Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations pierreick on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Transformer Differential Protection. Homopolar component compensation

Status
Not open for further replies.

MURILLO

Nuclear
Aug 5, 2005
7
Hi colleagues,

Actually we have a 3 bank power transformer with the following data sheets:

3 x 420 +-2.5+-5% kV / 20 kV
YNd11
3x420MVA - OF AF

The 87TG protection has conected an old design (dated from 1970) to compensate the homopolar component in the low voltage side, that include three CT in every winding phase conected in parallel and an auxiliary CT open delta / Wye used to conect to the 87 relay. Does anybody know this way to compensate the homopolar component ? Has anybody references about it?

The point is that now we are working to replace this relays and the actually digital differential transforme relay has features to filter the secuence zero component (on high voltage side) and this complex circuit shouldn't necesary.

Thanks

 
Replies continue below

Recommended for you

Homopolar is a rather unusual term, I suspect you are referring to zero-sequence. Any decent numeric transformer differential relay can have all CTs connected in wye and the relay will do the math necessary to compensate for zero-sequence currents appearing on the wye side of the transformer. Many such relays also have the capability of doing a ground differential (known as REF in the IEC world) to detect low current ground faults on the wye side.
 
Yes I mean the zero-secuence, and I am agree with you that the actual transformer differential relay have this feature to compensate this component when one of the transformers side is conected to ground. The point is that now we must replace a very old GE design to compensate this zero-secuence and I am trying to know in deeply the function of this circuit to get the best solution. I am sure that the function is to compensate the zero-secuence in an electromechanical relay, but until now I have seen to solve this with a CT auxiliary wye-delta to eliminate this component.

 
Check the manual of the relay you are considering; you should find a connection diagram that shows the power transformer in wye-delta and the CTs in wye on both sides. The compensation is done with a numeric filter that removes the zero-sequence component from the currents on the wye side of the power transformer. Mathematically it is a conversion from phase to sequence quantities, set the resulting zero-sequence component to zero and then convert back to phase quantities and do a phase shift to compensate for the phase shift in the power transformer.

The relay could also accept CTs in wye on the delta side of the power transformer and in delta on the wye side, but you will get better information from the relay if you have wye connected CTs. You can also do ground differential (REF) if you have the CTs on the wye side of the power transformer in wye but you can't if you have delta connected CTs.
 
Thanks David but it is not the case, the relay is a STD manufactured by GE and not show this conection. If you want I can try to send oneline diagram.

Anayway thank you so much
 
Murillo--

If you're indeed using GE STD's, then the instruction books are available on the GE website. In the manual for the STD15C, several representative diagrams are shown which depict the correct CT configurations for various transformer types. These should be followed to present these single-phase relays with currents in the proper relationships for proper operation. If the CT configuration cannot be followed as depicted due to other conflicts, then auxilliary CT's may be used to present the proper phase relationships, but this complicates the installation.

In my experience with commissioning many transformer installations, the improper connection of CT's for these relays was probably the most common one I caught. Many designers and engineers do not totally understand the circuits. The pictures in the book are correct.

The second most common problem was connecting the current inputs to the proper terminals of the relay, or incorrectly assigning the tap values. It's confusing because Winding 1 currents go to terminal 6 and Winding 2 goes to terminal 4.

New electronic relays do not require the external configuration of the CT's. Everything is connected in wye and brought to the relay, and the phase shifting and conversion is performed by the relay based on parameters the engineers enter into the relay program. Makes life easier in many ways.

old field guy
 
OK, the STD is not a numeric relay, but is a single phase electromechanical relay; of course it will require CTs in wye on the delta side of the transformer and in delta on the wye side of the transformer. But, why oh why, in this day and age would you even consider a single phase electromechanical relay? Other than watching the disk turn, there is nothing that can be done with an electromechanical relay that a numeric relay can't do, but there are many things a numeric relay can do that an electromechanical relay can't. For less than the price of one of the three STDs that you will need you can get a decent numeric transformer differential relay; for a bit more than one STD you can get a very good numeric transformer differential relay. You can also get numeric relays that will go into the hole that your existing relay comes out of. If redundancy is a concern, you can install two numeric relays for less than your three STD relays.
 
If you want to understand the existing circuit, you should be able to find it in any good protection book. If you are replacing it with numerical protection, I would remove the aux CTs and wire the remaining CTs in wye. You should be able to find instruction manuals on line for most of the new relays, including wiring diagrams and instructions for setting the relay to apply the phase shift and zero sequence filtering. Always bring the load up slowly while watching the operating quantities in the relay.
 
Thanks to for your help, but the point its that we are going to replace the STD electromechanical relay for a new digital, with the new features, instructions the circuit have been working for twenty years and it works correctly until now that it have had spurious trip, concerned to your help that you are talking about. I think that we have forgotten the main question and it is : actually we have a zero-secuence CT compensating cirucuit in the delta side transformer, until now I haven't seen it and I think that this design date from 1970, I have studied differents protections books, articles and talking with colleagues and I haven't found any information about it. This is the resason to post it. I think that it is very difficult to undestand a circuit without the online diagram but there is not way to post it here.
 
See page 265 (12 in pdf). ftp://ftp.areva-td.com/NPAG/Chap16-254-279.pdf

I don't see an open wye/delta CT connection, but this is how CTs were generally connected to electro-mechanical relays to get both differential and restricted earth fault (also known as ground differential)protection. To analyze, apply an SLG fault in zone and out of zone and trace the currents.

 
Murillo, I was confused when you brought up the STD. Any decent numeric transformer differential relay can accept wye connected CTs and perform all of the phase, sequence, and magnitude compensations mathematically.

Somewhere there is FAQ on this site that explains how to post images. Try a search for image post faq.
 
faq238-1161 (with a word of thank to itsmoked)

Murillo: Like indicated by previous posters, modern differential relays do not need extra equipment (zero sequence shunts, interposing CTs, etc) to keep the zone stable. You can take all those stuff out, connect the CTs in star, and take it into your relay. Just ensure your internal settings are correct, and there you go. You might need adding metrosils, if your relay is using a high impedance principle for a REF-scheme.

Can you just clarrify:

Have you already changed the electromechanical relay with a modern relay? (Seems not)

What kind of trip do you experience? (Seems like a LV REF trip, high impedance principle)

Regards
Ralph


[red]Failure seldom stops us, it is the fear for failure that stops us - Jack Lemmon[/red]

Make the best use of Eng-Tips.com
Read the Site Policies at FAQ731-376
 
Dear colleagues,
I try to answer us:

David: The STD were brought up, twenty years ago, every year the Maintenance Dpt. test it to ensure that works correctly, in other case replace it for a new one.

David&Ralph: I can't use the that show the FAQ238-1161, my server filter some pages.

Ralph:
No, we have not change the electromechannical relay, we are in design process step. We will use a new digital relay with all the features that we are talking about in previous posters, and for this we launched the original question, we won't use the CT that we are talking about, and like it is a very strange CT differential conection I am trying to know more about it.

The trips that we are experienced are due to external faults.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor