Current Differential Shorting Test Switches 1

[proznick]

Hello,

I am designing a 10MVA hydro plant. I am using an SEL 387 for current differential. I had designed for shorting test switches between the CT's and the relay but was told that industry standard for the differential relay is not to install shorting test switches but wire the CT through terminal blocks to the relay instead.

I didn't quite see the logic in this as relays are to be tested regularily which would would be easier with the shorting test switches.

Does anyone have any insight into this logic?

cheers

RP

 

[cranky108]

I for one believe test switches would be useful.

I have run into consultants that don't put in test switches into there design, and when asked why they always say they forgot. Is there another view point other than a cost savings?

 

[davidbeach]

We always use test switches.

 

[scottf]

Given that test switches are relatively cheap, can't see a reason why they wouldn't be used.

 

[dpc]

We always try to put in test switches, especially with SEL relays. Of course, if someone closes the test switch with the relay in service, it is (probably) going to trip, unlike a standard overcurrent relay.

You need to disable the relay outputs in some manner before shorting a CT.

The SEL relays are tough to test without test switches since they are not drawout and do not have the removable paddles.

I've had discussion with plant engineers who were concerned about someone accidentally closing a CT test switch and tripping off their 1000 MW generator. That's a big drag for everyone. But in the end, the relay techs won the argument and got the test switches.

 

[scottf]

DPC...most test switches come with tag holes in the screws that hold on the covers. In critical installs, I've seen a lot of folks lock the covers in some manner, or at least place warning signs on covers.

 

[dpc]

Locks aren't too popular at most power stations, but you're right, that can be done. I think they are mainly concerned with risk of human error. Someone making a mistake and getting into the wrong test switch.

The warning signs are a good idea.

 

[slavag]

Good Morning.
For many years we putted test switch to all type of relays.
Today we stop with this prctic, we installed test switch only by customer request.
In many companies, we put attention, commissioning teams DON'T use them and inject current voltage from terminals directly.
What is a point. US/Canada used type of terminals with ring connection and very difficult connected to them with tester.
Eropeun companies used terminals like to URTK/S or UGSK with GS and easier connected to those terminals directly from testers w/o any additional plugs.
I see only one reason for the test switches, if you need test relay on the live feeder or replaced relay on the live feeder.
BTW, isn't so cheap. And if you have relay with 8,12,16,24 analog inputs you need not one test plug.
Just my opinion.
Best Regards.
Slava

 

[scottf]

slavag-

I'm admittedly biased, since I have responsibility for selling test switches to North American utilities.

With that said, a rack mount test switch with 30 poles is less than $500. I would consider that pretty inexpensive in a substation/control room application.

I think one of the main advantages is the self-shorting features for CT connections, which allows safe relay testing and facilitates safe shorting of CT secondaries in other events.

 

[slavag]

Scott...., Im agree. 30 poles is new for me, this price is great price.
Best Regards.
Slava.
BTW, are possible see link to this type of test switch.
Isn't promotion or selling, I hope.

 

[scottf]

Like to the GE switch offering is:

http://www.gedigitalenergy.com/products/brochures/FTRTswitch.pdf

30 pole rack mount (RT type) consists of 3 - 10 pole switches mounted in a rack mount assembly.

 

[oldfieldguy]

Test switches used with the newer, non-drawout relays are worth their weight in gold.

As one who spent a considerable amount of time testing protective relays, the advent of the microprocessor-based non-drawout relays brought a lot of challenges when a client wanted his "magic box" tested the dame way as its older, drawout electromechanical predecessor.

The need to disconnect and reconnect dozens of wires made a huge possibility of the testing creating more problems than not testing at all as wiring was incorrectly reterminated, terminals failed or screws were left loose.

Saving a few hundred bucks on the test switch installation is a hugely false economy in terms of future inspection, maintenance and troubleshooting.

old field guy

 

[ANGIBEW25]

Hi Proznick;
I am in total agreement with the group and Old Field Guy couln't have said it better.
I too come from the field and find test switches an absolute necessity.
As stated earlier you can prurchase the type that have "see-thru" covers, you can label each test switch and seal the cover. You will be able to read the labels thru the cover.
The money spent in adding test switches will pay you back ten fold in labor not having to lift any wires to isolate wiring.
I love SEL relays, but who wants to rewire every relay in a substation just to perform peroidic testing.

Good Luck.
Ang

 

[lz5pl]

To have or not to have (test switch) was already discussed some time ago. In my view both approaches have positive and negative sides.

With test switch:
Positive: Easy connection of test equipment (especially when there are company standard for pin-outs allocation).
Almost zero possibility to leave some circuit wrongly open or shorted after finishing of testing.
Negative:
Added cost - 500 USD doesn't seem cheap to me when installed in front of 1500 USD-overcurrent relay, for example.
Additional breaking point in sensitive circuits like secondary current and tripping ones.

Without test switch:
Positive:
Cheaper installation - specialized terminal blocks are installed anyway, with or without test switch.
Less space occupied on relay panel - this is not the biggest issue with modern relay protections, but sometimes should be considered.
Negative:
Needs qualified and responsible personnel for connection of test equipment through terminal blocks (well, relay engineer must be like that by definition!).
Original installation is modified during testing - some circuits are disconnected, some are shorted - hence higher possibility for mistakes left.

Considering both positives and negatives in our national grid was taken decision to omit test switches in newer installations. Using of specialized terminal blocks are mandatory - URTK/SP from Phoenix Contact for secondary current circuits, URTK/G for voltage circuits. Correct manipulation with shorting and disconnecting links is responsibility of test engineers. I am not the author of these requirements, but they fully correspond to my philosophy of circuits design and technology of relay testing.

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

 

[scottf]

lz5pl-

The cost of $500 is for a 30 pole rack mount. If you're just installing 1 relay, then you wouldn't need that many poles and could use a switch as cheap as $125.