Relay Test Switch Philosophy 1

[rockman7892]

I was interested to hear from others related to industry standard or experience/preference related to test switches on relay inputs & outputs. I have seen it done a few different ways in the past:

- All inputs and outputs wired out to two test switches (both sides of relay input/output)
-Test switch located at the positive input only for each relay input, and output (negative side) of each relay output
-Test switches only on outputs that go are connected to a load (IE breaker or LOR coil).

Is there usually an industry standard that is followed here or is it usually preference based?

Also when brining CT's into a relay cabinet that are connected to a shorting test switch before the relay is it usually common practice to bring the CT's into a shorting block in the cabinet first before hitting the test switch? Was curious if the shorting block was standard practice or if it was redundant with a shorting test switch in the circuit?

 

[thermionic1]

I'll take a stab at this. Test switch application varies a bit, but in general in a utility application having the following is generally considered a "must"

1) (3 ph) Voltage / Current on a single test switch. Va, Vb, Vc - Ia-, Ia+, Ib-, Ib+, Ic-, Ic+, Vn. This layout is fairly popular.

2) Both sides of the contact available. - This allows for easy testing with a relay test set. It's also preferable to keep all trips on the same test switch. Relay techs have paddles that are designed to work with the FT test switch. This also isolates that live blades and can help prevent an inadvertent operation due to test leads falling off, etc.

3) For inputs one side is fine. Some have the contact feeding the input on the (+) side, some the (-) side. Of course, the switch has to be wired to allow simulating the input (DC +>>52a>>Sw>>Relay). It is also a good idea to provide access to either the DC+ or - at a test switch (Relay power for example). This allows the station DC to be run through contacts on the relay test set and then back into the relay for testing. A must for 52a in reclosing applications, BFI, etc. Many distance relays have SOTF, LOP, etc that need to see 52a, else people are changing settings to test and at that point what is being tested.

4) It's very important that the blade of the switch goes to the outside system and the jaw feeds the relay. This follows the old electromechanical relays. I've seen on a number of times this wired in reverse and it defeats the test paddles that are used. I've questioned the designs and heard that the designers didn't want live blades. My response has always been that it's 67 Vac and 63 VDC to ground on the blades and if the person is not qualified to work near these small voltages, they have no business in a substation or opening the test switch cover. Having the hang ten test leads with alligator clips to the switch blades creates a much larger safety and Human Performance hazard.

5) GE "PK" test switches aren't doing anyone any favors. #1 they require special knowledge of how the switch works and if someone isn't careful, they can insert the test paddle in creating an open CT. #2 Many times the testing part of the system is not ordered with the switchgear (or it walked away) and if someone is walking in cold to a situation they very likely won't have the test interface. Same goes for Alstom, etc. An FT test switch can be used without the paddle, although not recommended.

The switches should be applied to anything that needs to be isolated for commissioning, maintenance or repair / replace. I have also seen a number of designs where there are all kids of test switches, yet one is forced to run test leads into the switchgear cubicle, behind the panel, etc. Our utility designs allow a relay to be isolated safely with the test switches alone.

Many years ago, I was testing digital relays for gas turbine generators, both of the major US manufactures. Neither design had a single test switch and 5 relays per unit. At the time, due to Union rules, an electrician had to be on hand to lift the leads to the relay and re-land them when I was finished. Years later, I was called back by the owners (IPP's) to provide NERC required relay testing. No electrician this time, but each wired had to be lifted and landed / rechecked again. I've also been paid handsomely to write step by step procedures for others to do the same. I guarantee you, no cost savings were realized by these designs.

Whatever is done, keep it consistent. I've also seen designs where the AC test switch was as mentioned above for some, but not all relays. This just takes more time to sort things out in the field and also presents a Human Performance issue. Additionally, if you are going to use 2-3 test switches, get the ABB one (FT-19R?) that has the clear cover. It is designed so if something needs to be open for isolation, the cover can go back on and the live blades are protected.

Most panel builders / switchgear manufactures have the ability to make engraved plates that (preferably) sit right above the test switch and are labeled for each pole designation. As an example, somewhere on the top would be the Test Switch #, what device it is for and then the Va, Vb or Trip CB 123(+) and Trip CB 123 (-), BFI in Rly ABC, etc. Whenever I step out of the utility world, I'm likely to find metal clad Swgr that has 6 test switches and nothing labeled. It may take an hour or so in these situations to find the drawings, identify what is what for a wiring diagram or layout drawing and then connect the test set.

Not all substations, panels and situations are the same. Sometimes it's just not possible space wise to have 2-3 test switches per relay. In that case there are terminal blocks available that have test facilities. Standard test leads in the field generally have a banana plug (4mm) ends. I know Phoenix contact makes terminal blocks that allow for safe isolation / shorting as well as facilitate connection to test leads. Terminal Blocks. If things are cramped, consider an FT for the AC and Phoenix for everything else.