Generator Reverse Power Relay Testing

[ksprague]

The following is a new procedure that was given to me to perform the testing:

After unloading the generator let the 32 Relay (Reverse Power) arm itself; which is accomplished by sitting unloaded for 45 seconds and adjusting the power factor setting of .97 to .99 on the lag side, this will trip the Turbine/Generator. Additionally when performing this step you may need to break turbine vacuum to aid in getting the generator to show reverse current.

I have a couple of questions.

1. If I have to manipulate the generator by adjusting power factor and opening the turbine vacuum breaker, how reliable is the relay scheme? In and emergency if I had time to do both of these things I would have time to just take the Unit off.

2. How would opening the vacuum breaker make the generator motor? At zero MW, wouldn't the control valves open to accommodate for any "drag" that the backpressure would create? Also, wouldn't the turbine trip on high vacuum before the Reverse Power Relay would trip the Turbine/Generator?

 

[ScottyUK]

Reverse power relays fitted to steam turbines are very susceptible to the small phase errors introduced by the instrument transformers. The motoring power is very small for a turbine spinning in near vacuum conditions, and it is quite possible for a heavy reactive load to cause the relay to behave as though it is reading an active power value which is solely due to the reactive load and the phase shift. It can cause mis-operation of the scheme, the worst case being the machine doesn't trip in a reverse power condition because the generator is still on the bars and delivering or absorbing reactive power. If the procedure is relying on flexing excitation to produce an effect similar to a reverse power conditon then it's a poor procedure relying on the imperfections in the CTs and VTs to fool the relay.

Breaking vacuum to deliberately introduce windage losses onto the turbine in order to test the reverse power relay is possibly the silliest thing I've read today. Windage losses on a big LP turbine will quickly heat the blading to the point where damage occurs. Why deliberately risk the machine?

What's wrong with bench testing the relay?


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

If we learn from our mistakes I'm getting a great education!

 

[rasevskii]

for ksprague-

One way of testing the reverse power relay, is to reverse the polarity of the PT voltage to the relay, block the trip, and increase the KW of the turbine to the tripping point and see that it works. It should work at a few percent of rated power, let us say around 5% and a time lag of some seconds.

This assumes it is a single phase connection for voltage, such as the old ASEA type RXPE or similar.

Make sure a qualified person is there to do that.

For a digital integrated generator protection, this would not work as usually there would be a three phase voltage connection for all the functions of the relay, which must not be disturbed.

regards, rasevskii

 

[cranky108]

I believe it is suggested that tripping on the reverse power relay is a good test of the relay, and the settings.
The problem is on steam turbines the reverse power setting is very small, and the consultant that set the relay dosen't always know if it will work. But the consultant has collected his money and moved on.
You probally need a review by the plant electrical engineer, if you have one. However, most of the plant electrical engineers that I know, generally don't touch the turbine relays, as they just haven't worked with that many of them.

Bottom line is a relay setting review every decade isen't a bad idea. And tripping on the reverse power relay tests the settings, the relay, and part of the control scheme.
(Probally more important if you are in the NERC footprint).

 

[dpc]

Actually, some units routinely use the reverse power relay to take a generator off-line in order to verify the steam valves are truly closed.

Normally the problem is the thing tripping when you don't want it to, like during start up.

I've never heard of deliberately breaking vacuum to initiate a reverse power test, but maybe it something unique to this particular unit.

As others have mentioned, the reverse power relay has a tough job to do because of the low power levels in relation to the CT ratio and measurement errors. I don't see any problem with doing a test as long as it is not putting the turbine at risk.

David Castor

http://www.cvoes.com

 

[ScottyUK]

I'm not suggesting that testing the relay is a bad idea per se, but having to adjust the excitation until it trips indicates that the relay can't do its job, or at least it can't do its job *reliably*. What use is the relay if a reverse power condition occurs when the excitation is in a part of the range where it inhibits tripping? The whole scheme needs looking at - better CTs, a lower burden relay, tighter settings, all have potential to improve the behaviour of this system.

Unreliable protection is as bad as no protection, maybe even worse because there's an expectation that it will perform its function.


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

If we learn from our mistakes I'm getting a great education!