Why do faults result in an arc on high-voltage systems? 4

[ee_123456]

Hello All,

I am a young engineer trying to wrap my head around arc flash protection within substations.

Here is a simple hypothetical.

Say you have a 138 kV 3-phase breaker with the following spacing:
Phase-phase = 6 foot
Phase-ground = 7 foot

Now if you got a jumper (let's be theoretical here) and shorted all the breaker windings together, you would have a 3 phase fault and it would produce a huge flash of electricity and an arc.

I am trying to understand, why does this arc form? From what I understand, the arc energy is heavily dependent upon your arc gap. However, at 138 kV LL, and given the dielectric breakdown of air is 3 kV/mm.
The phase-phase spacing being 6 feet or 1800 mm, I am calculating you'd need 5400 kV to ionize the air and produce that big flash of electricity that we typically see with faults. So then how are you ionizing the air? and producing an arc flash ? The same logic applies for say a 230 kV transmission line where a fault produces an arc, but the phase spacings can be like 10 or 11 feet long so theoretically your spacing is too high to ionize the air.

 

[davidbeach]

Something other than air enters into what you're thinking of as just air.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[ee_123456]

Lets say this is a aluminum terminal/bushing.

Are you saying that as the current is heating (and potentially melting) say the aluminum, that is affecting the dielectric breakdown therefore you have less kv/mm for the fault to arc?

The reason I was asking is because those arc-flash line to line and line to ground arc gaps as defined by many standards (IEEE standard C2) appear to be significantly shorter than typical phase spacing for 138 or 230 kV systems. It appears those standards are calculating air gaps using a fixed 10 kv dielectric strength. If we're saying that the air dielectric strength gets reduced from that (therefore allowing arcing over longer distances), wouldn't that mean those standards are under estimating the arc gap to begin with?

 

[davidbeach]

Something, that isn't air, enters the "air gap" space. Could be a bird, could be bird streamers. Could be certain rodents (but they're much more problematical at distribution voltages than at transmission voltages), could be vegetation. Could be an oops from a crew. Could be insulator contamination or failure. Could even be smoke. But, dry, clean, air is your friend.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[ee_123456]

if you don't mind me asking then, if we're saying that there's no problem with the fault arcing phase to phase or phase to ground due to the bird or tree that got in the way.
For example ANSI C2 Table 410
Then why do these standards list the arc gaps so low for a high-voltage system where the phases are so far away?
I attached a snippet. It looks like we're using 0.45 meters so 1.5 feet for 230 kV voltage where the phase spacing is 10 feet.

From C2-2017:
"Arc gap—calculated by using the phase-to-ground voltage of the circuit and dividing by 10. The dielectric
strength of air is taken at 10 kV per inch. See IEEE Std 4-1995."

 

[LionelHutz]

Arc gaps are supposed to break down to protect against an over-voltage and also break down first before a flashover somewhere else occurs.

 

[sushilksk]

Once arc is produced, air is no more insulator. It is chain of ionized particles. Elongation of arc will reduce the current but cannot make it Zero. When arc is extiguished at any current Zero, recovery voltage appearing across ionized particles will re-establish arc, because air does not regain its dielectric properties back instantaneously

 

[jghrist]

The spacing is large enough to prevent large overvoltages like switching surges and lightning from flashing across the gap before aurge arresters operate.

Now if you got a jumper (let's be theoretical here) and shorted all the breaker windings together, you would have a 3 phase fault and it would produce a huge flash of electricity and an arc.

If you connected the jumper and bolted it to the phases before energizing, then you would not get a huge flash unless the jumper melted. It would be a "bolted fault" with no gap. If the breaker were energized when you put the jumper in, it would flash when the jumper got close enough to arc over.

 

[dik]

Elongation of arc will reduce the current but cannot make it Zero

Is that why a Jacob's Ladder climbs and the arc increases until it collapses and starts again?


Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[Palletjack]

Calculate the breakdown V of clean DRY air as David mentioned.
Now calculate the breakdown voltage of “smog”,
Fog, torrential rains, etc..

Also, as mentioned, something else entered the gap.

