SEF protection 3

[linkect]

I want to detect a 2 Amp Primary SEF on an isolated 3 phase underground circuit. My primary current will be around 700-800 amps maximum when starting a series of water pumps. There is a requirement that the SEF at the 11kV outgoing feeder towards the Pump Station should be limited to 2 Amps. I can not detect that with a the specified OC/EF relay whose setting range is 0.05 to 2*Io with a 1000/5 A CT. This relay has a seperate input for a Balanced Core CT.

Could anyone advise me on how to accomplish this task? I am used to 20% and above EF settings. I saw something about Interposing Balanced Core CT's which I think may be useful.

 

[davidbeach]

The only way you can get there is to use a zero-sequence CT around the three phase conductors. If you try to use the 1000:5 phase CTs you will be looking for a single that is down in the noise. Your aux CT won't extract the signal from the noise.

 

[isquaredr]

Are the water pump motors 11kv rated or are you transforming down at the pump station. 2-amp primary out of balance seems a very small value to detect and I wouldnt be suprised if spurious trips are as a result of such a setting

 

[linkect]

Thanks Davidbeach, I was trying to avoid the process of fitting new relays on site, slotting holes on the new panels etc. It seems there is no way out other than the Zero sequency CT method.Which relay would be most suitable? Should I sue the Startco as on the 3.3kV side ( see below)?

Isquaredr: The 11kV circuit in question is from one 11kV switchboard 3km away to a local 11kV switchboard. Pumps are supplied through 11/3.3kV transformers at 3.3kV. A Startco Ground relay with NER monitoring has been proposed for the 3.3kV side as the transformers are NER grounded on the 3.3kV side. This relay can give us the low settings we are looking for.

 

[davidbeach]

Never heard of Startco, so I really can't comment.

 

[ScottyUK]

Echoing what David has already said, we reduced the settings on a residually connected Areva MCSU sensitiv E/F relay following a coordination study carried out by a third party. It led to a lot of nuisance tripping and we concluded that the CT matching under transient conditions such as motor starting was not good enough to allow use of such low settings.


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

I don't have an attitude problem. You have a perception problem...

 

[RalphChristie]

linkect:

In the mine I am working, we are using sensitive earth fault protection on the trailing cables of our mobile earth-moving equipment. The earth-fault current is limited to 25A on the 6.6kV-systems to reduce the step and touch potentials during fault conditions. Our setting is 400mA (0.4A) for 1second.

Like already mentioned, you'll have to use a core-balance/zero-sequence/ring-type CT around the three conductors.

A few comments:

The setting is based on trial-and-error. You'll have to measure the capacitive charging current in the cables and set the relay above that value.

Motors starting can be a problem, you'll have to decide either a higher setting or a longer tripping time-delay.

I can not comment on the Startco-relay, because I am not familiar with it. We have used Vigirex and Adit-relays for a long time without any problems, recently we changed to a SEL-relay (SEL351A) with a small neutral channel current input. I suppose most relay manufacturers will have a similar function in some of their relays.

Invest in some kind of fault-locating instrument. Many times it is not possible to find a damaged cable with the eye, due to the low fault-current. We use a surge-generator (thumper) - sorry to all the cable manufacturers - to locate our faults. In our environment it is better to risk a chance of damaging a cable more and locate the fault quickly rather than keeping the machines down for a long time looking and searching for the fault-location.

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

 

[isquaredr]

If your 11/3.3kv transformers are Dy vector group why do you require SEF on the 11kv u/g supply. For SEF on the 3.3kv side and to eliminate ct matching errors associated with a residual scheme why not feed into your SEF relay from a ct situated between the star point and NER.

 

[slavag]

Hi Ralph.
I would like ask you, why you don't use directional sensetive earth fault functionality?
In your application you need all time meas. capacitive current and change settings accordantly.
Regards.
Slava.

 

[RalphChristie]

Slavag:

In the new SEL-relays we are using, directionality is an option, but we do not use it. The cables always stay the same length - 1000m - a cable-reel is installed on each machine for propel (travel) issues.

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

 

[slavag]

Thanks Ralph.
Regards.
Slava