Theory question: why do Petersen coils help?

[Curious nerd]

AC distribution systems are sometimes designed with a safety device variously called a "ground fault neutralizer", "arc suppression coil", "Petersen coil", or "resonant earthed neutral". If you Google, try every possible mis-spelling of "Petersen".

It adds inductive reactance to cancel out the capacitive reactance of the power lines. This is supposed to mitigate line to earth faults.

Nobody explains how, so the reason must be obvious and I'm stupid for having to ask. I would have expected capacitive current, at 60 Hz and the low capacitance of power lines, to be a really small problem compared to the real component of the fault current.

Guess #1: When a line hits a tree, the tree is a capacitor as well as a resistor, so tuning the circuit can reduce the problem to the tree's conductivity.

Guess #2: If the line develops an arc to earth, then if it's running with current and voltage in phase, both go to zero at the same time and it's much easier for the arc to self-extinguish. At zero crossing there's no current to keep the air hot and ionized, and no voltage to restrike the arc. So, a tree branch swings in the wind, touches the line, draws an arc, swings back, and the arc goes out instead of being self-sustaining.

Guess #3: For the same reason, the protective devices can work better.

There are lots of references to the "what" of how the system works, but I've had trouble finding the "why" behind its benefits.

Thanks for your patience. It's not just me, there was someone here in 2008 asking a similar question. That discussion died from a language barrier.

 

[Phil Corso]

CuriousNerd...

Most of electrical-faults are arcing-faults. The damage can be substantial… visualize a welding machine! And as you said the P-coil has other names. Another is the Resonant-grounding system.

Its principal advantage is the fact it can substantially reduce damage to iron-structures by reducing arc-effects quickly. A second advantage is that unlike a Hi-R grounded system (also employed to minimize iron-damage) the arc can be extinguished before restrikes can occur.

A disadvantage is that it must be used where system parameters, like phase-to-ground capacitance, is fairly constant. Thus its not likely to used in areas where tree-strikes are plentiful!

Regards, Phil Corso

 

[Curious nerd]

Thank you, Phil.

What makes adding inductance reduce arc damage?

Guess #4: the coil makes a parallel resonant circuit with the system capacitance, which reduces current to theoretically zero. But that can't be right, can it, because there are so many cheaper and simpler ways to prevent current from flowing?

 

[GoldDigger]

Thank you, Phil.

What makes adding inductance reduce arc damage?

Guess #4: the coil makes a parallel resonant circuit with the system capacitance, which reduces current to theoretically zero. But that can't be right, can it, because there are so many cheaper and simpler ways to prevent current from flowing?

What you want is to be able to draw normal current through the load resistance which is in parallel with the capacitance but still provide a high impedance when a point on the transmission line is shorted.
Limiting the current using reactive elements means that they do not have to dissipate enormous power the way a resistive element would.

 

[Curious nerd]

This is helping. Thank you. I'll stare at circuit diagrams again.

 

[topgone]

This is helping. Thank you. I'll stare at circuit diagrams again.

There is a simpler way in how Petersen coil works:
In an ungrounded system, the system is actually referenced to ground through the capacitive reactance of the system. When a ground fault occurs, the faulted phase voltage goes to ground potential, while the remaining phases increase to 1.732 times the line to line voltage. The fault current which is very capacitive (leading the voltage by 90 degrees) is three times the normal charging current of the system.

By grounding the neutral of the system with an inductive reactance, the resulting current becomes near unity(system capacitive current gets cancelled by the inductive grounding current)->meaning, the current is in-phase with the voltage. The fact that the resulting current will be almost in-phase with the voltage, there will be lesser chances of arc re-strike because both reach the zero magnitude at the same instant.

 

[Curious nerd]

Topgone, thanks, I think it's beginning to get through to me. Reminding me that they're isolated from ground helped things fall into place.

So it's not like a solidly earthed system, because in a solidly earthed system and especially MGN, the fault to earth would follow through wet dirt to the grounded neutral, and fault current would be high and ohmic?

In one of the systems I'm struggling to figure out, a line to earth fault becomes a line to line fault carried by the displacement current between the suddenly energized dirt and the remaining healthy overhead lines? The arc is therefore fed by an almost pure capacitive circuit, and if there were no capacitance to earth there'd be little or no current? And if there is an arc, there's current keeping it hot while the voltage crosses zero, and voltage re-ionizing it when the current crosses zero, so the arc is more likely to keep going like the ones on Youtube?

So is there a double benefit from the tank circuit? It limits current through the fault circuit with its high impedance (but why not a breaker?), and on top of that it power-factor corrects the current so the odds are better that the arc will self-extinguish?

So the customers are happy, because their motors aren't banging on and off in rhythm with the recloser because there is no recloser because it's not needed? Or there is one with a high trip point?

Am I even getting close?

 

[Sahib]

In a solidly earthed system and especially MGN, the fault to earth would follow through wet dirt to the grounded neutral, and fault current would be high and ohmic?

Yes, but the fault current is generally of inductive nature.

In one of the systems I'm struggling to figure out, a line to earth fault becomes a line to line fault carried by the displacement current between the suddenly energized dirt and the remaining healthy overhead lines?

That is a disadvantage of power system one with Peterson coil: if single phase low resistance ground fault is allowed to persist for a long time, it is liable to be converted to phase to phase fault.

The arc is therefore fed by an almost pure capacitive circuit, and if there were no capacitance to earth there'd be little or no current?

May be. But if the ground fault has low resistance, considerable purely resistive current could flow.

And if there is an arc, there's current keeping it hot while the voltage crosses zero, and voltage re-ionizing it when the current crosses zero, so the arc is more likely to keep going like the ones on Youtube?

The arc is really of intermittent nature and that causes over voltages in the power system.

So is there a double benefit from the tank circuit? It limits current through the fault circuit with its high impedance (but why not a breaker?),

It permits purely resistive current to flow through the ground fault and in such a case, per Swedish wiring regulation, when the ground fault resistance is less than 3 k Ohms, the MV network must be disconnected within 5 seconds.

it power-factor corrects the current so the odds are better that the arc will self-extinguish?

Yes.

So the customers are happy, because their motors aren't banging on and off in rhythm with the recloser because there is no recloser because it's not needed? Or there is one with a high trip point?

Reclosers are still needed to clear transient line to line faults.

 

[Phil Corso]

CuriousNerd...

The arc itself is a restive!

Phil

 

[Curious nerd]

That's a big part of my mental block, in fact. I have trouble looking at a short circuit through plasma and seeing it as capacitive current.

Thanks for all the real-world power engineering expertise.