Autotransformer teritary 5

[Mbrooke]

Why do utility companies spec tertiary windings on auto transformers? I'm told because of zero sequence current, however a member on here mentioned that a tertiary can be eliminated with no ill effect. How does this work? Is it system/specific dependent?

 

[davidbeach]

Without the tertiary the transformer does not source zero sequence current, but does pass it from side to side. With the tertiary the surrounding system is more effectively grounded and the voltage rise on the unfaulted phases during a ground fault is reduced. If your system design is based on the assumption of being effectively grounded you may not want to eliminate the tertiary. If your design standard is more aligned with a delta type system and the overvoltages they can produce then the tertiary may just be extra baggage for you.

Without the tertiary a three-phase transformer may exhibit a certain amount of phantom delta and source some amount of zero sequence current. If the bank is comprised of three single phase transformers there is no phantom delta.

 

[Mbrooke]

If your system design is based on the assumption of being effectively grounded you may not want to eliminate the tertiary.

That might be it. Everything from a relaying and transient perspective is based on the assumption of effectively grounded.

Without the tertiary a three-phase transformer may exhibit a certain amount of phantom delta and source some amount of zero sequence current. If the bank is comprised of three single phase transformers there is no phantom delta.

This is what has me so intrigued, these tertiaries are also present on 3 phase units when in theory they are not needed. Maybe the phantom delta phenomenon is not enough, I guess.

 

[davidbeach]

It's also really tough to feed the station service from a phantom delta... ;-)

 

[Mbrooke]

Well, not unless you have distribution at your bulk transmission substation

 

[prc]

Mbrooke, It is true that removal of stabilizing winding will increase Z0. But with 3 phase 3 limbed cored units, this increase is negligible or many times nil as the flux from three limbs of core (all unidirectional with zero sequence flux) will generate a circulating current in tank (Fantom delta) So it will not affect the grounding.(X0/X1< 3) In India we eliminated stabilizing winding from all 3 phase 3 limbed Yn/yn transformers since middle of 1970s. Thousands and thousands are in service for the past 40 years and we are yet to notice any surprises. We took this decision based on experience reported in US those days. See 1959 AIEE Paper by B A Cogbill ( Transformer engineer at GE Pittsfield factory) "Are stabilizing winding necessary in all Y connected Transformers". Those days both in US and France there were many EHV auto transformers banks working without stabilizing tertiary. Data were reported in Doble conferences of those days by P S Bellaschi ( A Westinghouse veteran who prolifically wrote on transformers during last century - pioneer who developed impulse testing for transformers in 1930's) In Italy they removed tertiary from their 400 KV auto transformers during 1990s.Turkey followed the practice later. In India we avoid supplying station loads from tertiary. With STATCOM and power electronics , reactive compensation through tertiary is no longer attractive. When we eliminated tertiary, we measured Z0 with tertiary and then opening out tertiary of transformers. With normal excitation, you can measure V0 at the corner of opened out delta. One issue that used to create problem in early days. An IEEE standard on Application of tertiary winding is under preparation PC57.158. This is under draft stage for the past 10 years.

 

[Mbrooke]

In India we avoid supplying station loads from tertiary.

Wondering if you can shed light on the reasoning behind this.


BTW, excellent information!

 

[marks1080]

If you use the tertiary you save costs on your station service transformer.... Now that doesn't mean the auto couldn't have been cheaper without the tertiary but the reasoning behind why the tertiary was there in the first place meant that you might as well use it for the station service. I guess the real question is does the money you save on your station service transformer (think about stepping down 500kV at a switch station to station service level voltages)outweigh the cost of an auto without the tertiary - and yes this is a horrible example cause why would you have an auto at a SWITCH station but you get idea...

All of this is moot if your system relies on the zero sequence filter...

 

[cranky108]

If you specify a tertiary winding for station power, do you specify the voltage? Or do you specify a special distribution transformer for that tertiary?

