the ramped DC bus voltage that makes the long spark growth possible, over a very long time, maybe up to 10ms. The sword charateristic shows above 400 kHz resonant frequency.

Thanks a lot，I know these features. now we are trying to give a explanation by using physics. Why does the ramped DC bus voltage generate the sword shape.

[Mads Barnkob]

You will find the most information from Steve Ward and Gao at loneoceans.com

Generally it is the ramped DC bus voltage that makes the long spark growth possible, over a very long time, maybe up to 10ms. The sword charateristic shows above 400 kHz resonant frequency.

Thanks a lot，I know these features. now we are trying to give a explanation by using physics. Why does the ramped DC bus voltage generate the sword shape.

Read my reply again, that was not what I said.

Only someone with deep knowledge of plasma physics could really explain it. As you found out, there is no research with cross field teams to experimented with the QCW / VTTC spark formations. But the phenomenen has been known since the age of tubes.

[John123]

I wonder if its magnetic in nature.

[HiVi]

I always found this sword sparks really interesting.
I think that it will be much easier to explain once the phenomenon is captured in real high speed video. There are some youtube videos in slow motion but still not slow enough to see clearly the propagation of plasma channel.
But some food for thought: (look at image stolen from youtube in attachment)

• It propagates quite slow - approximately half the speed of sound in main direction.
• First small arc that ionizes the air is of classic shape, which is almost always visible at bottom.
• Few moments later it gets the shape of sword arc with almost straight direction.
• Plasma channel is always of serpentine shape, with tiny arc branching at 90 degree from main direction, whenever the channel localy changes direction.

Suggestions for research:

• use stereo vision so you can reconstruct plasma channel in 3D
• always measure atmospheric conditions, or even better, if possible experiment with different conditions
• measure EMF

With this you should get good indices to reconstruct what is happening.

[Uspring]

I have 2 ideas regarding sword arcs, which are notable due to their straightness and lack of branches and seem to occur only at high frequencies and slow rampup:

1. The "sideways breakout hypothesis".

Arcs are rather thin and cause strong fields pointing at a 90 degrees angle to their main path. This can result in branch offs. This is a voltage dependent effect as can be seen e.g. in ground arcs, where branches die off immediately after the arc hits the ground. The voltages decrease considerably in such a situation as was measured by Hydron. https://highvoltageforum.net/index.php?topic=117.msg735#msg735
Slowly ramped up arc voltages, such as in a QCW are in a sense low voltage arcs. The arc conductivity relies on charge carriers being generated, which arise from a high temperature and/or high voltages. Most of the arc, except its tip, is hot and charge carriers are generated by air molecules hitting each other at high speeds. Near the tip the arc is cold and charge carriers are generated by the electric field, which accelerates spurious electrons so that they will free other electrons once they hit air molecules. This will cause an avalanche of electrons and establishes conductivity. In contrast to the latter effect, the heating up of the arc requires a lot of energy and is thus a slow process. In most (non QCW) TCs the top load voltage is ramped up quite quickly, which doesn't allow the arc temperature to settle at the value it would have as if it would be stationary, e.g. in an equilibrium state. The arc is more resistive since it is colder and it will have less voltage at its tip, but it can still grow if there is enough voltage in the top load. Thus long arcs are possible even with short bursts, but they will require more voltage.
QCW coils ramp up slowly, so they need less voltage for a given arc length. This will reduce the likelyhood of a sideways breakout.

2. The "random walk hypothesis".

The direction the arc will choose at its tip, where it will grow is of statistical nature. As pointed out, it depends on spurious free electrons near the tip, which will be multiplied rapidly due to the electric field there. A rapidly growing and "overvolted" arc, as described above, will have a stronger electric field at its tip. There will be a direction, where the field is strongest, i.e. forward but around this direction there might also be a region, where the field is large enough to cause avalanches. The arc will grow into the direction, where spurious electrons happen to be. In slowly growing arcs the voltage at the tip is lower and the only remaining direction, where the field is large enough to cause avalances, is forward.

Sword arcs also seem only possible at frequencies of 400kHz or above. Large frequencies move the arcs space charges back and forth more often than lower frequencies. They will cause a higher current and consequently a higher arc temperature. These arcs will be more conductive and therefore longer. Or, vice versa, will require less voltage for a given arc length. So high frequency arcs also fall into the category of low voltage arcs.

These are just ideas from what I know about arcs and AFAIK sword arcs or even standard TC arcs haven't been investigated intensively. It certainly would be interesting to see in a high speed video e.g., whether arcs branch at the tip or sprout out sideways later. And the region of ramp up speed and frequency, where sword arcs appear, hasn't, to my knowledge, been explored yet. I'm certainly interested in the results of your experiments.

[k42]

You will find the most information from Steve Ward and Gao at loneoceans.com

Generally it is the ramped DC bus voltage that makes the long spark growth possible, over a very long time, maybe up to 10ms. The sword charateristic shows above 400 kHz resonant frequency.

Has anyone estimated the conditions inside the spark - such as temperature, pressure, density...?

If we can estimate those, we can try to estimate some characteristics of the plasma.

My hypothesis is: does the "sword-style cutoff frequency" happen to match any of the characteristic frequencies of the spark plasma?

[Uspring]

Tesla coil arcs are relatively cold arcs, as e.g. compared to arc lamps. If they are hot and conductive, they tend to stretch out until they almost extinguish themselves. A rough estimate of the temperature can be obtained by conductivity measurements such as here https://highvoltageforum.net/index.php?topic=117.msg827#msg827 and looking up the corresponding temperatures in a table. IRC that resulted in about 5000 K. I don't have access to the table at the moment.

The pressure is atmospheric pressure. These QCW arcs grow slowly and don't compress the air. They don't bang as loud as the non QCW arcs.

Plasma frequency is related to the free electron density. For e.g. 400kHz that would be about 10^9/m^3 or about one per cubic millimeter. TC arcs generate a _lot_ more free electrons.