Ferroresonance vs transfomer size

[Mbrooke]

Are larger size transformers les susceptible to Ferroresonance? Can 3 phase padmount and overhead transformers be ordered with modified saturation curves to prevent Ferroresonance?

 

[magoo2]

Larger kVA sizes are less susceptible to ferroresonance, but it can still occur with them. We had a 34.5 kV grd-wye grd-wye 2000 kVA unit experience ferroresonance last year with very small amounts of cable. These have the 5-legged core construction. It is more likely to happen with smaller transformers like 500 kVA, but it still happens with the larger sizes.

For 25 kV and 15 kV distribution, the larger sizes are more immune to ferroresonance, all other things being equal and the 15 kV is more immune than the 25 kV or 35 kV.

You can probably ask for a three phase padmound with three single phase core and coils and it will be immune to ferroresonance because, when you gen an open phase condition, you don't get coupling from the adjacent core leg.

I don't think it's practical to ask for a modified saturation curve. What you need is a linear excitation curve for the transformer over the possible range of voltages. I'd be interested in hearing what the transformer manufacturers tell you about this.

 

[Mbrooke]

Would a linear curve help over a modified saturation curve?

Also, you scenario peaks my interest. You say Grd Y Grd Y? Was this amorphous iron? My understanding is that wye wye is practically impossible to induce Ferroresonance?

FWIW most of the our units in service are 3 legged.

 

[magoo2]

What part of the world are you from?

I'm in the U.S. and the standard wye-grd wye-grd distribution transformer is a 5-legged core design. The 3-legged core design has some of the same concerns about ferroresonance. It also has significant tank heating concerns with open phase conditions. We had used this before the 5-legged core design.

That being said, with the 5-legged core, we normally get around a 2.1 per unit voltage on the open phase when ferroresonnce occurs. Ours is a 4=wire multi-grounded neutral distribution system. You could pursue that with your transformer vendors but you would probably need to have a linear sat curve up to about 2.5 pu or so.

It was originally thought that a wye wye (i.e., grd-wye grd wye) was immune from ferroresonance. Even Hopkinson, who published several papers on this subject based on TNA results, was fooled by this. That thinking was based on his tests with banks of 3 single phase transformers. You still had the same L and C, but no voltage for the open phase because they were not magnetically coupled, so no ferroresonance.

With a three phase transformer, whether it be 3-legged or 5-legged core, you develop about half voltage under open phase conditions on the open phase due to the magnetic circuit. This is a high impedance circuit but it will be sufficient to produce ferroresonance if you open one phase under unloaded or lightly loaded conditions.

My comments were worded in general. It applied to the common silicon steel as well as amorphous cores. We have done tests on both.

 

[prc]

The chance of ferroresonance in distribution transformers is incresed as primary voltage is incresed (>12 kV,single pole switching is used,cable length on in put side is increased, transformer size is reduced and secondary loading is light.Certain connections are more prone to ferroresonnace like delta/star,delta/starG and delta/delta. Chance of resonance is least with starG/starG and open star/open delta connections.4or 5 legged core transformers with star G/starG are more susceptable to resonnace.Ferroresonance cannot be prevented by using a lower flux density in core as the overvoltages still take the core to saturation.A good tutorail on the subject is available in IEEE Standard C57.105-1978-Guide for application of Transformer connections in 3 phase distribution systems.Type of CRGO used will not affect the phenomenon.

 

[magoo2]

CRGO??

 

[Mbrooke]

Inductive tank heating is one of the reasons I avoid Wye grounded-wye grounded. A delta primary is chosen to combat this and thus a 3 core design is usually found. About the only time I uses a grd Y grd Y design is when stepping 33kv to 12kv via padmount, and this is acceptable because the primary protection is 3 phase make/break via SCADA recloser as apposed to cutouts.

I have heard that frequently about Ferroresonance, where voltages over 12kv are more susceptible and is often a driving factor behind intentionally designing underground and sometimes overhead systems at 12kv rather than 22kv.


