Two LV bus bars with Transformer Different Phasing / Phase Rotation

[SMA7875]

Hi,

We have come across an issue where we need to connect two 132/11kV power transformer LV (11kV) bus bars in parallel through an interconnector. The vector group of both is YN0yn0 however the HV bushing of one of the transformers is connected as Red phase to C Bushing, Yellow Phase to B Bushing and Blue Phase to A bushing of the transformer. Red, Yellow and Blue are network or system phases whereas A, B and C are transformer HV bushings. LV Connections are a, b, c following the HV A, B, C which means on LV side c will be red, b will be yellow and a will be blue. Now the other transformer has different configuration i.e on HV side Red to A, Yellow to B and Blue to C which means on LV side a is Red, b is Yellow and c is Blue. When the transformers are run independently this is fine however if one of the transformers is out of service and the LV bus bars are interconnected, the phasing will be different. We can’t change the HV phasing due to site constraints. We can change the LV cable connections so the question is will this solve the problem? I have attached a diagram as well. Your help is much appreciated.

 

[stevenal]

I fail to see a problem. Red phases with red, yellow with yellow and blue with blue as drawn. Just to be sure, check voltage across the open switch before closing.

 

[SMA7875]

Stevenal,

Thanks for your response. During the normal operation when the bus bars are running separately with the interconnector breaker open, its OK, however when one of the transformer is out of service, and we have to close the interconnector breaker, the previosuly Red phase on, say, GT5 will now be Blue phase from GT3, in my opinion reversing the previous phasing and the direction of motors. If we swap LV cables on GT5 to make a - blue and c - red to match GT3, we will still have the same problem when we connect the two LV Bus Bars.

 

[Marmite]

I'm with Stevenal. Your second transformer has a Red-Blue cross on both HV and LV sides which has no bearing on the overall HV-LV phase relationship as the vector group is Yy0. If it was a Dy transformer then that would be different. It's not unusual to come across this kind of situation. It's usually done to get the best physical disposition for the 132kV bars/connections.
If you have accessible 11kV connections you can make sure the two sides are in phase using phasing sticks, or potentially by comparing the two VT outputs if you have 3 phase 11kV VT's on the switchgear, or even auxilary transformer secondary outputs if fitted. The first option is the one to go for if you can as it gives a definitive answer.
Regards
Marmite

 

[stevenal]

"the previosuly Red phase on, say, GT5 will now be Blue phase from GT3". Your drawing does not indicate this. GT5 red (GT5 a) connects on through the interconnector to GT3 red (GT3 c). Either the drawing is incorrect or your understanding is.

 

[stevenal]

Sure you uploaded the right drawing?

 

[SMA7875]

I understand that for a Red / Blue HV cross for GT3 would be matched by an Red / Blue LV cross on GT5 as both are Y0y0. However when we ran simple voltmeter tests on VT secondaries (yellow phase earthed, I am in UK), it didn't phase out and the vector diagram from the readings showed something strange. It appears that in addition to a Red / Blue HV cross there is a Blue - Yellow LV cross on GT3. To make matters more complicated, it would mean that blue and not yeallow pahse is earthed on VT secondaries for GT3. To replicate GT3, We temporarily earthed blue pahse VT on GT5 (it is to be commissioned), crossed yellow and blue wires and the VT reading matched up. My question now is if we cross our LV (11kV) cables for GT5, earth blue phase on GT5 VT and connect the two boards, would it phase out or do we need to replciate HV cross AND LV cross to make it work.

Thanks for your suggestions.

 

[Marmite]

It's getting a little confusing, but if I'm following correctly you currently have clockwise rotation on the 11kV fed from GT3 and anticlockwise rotation on GT5 which would account for the odd readings. As GT3 is existing you need GT5 to have clockwise rotation too. On the transformer 11kV cable box cross the cables currently terminated on b&c terminals, or do the same on the 11kV switchgear instead if that is easier. Remember to check for repercussions with the transformer differential protection scheme. Ideally check GT3 & GT5 phase rotation with a rotoscope on the VT output or aux tx output to confirm the different phase rotation.
Regards
Marmite

 

[SMA7875]

Thanks Marmite,

I am also thinking on the same lines as the results suggest it. My biggest worry is if we need to alter the HV side of GT5 which would be a disaster as it is all built and bus bar connected in a tight old yard. I am not very good with vector diagrams and phase rotation. Is there a way to make sure, on paper, by a vector diagram that after this LV cross on c-b terminals for GT5, it will work in correct phase sequence. I am trying to get a porper phase sequence meter but have to report back in black and white with a vector diagram as well.

Appreciate if someone could help.

