low votlage ground fault testing 1

[alehman]

One of my clients recently made some modifications to the ground fault detection system on a low voltage switchboard. We have ground fault on the main breaker and feeder breakers, with zone selective interlocking. These are draw-out mounted breakers with internal residual ground fault sensing. There are cell position contacts and breaker auxillary contacts wired in the neutral CT circuits. The wiring of these was changed when the bus was re-configured.

I have recommended primary injection testing of the revised ground fault system. Due to the complexity of the system, tseting will take a number of hours. My client insists that the switchboard cannot be shut down for sufficient time to perform the tests. Does anyone have ideas on alternative methods to test the neutral CT circuits without shut down?

 

[busbar]

If the client does not already have a philosophy of applying current-injection testing—associated with the larger issue of periodic power-circuit-breaker maintenance—it may be a hard sell now.

A “no outage” insistence may mean, 1) business is good and he doesn’t want to slow production, or 2) scheme testing is a waste of money and he’s sure it’ll be fine. It is likely the client has not experienced a significant unscheduled outage that had a serious impact on production reports. Adding external {neutral} CTs to power-circuit-breaker lineups is not my idea of ensuring flawless protective-scheme operation.

Unscheduled testing can be tedious if one of two scenarios prevail. Besides the pressure to resolve possible problems when the facility is dark, someone has to intelligently determinate what problem external to the switchgear and its communication scheme has surfaced? If nothing can be found, who decides to go ahead with the “burn-and-learn” approach? [In this case, you are the default expert {whipping boy} for almost any undesirable outcome.]

Do your best to make the client understand that the no-test decision is risky, and may cost at least an order of magnitude more if “debugging” is required in conjunction with an unscheduled outage. Too often managers have to learn from their poor choices, although it [realization of a hard lesson] may not happen the first time around.

 

[alehman]

Busbar - thanks for the excellent response. My client is using excuse #1. They are concerned that connecting and disconnecting numerous temporary power feeders may pose more risk to their facility than not testing the equipment. They've already had one major "burn", and seemed for a while to have learned. But the memory is fading quickly, while the lawyers (including mine) are still sorting it out.

The neutral CT's are part of the original ground fault detection system. More were added when the system was expanded. I think the design is sound. My client and I have little confidence in the installer however.

 

[bass1mj]

It sounds as if you are using M&G Masterpacts. I know these breakers have an external test facility.(Plug in unit). I don't know if it will help if you disconnect the shunt trip on the breaker for the test (Dangerous when in cct). Maybe an investigation into the breaker workings can help.

I would have preferred to do injection testing.

 

[alehman]

The breakers are Cutler-Hammer DS-II, but most LV breakers use similar GF systems. We have tested the individual breakers (out of the switchboard) with primary injection.

My concern is that there are breaker aux contacts and cell position switches involved in the neutral sensor circuits and ZSI circuits. The wiring is somewhat complex and has not been tested.

Would it be possible to do secondary injection at the neutral sensors while they are in the circuit and live? At present, there are no loads connected to the neutral bus. If this can be done without damage to the sensors or trip units, we could at least verify portions of the circuits.