In your experience, assuming equal cable and line lengths, how much more likely is ferroresonance to occur at 12kv vs 23kv?

 

[prc]

Sorry-CRGO-Cold Rolled Grain Oriented Silicon Steel-type of silicon steel used in transformers. With YG/YG transformers of 3 limbed core, tank heating with un balanced load may be less as the tank will act as a virtual delta by the currents from zero sequence flux.

 

[Mbrooke]

Sorry-CRGO-Cold Rolled Grain Oriented Silicon Steel-type of silicon steel used in transformers. With YG/YG transformers of 3 limbed core, tank heating with un balanced load may be less as the tank will act as a virtual delta by the currents from zero sequence flux.

This is one of the reasons I avoid wye-wye. A cable fault or blown cutout can destroy a pad-mount in this scenario.


..............................................

Basically I have 2 options at hand for an none network underground padmounted and subsurface installation:

1. Run 4 22kv circuits via overhead (spacer cable) transitioning to underground leading up to PME S&C pads with SCADA switching and outgoing fused URD feeding padmount sub loops.

2. Run a 33kv over head line (spacer cable) with an extended tie (alternate source) over to SCADA recloser transitioning to 33-12kv pad mounted step-downs which will feed into fused pads outputting 12kv sub feed loops.

Situation 1 would be the most ideal taking a step out of the process, however my understanding is that a combination of long over head line lengths, underground URD plus higher voltage equates to greatly increased risk.

My thinking is, if ferroresonce can be mitigated by design 22kv would be a more practical choice.

 

[VTer]

You could consider a triplex core wye wye connection to help avoid ferroresonance and also avoid any tank heating due to unbalance in wye wye connection.

"Throughout space there is energy. Is this energy static or kinetic! If static our hopes are in vain; if kinetic ù and this we know it is, for certain ù then it is a mere question of time when men will succeed in attaching their machinery to the very wheelwork of nature". û Nikola Tesla

 

[Mbrooke]

That is a good idea, and not one I would discard for stepping down sub transmission to distribution.

I want to avoid wye-wye at the MV-LV level because it passes zero sequence (neutral load) currents.

 

[krisys]

CRGO stands for Cold Rolled Grain Oriented, type of speacial sheat steel commonly used to build the transformer core.

 

[Mbrooke]

So CRGO used will not increase of decrease the risk?

 

[magoo2]

So far we've been talking like ferroresonance is ferroresonance. However, the level of overvoltages is considerably different based on the transformer connection. As I mentioned before, with a grd-wye grd-wye connection, you can get up to 2.1 per unit voltages. With a delta wye-grd connection, the values are more like 5 per unit or more.

Thus, the primary connection is an important factor for any distribution voltage. In general, any ungrounded primary connection is more likely to have ferroresonance than one with a grounded-wye primary. For each major voltage classification - 15 kV, 25 kV and 35 kV - this will be true, and for each of these classifications, the higher voltage is more likely to result in ferroresonance.

Our base of three phase units are largely grain oriented silicon steel transformers. We have some units with amorphous steel but most of what we have are the plain vanilla grain oriented silicon steel units.

It seems obvious from some of the comments that this is a non North American installation.

 

[Mbrooke]

Excellent replies in so far!

So researching this some more I have come up with the conclusion that perhaps the best safeguard is breaker protection over fuse protection in all scenarios. Protecting underground cables via switches or fuses would eliminate Ferroresonance for a cable fault, and reclosers/gang switches for over head transformers where units may be susceptible.

The only risk I can think of is bay-o-net fuses blowing in a faulted pad mount, but in theory would be rare. Out of curiosity, in that regard, is there such a thing as relay that can detect Ferroresonance and trip a 3 phase device?

It seems obvious from some of the comments that this is a non North American installation.

Why do you think so? Just wonder lol .

 

[Mbrooke]

*edit, switches or breakers instead of fuses. My mistake.