maintenance requirement for digital relays?

[3rdkill]

Hi,

Im doing a small project regarding the maintenance procedure for digital relays for our company, at the moment as per our summary of tasks for digital relays we only test the inputs and outputs cause these relays have the ability for self checks.

Im suggesting for a revision of our summary of task so we can include the testing of the full scheme including the logic behind the operation of that contact.

I need some help from you guys to convince them to include this on our summary of tasks as i have to defend this in front of our bosses... any input would be greatly appreciated...

P.S.
We have encountered at least two relay fails that the self checks didn't pick up for the past 3 years thats why i've decided to have a punt on this one...

 

[cranky108]

What kind of failures? What type of relays?

 

[rbulsara]

Only way to make sure that relay AND the wiring works right is to perform functional tests, under controlled conditions. This may or may not be possible in existing systems.

The bench test and self-test are a must but they do not check the external wiring to breaker controls.

Rafiq Bulsara

http://www.srengineersct.com

 

[cranky108]

In the documents (draft) for NERC PRC-005-2, there is discussion that you know what the self test actually tests. But most relay manufactures presently don't give this information.
Therefore we should ask whenever there is a clame the self test dosen't test everything, what type of relay.

I would expect the inputs, and outputs would not be tested. A meter test, and pulseing the outputs, is a big importance. Also descrete inputs must be tested.
But beyond that, what isen't self tested?

 

[Zogzog]

ANSI/NETA has detailed test procedures for microprosser based relays but the procedures depend on the relay type. If you want some ammo to present the ANSI/NETA specs are your best bet.

 

[NorthSlope]

What were the relay failures that the self check did not catch?

Solid state relays generally work perfectly if they are set up properly to begin with and nobody "lets the smoke out".

I've been setting and testing them for going on 15 years now and have never seen one that passed initial test fail in service.

 

[dpc]

This has been discussed not too long ago in this forum. You might try doing a search.

I've been setting and testing them for going on 15 years now and have never seen one that passed initial test fail in service.

They are reliable, but our luck has not been quite that good. We have seen a few fail.

 

[oldfieldguy]

My experience with microprocessor relays has been good, but no perfect. That company that makes the blue ones popular with many utility companies found out several years into the game that one of its first offerings reacted poorly on distance protection on short lines.

First off, let's assume you did a very thorough job of checking exterior devices for proper function and you also checked every bit of logic you've programmed, and all that has been working properly. Now you're in service, and problems still come along.

We had relays just quietly die. That got us to start assigning a "common trouble" alarm into the SCADA where heretofore we'd just gone on breaker trip status. We didn't have "critical failure" outputs on the old electromechanical relays. With the new relays you can have several conditions to monitor, some by direct communication to the relay, such as loss of remote device communications, etc., AND you have a set of contacts that tell you that the relay has failed, up to and including loss of power. That contact needs to go to the SCADA and thence to an operator.

I've had more than one case where a test technician inadvertently loaded a different setpoint file into a relay than the one intended.

And as others have pointed out, problems with exterior devices and wiring need to be checked periodically, too.

old field guy

 

[3rdkill]

thanks to the reply guys, the failure we've experience last time is with regards to a SEL 421 distance impedance relay, there's a fault in the line and the relay respond to it correctly but the output contact didn't close. at the moment like cranky 108 is saying to test the outputs were just pulsing it and check the external wiring if its correct, the problem is for micom/areva relays, im not sure if there's a way we can force it's output like what we're doing on SEL and GE enervista relays, have anyone done that before?

 

[3rdkill]

have found the way to pulse outputs for micom and areva relays, have to dig on the manual under commissioning tests...