 

[ee_123456]

All,

I was reading this paper by EPRI.

https://www.epri.com/research/products/000000000001022632

It turns out a few things.
1. Most typical gap values have been calculated using much lower voltage equipment (under 15 kV).
2. Arc pro which limits gap length is not verified for transmission and substation scenarios
3. For transmission line and substation situations, the gap length can reach 2 ft to 12 ft and more. Meaning it doesn’t make sense to apply the dielectric breakdown of clean dry air as is calculated using the standard methods to drum up the arc gap.

 

[ee_123456]

Does the conductor melting cause an arc due to molten material causing air to be more easily ionized? I’ve seen a video where a tree fell on a power line and once the tree fried, a huge arc was formed.

 

[sushilksk]

IMO, arc cuases melting of conductor. After melting, molten material will go down by gravity.

 

[ee_123456]

I think I now finally understand.

Arc refers to fault across air. This can happen if conductor gets too close together, or can happen if say a tree falls onto conductors, tree will melt down and then there will be gaps in tree which will cause rapid ionization of air. And once air is ionized, arc can enlarge as air cannot be immediately de-ionized.

Most faults in power system are not bolted but rather are arcing fault.

However, I have come to conclusion that an arc can start from small air gap but get larger if the tree for example melts and gaps begin to form. You can theoretically have phase-phase arc evne in 230 kV transmission line (which is like 10 foot apart). Is my understanding correct here?

 

[bacon4life]

Also, the gap can start at a distance of zero when a switch opens. YouTube has several examples of 30+ foot arcs through ionize air.

 

[JezNZ]

Most faults in power system are not bolted but rather are arcing fault.

A bolted fault is for most intents and purposes a theoretical fault.

It will only occur in practice when a mistake has been made (such as earth switch left closed or commissioning/maintenance procedure error).

Other answers have given good reasoning for why arc faults are most likely, often developing from a short or partial short, the cause of which is then removed or destroyed.

 

[davidbeach]

Many real faults are bolted faults to within the tolerances of the system fault model used. Certain phase-ground faults are less like bolted faults, but nearly all multi-phase faults are bolted faults as far as the fault magnitudes are concerned.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[crshears]

The distance between arcing horns is usually a fractional ratio of the equipment the horns protect so that any over-voltage will induce flash-over at the horns rather than anywhere else, therefore damage that occurs due to arcing will be to the horns, which are easily and inexpensively replaced, as opposed to the much more expensive and generally much less easily repairable transformer, or generator, or cable, or pothead, or bushing being protected.

One example of "something other than air": grass fire under a 28 kV feeder caused feeder trip due to ionized smoke particles greatly reducing dielectric strength of normally much cleaner air between feeder phases and ground.

Tree hung up on feeder may cause fault but without much arc.

Or: closure of switch to make parallel between two different feeders takes place, but one phase of switch is out of adjustment and doesn't go into the jaw, if visual inspection is not performed, when parallel is broken using a different switch, a dirty yellow power arc will develop across the higher resistance improperl closed phase and will soon cause some mayhem.

 

[Gr8blu]

ee 123456: Power systems (read "substation" or "outside the building") exist in atmosphere, which is not "just" air. It is everything floating around in the air as well - such as dirt or debris, moisture, ionized particles, the occasional low-flying alien, etc. More importantly, there are factors that affect the insulating properties of "air" that have nothing at all to do with the odd bit of animal, vegetable, or mineral matter carried along on the wind. These include: altitude, humidity, temperature, relative movement of the gas (read "wind"), and the chemical composition of the gas mixture itself. Then we add in the effects of actual "debris" which may or may not be conductive.

The presence of ions decreases the insulating properties - as does a reduction in density caused by altitude or temperature increases, and an increase in moisture. Some gases are better insulators compared to others - hence the use of inert gases as coolant for some types of electric machines.

Like anything else, electrical energy will follow the easiest path to get from one point to another. If that happens to be across a small distance occupied by ionized air, then that's what happens. If it's by traveling the long way round via a metallic conductor, then takes that path. A longer, cleaner gap through a nominally-insular material (like atmosphere) is typically the "hard" way.

Given enough voltage (e.g. the "push"), an arc will cross any length of gap if the conditions are right. Think about it - that's what a lightning strike is, in essence. It jumps from one potential to another across THOUSANDS of feet of open atmosphere.

Converting energy to motion for more than half a century