It is also possible to use a station power PT for station power. But rarely have I seen a location so remote that the tertiary was the only source of station power,

 

[Mbrooke]

@Marks1080: Exactly, if no distribution is present, that is one reason behind the tertiary. Even with 2.4-34.5kv present it is still utilized as it provides very reliable station service. There is no risk of loosing it from clearing a distribution bus or subtransmission bus.

@Cranky: To answer from my point of view the tertiary is speced at a common voltage. Sometimes it is also used to feed subtranmission with a complimentary zig-zag and feeder regulators where need be.

 

[marks1080]

cranky you don't need a remote station... any HV switchstation is an example. If you have a 500kV switch station next to a city or out in the boon docks, it doesn't really impact how you get 600V.

 

[davidbeach]

Where you get station service matters during outage restoration, particularly if dealing with an FOA transformer that can't support load without the cooling system running. If the cooling system is run from the tertiary you can get it running almost immediately after energizing the transformer if there isn't an alternate station service available. Be glad to see the end of the FOA transformers, but I'm sure I'll be long retired before the final one gets decommissioned.

 

[prc]

General thinking is any fault in the tertiary distribution circuit can affect a major power flow say 500-1500 MVA, if you take power from such large transformers. So auxiliary supply is provided from two sources- one from the station bus and another from near by distribution lines. In large substations, a standby diesel generator set also will be available. In India,OFAF unit coolers are used only with large GTs in thermal power plants to save foot print. General trend is ONAN/ONAF cooled units ie without forced oil cooling, up to 1500 MVA banks. Due to excess loss capitalization followed by utilities, manufacturers have to adopt current density below 2A/mm 2 (earlier 3-4A/mm2)With such current densities forced oil cooling is not attractive. Oil pumps apart from maintenance liability, increases auxiliary losses, create chances for metal dust entering in to windings etc,etc.

 

[Mbrooke]

Why not from all two or 3 autos? Also if you fuse the station auto (yes, not always doable due to fault current) you eliminate the risk of loosing the auto during a fault, but even then the risk is minimal- worse case to that being the loss of an auto which is very tolerable under normal conditions.

 

[prc]

Well I may not be able to comment on it further.Another problem esp with 765 kV autos is that tappings are on neutral end and hence the tertiary terminal voltage will change with tap changing.(variable flux regulation) Another solution found in Sweden is to provide a small winding on lower yoke of the core in EHV transformers to get small power (500 or 1000 kVA ) at 230V.
Another point tried in Italy along with tertiary removal was elimination of OLTC from 400 kV autos. In India also it is seen that when the grid became strong and fully interconnected, tap changing ( on 400 kV side for 400 kV variation) has no effect on voltage change. Then Why OLTC should be there in such interconnecting autos? Can't the voltage regulation be done through OLTC in downside transformers?

 

[Mbrooke]

My mistake, I meant placing station service on each of the auto territaries when they are present if not how it was read. Instead of one on the tert one on the dist bus.

Another point tried in Italy along with tertiary removal was elimination of OLTC from 400 kV autos. In India also it is seen that when the grid became strong and fully interconnected, tap changing ( on 400 kV side for 400 kV variation) has no effect on voltage change. Then Why OLTC should be there in such interconnecting autos? Can't the voltage regulation be done through OLTC in downside transformers?

We all think alike! I have evaluated removal of tap changers for some time... How did it work out in Itlay? I take it was a success.

 

[prc]

Italy they are procuring 400 kV autos without tertiary and OLTC for more than 20 years. In India also we have couple of 765 kV and 400 kV autos without OLTC.

 

[Mbrooke]

Honestly, the only place you really need OLTC as when stepping 33kv to 11kv or 132kv to 11kv. Am I right in thinking this?

 

[mgtrp]

My former utility had them for transmission intertie to control VAR flows between the transmission systems (different voltages). We would have struggled to operate the system without OLTCs!

 

[Mbrooke]

Was the system loaded near it limits?