# Why does electric arc in a switch prefer a curved path over a straight path?

Recently I found this video of a 500 kilovolts line being opened under load.

When the switch contacts are pulled apart an electric arc predictably starts. While the contacts are close to each other the arc runs along a straight path between the contacts. Then as contacts are pulled further apart the arc starts to bend and turn into a steep curve and its length becomes several times greater than the distance between the contacts. Then finally the arc just fades out.

That doesn't make sense to me. As I see it the arc should take the least resistance path and that's clearly a straight path, not a steep curve. Even more, if the arc takes a curved path why would it fade out suddenly instead of just taking a less curved path of less resistance and continue running?

Why does the arc behave this way - first prefers a curved path and then suddenly fades out?

• There are plenty of answers here, but physics might be able to give a better answer. – Kellenjb Feb 3 '12 at 12:19

This was a comment but the links were too long.

As well as what others have said - look up "magnetic blowout" and be suitably amazed. More for DC but certainly not only. A magnet is used to deflect the arc so it lengthens and fails

Equipped in even very small and common switching devices. Many of these and these

Even Tesla did it :-)

Interest only - from here

EXPERIMENTS WITH ALTERNATE CURRENTS OF HIGH POTENTIAL AND HIGH FREQUENCY.

BY NIKOLA TESLA.

A LECTURE DELIVERED BEFORE THE INSTITUTION OF ELECTRICAL ENGINEERS, LONDON.
With a Portrait and Biographical Sketch of the Author.
NEW YORK: 1892

• Wow. So it doesn't depend on convection, does it? What happens if there's a failure in the circuit powering the magnets? – sharptooth Feb 3 '12 at 11:59
• @sharptooth - As above - this is AS WELL AS convection. More in DC use but AC also. Often permanent magnets in small switches. In very high voltage systems the arc will always stop. Eventually :-). – Russell McMahon Feb 3 '12 at 12:14
• But in the video there isn't such a system, right? – clabacchio Feb 10 '12 at 10:43

There are two combined phenomena:

1. The current chooses always the least resistive path, that is not necessarily the shortest, as can easily be proven by physical circuits;

2. Such high voltages and currents have an ionization effect on the surrounding air (electrons are stripped off of the atoms) that cause it to become more conductive in the area where this current is flowing, but at the same time also hotter; this hot air is lighter than the surrounding colder air so it starts going upwards, but still leaving this "conductive" path in which the current continues to flow.

This process ends when the path of more conductive air becomes too resistive for enough current flowing in it, and the ionized air goes up, replaced by "normal" and less conductive air, which is not conductive enough to create the arc. Probably the arc was caused by an event, such an overvoltage, or simply, as in one of the videos, an object that lowered the resistance between the two contacts; OR, like the video in the example, a switch that is opening. When the arc fades, it's also because this triggering event has ceased.

• Why does the arc suddenly fade out then? – sharptooth Feb 3 '12 at 7:29
• I believe it stopped because an upstream station disconnected the power. I think if it had dissipated due to increased resistance, it would have restarted at a lower point (like a jacob's ladder). – W5VO Feb 4 '12 at 3:36
• @W5VO but if you look at the video posted by Nick, the ladder stops until a new event cause it to start again. – clabacchio Feb 4 '12 at 8:59
• it looked to me that by the time the long-path discharge extinguishes, the gap between the open contacts has been widened to the point that the 500kV can no longer just punch through the ordinary air in the gap. – JustJeff Feb 4 '12 at 20:42
• @clabacchio - i must stick by my point. The path of the discharge may be longer, looping above the gap as it does, but that path is composed of hot, ionized air; even though it's longer in distance, it's lower in resistance than the cold air in the direct path in the gap. If you've played with high voltage, e.g. even a neon sign transformer, you should know that you can always draw a spark out longer than the gap needed to strike the arc, and once the spark is lost, you have to bring the conductors in closer to get it started again. – JustJeff Feb 4 '12 at 23:20

The air ionises initially and an arc forms. Being air, and being hot, it rises.

The ionised "tunnel" air rises and "breaks" at which point the arc is extinguished.

• Does the switch in fact rely on convection for breaking the arc? – sharptooth Feb 3 '12 at 7:35
• @sharptooth In this case it seems that :) and the fact that theoretically, when the arc fades, the contacts are too far each other for creating an arc – clabacchio Feb 3 '12 at 8:15
• @sharptooth - Not in general, no. The posted video was a system failure. Generally, switches are designed so that they open very rapidly, and to a wide enough distance to reliably and safely quench ("put out") the arc. There are actually some high-power switches that intentionally create an air-jet between the contacts to more rapidly stop the arc when the switch opens. – Connor Wolf Feb 3 '12 at 12:00

The arc ionizes the air. The air has finite resistance, so it heats as current flows through it. As it heats, it becomes more buoyant and raises. Current simply follows the path of least resistance.

Jacob's Ladder is a visual effects device, which works on this principle. Some lab scenes in Frankenstein movie feature it. There are a few videos of Jacob's Ladder on YouTube (here's one).

EDIT: Look carefully at the beginning of the experiment in OP. You'll notice that the arc starts with something burning in a straight horizontal line. There was a conductor which burned apart, and that had established the initial arc (tunnel of ionized air).

• Why does the arc suddenly fade out then? – sharptooth Feb 3 '12 at 7:29
• @sharptooth The description of the YouTube video in your OP says that the arc was fed from a reactor. I'm not sure what the term means in power transmission systems, but I guess that reactor stores energy as magnetic field. At the end of experiment, the energy stored in the reactor may have dissipated. The crew may have also cut the power supply that fed the arc. – Nick Alexeev Feb 3 '12 at 7:37