 

[Marmite]

Referring to your original diagram, everything works. GT3 is in phase with GT5. You then stated that GT3 has a y-b cross on the LV side. To make GT5 phase with GT3 you now need to replicate that y-b cross on the LV side of GT5, which again referring to your diagram means swapping the cables connected to terminals b&c on the transformer. You are fortunate that you have a Yy transformer which means you can sort it all out on the LV side. The point I made about the differential protection is that if you change the HV connections within the protected zone bounded by the CT's you will need to change the corresponding CT secondary connections to maintain stability.
Regards
Marmite

 

[mgtrp]

Before swapping too many primary connections, are you certain that the crossed phases are on the primary, and not on the secondary wiring of the VT? This is easily checked if you open one feeder and close the tie breaker - you should see perfect synchronism between the two VTs.

 

[stevenal]

I agree it's confusing. I would correct the drawing before trying to figure out what to do next.

Phase rotation is determined by the system, and cannot be altered with transformer connections. Diagram indicates it is RYB, but even if incorrect, all that matters is that motors turn in the right direction.

With wye/wye transformers you can make any corrections needed on the low side, leaving the high side alone.

Of course if you swap phases at one of the transformers, the loads on that bus will also need to be swapped. (assuming motor rotation is correct now, and only a single swap was done)

Whether or not differential relay CT secondaries need to be swapped depends on CT location. If using bushing CTs and you have stability now, no changes need to be made.

I'm concerned with swapping VT secondary connections around to check for phasing. This depends entirely on how things are wired, and we already know that at least one drawing is wrong. I would have some qualified people use a set of phasing sticks across the open switch prior to closing it.

 

[ScottyUK]

Can you earth the star point of your VTs rather than a phase, just as a temporary measure? If so, repeat the phasing checks and post the results. I have a slight suspicion that you are getting some weird results as a result of how your VTs are earthed.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[gvasiliou]

If 132 kV Side is not yet connected to the grid or can be easily isolated you may try this if you like.
Isolate both GT3 and GT5 from 132kV grid.
Force closure of "interconnector" and GT3 and GT5 breaker.
Find a smart way to supply 480V AC to the primary side (132kV) of GT3 or GT5. You should get 40 V AC on the 11 kV Side (assuming 30kW iron losses per transformer you need something like 140Amps at 480 Volts to energize both transformers).
With your multimeter check each phase of GT3 vs to GT5 at 132 kV Side (480Volts) . When you have a phase match you should read zero in you voltmeter. When phase doesn't match you should read 480V.
In this way you check (with one test) both primary and secondary crossings if any.

PS : By the way, it seems from your drawing that GT3 and GT5 has opposite phase rotation. GT3 has a-b-c rotation and thus can be matched only with one of the combinationa a-b-c or b-c-a or c-a-b. But GT5 has a-c-b rotation which is opposite to GT3.
(opposite rotations : c-b-a / b-a-c / a-c-b)
Recersing the LV side of the transformer i believe should do the job. Reversing high side would make the installation perfect. I would go for matching primaries and avoid tricks to secondaries.

 

[SMA7875]

Many thanks for your responses. The problem was all 132kV connections for GT5 are done, would be a disaster to chnage, we couldn't switch off or change VT earth for GT3 and the 132kV tie breaker due to load demands however we have done some more voltage measurments between GT3 132 and 11 and GT5 132 and 11. We already had the data between GT3 and GT5 LVs. It is evident now that there is a Blue-Yellow LV cross on GT3. The configuration now is

GT3 132
A-Blue
B-Yellow
C-Red

GT3 11kV
a-Yellow
b-Blue-earthed
c-Red

GT5 132
A-Red
B-Yellow
C-Blue

GT5 11kV
a-Red
b-Yellow
c-Blue

What we think we need to do (and thank God it doesn't include any 132kV modifictions)

GT5 11kV (Cable Swap)
a-Red
b-Blue
c-Yellow

Secondary Side

VT
a-Red
b-Blue-earthed
c-Yellow

We will swap the GT5 Differential CT connections Yellow-Blue (SIEMENS Duobias Relay) as well as LV Directional Overcurrent ( MiCom P142 Relay). Also a setting change for GT5 AVC (Microtapp Relay). Not sure about the LV power transducer (Alstom iSTAT500) but I presume if we are using L1, L2 and L3 in the same sequence it doesn't check the phase rotation etc. Will give Alstom a call.

Lesson learnt; confirm the phasing and rotation in early stage, never rely on visual connections to assume phasing (as there was no drawing or info available).

Could you please send me links to literature about phase rotation and how to draw it?


Many thanks for your responses. It really help to confirm our understanding.

 

[stevenal]

You cannot correct at GT3? Looks like you will have a system where color x will not necessarily phase with color x, and the relationship of that color is not defined as related to system phases.

Also looks like may end up with a system that does not phase across the switch. This will force you to drop load before making the transfer. If correctly phased, you can instead do a make before break transfer in order to remove a transformer from service while keeping the load on.

I think you may be off track with your suggested differential relay reconnections. See my post above.