Current unbalance protection - advice needed

[lz5pl]

My customer requires to put in operation unbalance current protection function (relays are Siemens 7SJ62, function is Negative Sequence Protection (46). He cannot give me any information about acceptable unbalanced load levels in his grid. Grid is industrial, 6,3 kV, feeders supply 2 MVA, delta-wye, 6/0,4 kV transformers and 6 kV capacitor banks. No any directly fed large motor feeders.
Frankly speaking I haven't any experience with determination of setting values for 46-type protection. Could you advise me on the criteria for determination of setting values?

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[slavag]

Hi.
Maybe used criteria and avverage setting, that used for the industrial generator 46 protection. with two stages of DT. 7-8% for alarm and about 12-15% for trip. I think it's make sense.
But will wait to other recomendations.
Regards.
Slava

 

[ScottyUK]

The negative phase sequence current limit is inherent in the design of the machine so NPS relay setting requirements are usually defined by the generator OEM. They very considerably between designs of similar MVA - I've never seen a rule of thumb to estimate the setting and getting it wrong could result in a damaged rotor.



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

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

 

[521AB]

For the trip level (trip if fault)I agree with ScottyUK,
But for the non-trip level )security level, no trip if no fault), use the disturbance recorder when the system is in a "typical" operation and take screenshots of voltages and currents.
From the disturbances you can see the minimum level for your settings, because you don't want to trip for normal unbalance.

You can use your results to estimate the alarm levels, a least.

 

[slavag]

Scotty, 521AB.
Yes, each generator have NPS limit according to design.
And term "typical" what I used seems not correct. Agreed.
From time to time I used " excrtapolation" right or not, it's other story. Please check all types and sizes generators, what you know, you see NPS trip settings are higer ( for DT and IDMT). I'm not sure, but I think, for DOL MV motors too.
Now, please see OP, isn't generator , it is industrial grid w/o DOL motors. I assume w/o single -phase load too. trafos is delta/star. What is reason for the unbalance current???
According to our recorders in many such factoties, optimal level for the NPS alarm is about 5-8% ( in above post I wrote 7-8%), for trip level I used 70-100% more, with time delay of course.
In such grid I see only one reason for the NPS, disconnect one of infeed conducrors or not clear non symetrical fault???.
Regards.
Slava

 

[lz5pl]

According to me one of the reasons for that requirement for 46-protection is some maximalism of the customer. He would like to use as much of installed functions in the relays as possible. Nothing bad acc. to me, but it may lead to confusion in some cases. For normal feeders - 6/0.4 kV trafo and capacitor bank I also cannot see much sense in 46 protection, but I would propose values given by Slava with time delay coordinated with usual 51 protection. I use ELPLEK software for modeling of the grid, it is possible to insert also such protection.

There are two other feeders which concern me. One of them is a big rectifier - 13 MVA (!), fed through 6/0.26 kV transformer. Another feeder is compensation bank 3,8 MVA, consist of three stages for filtration of harmonics from the rectifier. I wonder whether some more specific measures should be taken for protection of them by 46 protection. According for me at least for the compensation it is not necessary, but for the rectifier I should ask the manufacturer which unbalance levels are acceptable. But if you have some ideas, any advice will be very helpful.

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[lz5pl]

521AB, thank you for the advice about disturbance recorder. I will discuss it with the customer. I am afraid they haven't disturbance records, as their existing protections on the grid are SEPAM and I am not sure whether they are fully numerical or only digital without fault recording capability. Anyway I appreciate the desire of these people to have the best and will try to help them even over the scope of that current order.

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[AMBMI]

lz5pl,
I think you should consider the 46 element as an important part of the compensation bank protection. The bank should consist of many single phase elements (capacitors) working in groups connected in series and in parallel; when some of them fail in a phase we get a negative sequence component (phase umbalance). I know that some compensation bank protective relays are using exactly such principle.
(happy easter!)

 

[lz5pl]

Good point AMBMI ! I should check the number of elements of the battery. The bank is complete delivery, so there is installed unbalance protection fed from a CT between star points of both sections of each stage. I will ask for factory settings of these protections (they are three relays, one for each stage). My task is to set protections from the switchgear side, so in this case I must coordinate with factory protection of the capacitor bank.

Thank you very much for Easter wishes. In our country it will be in April, as we are Orthodox Christians and calculation of the exact day usually differs from Catholics. But it is good to keep in mind Easter holidays in advance

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[slavag]

Oopp. Next customer with this: Relay are include those functions, I would like use them or I read, it's very important in case of my plant;-).
Plamen, I don't understand your Q exactly,sorry, he want use 46 for each bay!!!!! Try explain to him, only for alarms or may be as phase conductor disconnected.
AMBMI, good point, but have take in account: Motor unbalabce function and Capasitor Bank unbalnce function are not same. Usually you need use special capasitors bunk protective function 46, based on the CT at the neutral point and special overload curves.
Rectifier, maybe will problem. Here must used 521AB's recommendation.
Regards.
Slava

 

[lz5pl]

Exactly slavaq, 46 for each bay. When customer pays he is right to want the best (even in his understanding). But I am sure we will reach some common understanding, because the guy is quite competent, really. He is just happy with the bunch of available functions, but after the second look I am sure we will find the best compromise.
We have factory installed 46 protection in capacitor bank cubicles, I should back-up it with my relay from the switchgear.
My biggest concern is rectifier, and possibly also harmonic compensation filters.

By the way, another requirement was to use also Thermal Overload Protection (49) for capacitor bank feeders. For trafos and cables it is clear, but actually I have no idea about capacitor damage curves. Well, I cannot pretend I am very experienced in setting calculations, as my experience is mostly in utility grids, not industrial ones. May be I am missing something, but specially for capacitor banks I cannot see much sense in 49 function.

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[AMBMI]

lz5pl, I'm pretty sure that ABB and Cooper are publishing the capacitor damage curves (with dfferent damage probabilities %).
These curves were available inside the Cooper V-Pro II software.

 

[AMBMI]

Slavag,
regarding the capacitor protection two strategies can be used:
-a F46 on the bank connection
- a simple overcurrent on the neutral connection of the bank
I agree that the rotating machine unbalance protection is different.

 

[slavag]

Hi.
AMBMI, yes at the "old" systems usually used simple overcurrent relay. In the new relay I see also some "new" version with compenstation of natural unbalnce current.
Possible also other option, 3-phase capacitor bank unbalnce function, is depend on constaraction of CB.
Know about it only in theory, not used yet in my practic.
Plamen, no, Capacitor bunk protection is special protection
and have OL curve( 49) have another standard at the ANSI, IEC standards, is 51C and must also include undercurrent protection I<. Very important function for the capacitor bank and in additional you need add standard 51, 49 for the cable to capacitor bank.
Regards.
Slava

 

[RRaghunath]

lz5pl,

When protection for Capcaitor bank unbalance is taken care by the respective manufacturer, I agree that there is no need to set 46 element in the transformer feeder relays. It is also not prescribed by any of the protection reference books such as IEEE 242 (Buff book), NPAG.

More protection need not necessarily be better for the feeder / equipment. I would prefer not setting the element.