Geomagnetically Induced Currents 2

[Mbrooke]

Does transformer connection make a difference in GIC currents and their mitigation such as delta wye 345kv-69kv vs a typical Wye Grd auto? Also are GSUs effected by GICs such that it needs to be evaluated?

 

[magoo2]

Any large transformer with a connection to earth is susceptible to GIC. So that would include delta wye-grd, wye-wye and autotransformers.

Our GSUs are mainly delta (low side) to wye-grd (high side). In the U.S., a New Jersey utility had a GSU failure because of GIC. This has been maybe 20 years ago.

 

[Mbrooke]

What about a wye grounded secondary (69kv) traveling down a 69kv delta-wye distribution power transformer? Technically its grounded at only on point, so GICs can not flow, unless the LAs conduct it?

 

[magoo2]

For a distribution power (substation) transformer, these are less susceptible because of their smaller MVA size and there are more of them in an area compared to GSUs or large autotransformers.

The flow of GIC is a dc current that tries to go to earth. With a wye-grounded connection on one side of the transformer and a relatively low impedance to earth, the dc flow produces lots of harmonics in the transformer.

There are a number of rules of thumb and generalizations about GIS, including
- locations closer to the earths's magnetic poles are at higher risk
- long transmission lines running east-west are most suceptible. North-south transmission lines are relatively immune to GIC.

There's a number of free EPRI reports available on the subject of GIC and I'd encourage you to look at some of them. Just search EPRI reports on GIC.

 

[cranky108]

It also depends on several other factors, like ground resistance, direction of the line, neutral resistance, etc.

I doubt there would be much voltage across most distribution lines.

 

[prc]

The effects of geo magnetically induced currents on transformers are not that devastating as being considered so far esp for 3 phase 3 limbed transformers. Please see the latest tutorial on the subject-IEEE Std C57.163-2015 Transformer capability under geomagnetic disturbances. Wherever there is a grounded star winding, DC currents induced on earth layer will enter through neutral of transformer, come out through line terminals, travel over the transmission lines and enter in to a remote transformer and again enter to ground through the neutral of the remote transformer. In that process both transformer cores will get magnetically saturated, drawing appreciable excitation currents and resulting in over heating, vibrations and noise.

 

[MatthewDB]

How do series compensating capacitors do at blocking GIC? I would think at 345kV and higher, the voltage rating of the series of the capacitor would be high enough to stand off the induced voltages.

 

[davidbeach]

We have a GIC detector on one GSU. All it ever detected was asymmetric firing of a nearby Thyristor Controlled Static Compensator (TCSC). Periodically one stage would not fire on both the positive and negative halves of the wave. That 2nd harmonic would be detected, and would cause the transformer to growl fearsomely, but never caused much trouble. Often it was reports of the noise that would cause people to dig into what the GIC monitor might have discovered. The TCSC was removed (or deactivated) a few years ago and there's been nothing of note since.

 

[Mbrooke]

Is the GIC detector a hall-effect sensor on the neutral or am I thinking of something else?

 

[Mbrooke]

Also, regarding the system as a whole- how would isolation interconnecting transformers vs auto interconnecting transformers make a difference during a GIC event? By interconnecting I am referring to 345kv to 115kv at bulk substations, ect.

 

[prc]

As far as GIC is concerned, whether two winding or auto connected, effect is same. It requires DC current flow through one winding alone to cause core saturation. Siemens is marketing a device that can be fitted on transformer neutral grounding line to block DC current. In India, being near to equator, we never had GIC issues. But nearly 15 years back we had a similar problem. One +- 500 kV HV DC line was operating normally, but when one conductor was out they tried to transmit half the power by passing current through ground return path. Near to one end station, there were rocky areas that gave very high resistance to DC currents. Then part of DC went to AC transformer neutrals at near by station went up and travelled over 132 KV AC lines and at remote end came out to ground through another transformer neutral and continued the journey to the other HV DC station! AC transformers started howling creating panic everywhere!!

 

[Mbrooke]

PRC: Did not know that was possible- but it makes perfect sense. Monopole operation has to be considered carefully.


Bare with me here- in regards to detla wye vs auto- say there were GICs on the 345kv system- they would not pass straight through to the 69kv system- correct? Rather current would have to find a wye secondary at one point of the 69kv system, up through the 69kv system- then out another wye point X distance away, correct?

 

[marks1080]

Put a CAP on the neutral of the unit to block GIC currents, not in series... In fact, I think it's safe to say that any power system where all transformer neutrals are grounded through a cap is essentially protected from GIC.

 

[Mbrooke]

How big the cap and what if any changes to protective relaying?

 

[HamburgerHelper]

Even if the cap was big, I think you could get around that by putting a new inductor in series with the capacitor to make the overall grounding unchanged. The cap isn't compensating anything. It is just there to block DC.

The thing that is odd about GIC is that the transformers aren't damaged if they go into hard saturation. If they saturate, they will trip on saturation current. The damage is done if they semi-saturate. Just enough to not trip but enough to cause damage if sustained.

The best solution in my opinion is to just shut the system down. The world doesn't end if the power is out for a few hours and that is cheaper and preferable than trying to ride it out.

 

[cranky108]

A capacitor introduces some impedance, which could make detecting ground faults more difficult.

Placing an inductor in series just introduces more impedance, so not much help.

Needless to say, delta systems don't have a GIC problem.

If one were to not load a transformer fully it is possible to not saturate the core during a GIC event.

 

[marks1080]

From the point of view of protecting a power system from solar flares, I think its probably better to just accept the temporary blackout every 40 or so years. From the point of view of protection a power system from an EMP style attack I think the discussion gets a lot more interesting... I think the military complex can do a better job of convincing policy makers to install neutral caps vs. power system engineers hahaha.

 

[HamburgerHelper]

Cranky,

I meant to use an additional inductance to cancel out the series capacitance so you wouldn't change the fault current. Xc = Xl, Xc*(-i) + Xl(i) = 0.

Transformers are the least saturated when heavily loaded. I don't know if loading them more or less would help. Less loading would mean they would more likely trip out due to further saturating. Heavy loading might help pull them out some of saturation. You don't want them to sit in the region of semi-saturating.

 

[Mbrooke]

Sounds counter-intuitive tripping transformers out during a flare to load them up, but then again, it protects some of them from failure in the process.

 

[bacon4life]

Yes, a delta/wye transformer blocks GIC on the delta side. GSUs typically have high side wyes, so they are paths for GIC flow. Distribution substation transformers around here typically have high side delta windings, so they are not GIC paths to the transmission.

The blocking capacitors on transformers neutrals include an arc gap/surge arrestor bypass that activates during fault conditions. This one from Siemens looks to be in cabinet of 1 m by 1 m by 2m.

https://w3.siemens.com.br/topics/br...dfs/hvdc/DC_Blocking_Device_20130920_A06a.pdf

A military EMP event is totally different; the typical mitigation actions for a solar flare GIC provide negligible mitigation against an EMP attack.

There is little consensus as to what level of solar flare will cause enough GIC to damage large portions of the power grid. It would be interesting to figure out prearranged criteria for making a decision to shut down the grid. Intentionally shutting down the grid trades off the possibility of saving the grid from long term damage, with the certainty of death and severe economic damage from a temporary blackout. According to Wikipedia, the US east coast blackout in 2003 contributed to almost 100 deaths.