Network protector question

[Chrisflhtc]

I have a 4500A network protector Richards Type 147NP that is lightly loaded (270A) it has opened several times in the past few weeks. This feeds a few large UPS' for one of our data centers it opened each time during a period of lower than normal load.The current dropped to around 135A and held there for almost an hour or so then tripped open. There were no anomalies in any of the scada readings. Is there a relay that looks for too low of a load?
TIA
Chris

 

[infinity]

Is this a brand new installation?

 

[GoldDigger]

I have a 4500A network protector Richards Type 147NP that is lightly loaded (270A) it has opened several times in the past few weeks. This feeds a few large UPS' for one of our data centers it opened each time during a period of lower than normal load.The current dropped to around 135A and held there for almost an hour or so then tripped open. There were no anomalies in any of the scada readings. Is there a relay that looks for too low of a load?
TIA
Chris

The function of a network protector is to prevent power from flowing in a reverse direction through it. It will open when reverse power flow occurs and will not close under conditions where reverse power flow would happen as soon as the contacts close. It identifies reverse power from a closed condition by comparing the phase angle of the current signal from the Current Transformers to the phase angle of the voltage on the lines. It identifies the potential for reverse power in an open condition by comparing the phase angles of the voltages on either side of the open contacts and the amplitude of both voltages.

If you have something on the load side which is capable of putting energy back into the system for a short period of time, and the "normal" load current is too small, the burst of output power can drive the current/power through the protector negative for long enough to cause it to trip. I think that this condition on just one phase will be enough to trip it.

Some of the things which could supply energy into the circuit from the load side are:

1. PV grid tied inverters.
2. Large motors forced by their loads to rotate faster than synchronous speed.
3. VFDs which have Active Front End / Regenerative breaking circuitry, when the controlled motor is shut down and braked.
4. Possibly large UPSs under certain strange conditions. No idea about that, but if they are the main loads, I would have to suspect them.

Your steps to mitigate this, once you identify the source of the reverse power are to either increase the static load to more than the reverse power or to reduce the reverse power available from whatever source you have identified.

 

[iceworm]

I have a 4500A network protector Richards Type 147NP ... ... Is there a relay that looks for too low of a load?

I am always a fan of geting the book on the device - and seeing what that describes

ice

 

[GoldDigger]

I am always a fan of geting the book on the device - and seeing what that describes

ice

I looked at the manual, and that did not help at all. It assumes that you already know what a Network Protector does. Next help from Google was this though. Much better. From there just an application of logic.
Note carefully that the 4500A rating does not meant that it will open under 4500A of overcurrent in the normal direction. It is not an OCPD, it is a power direction detector.

I suppose that it may not be able to tell the difference between 130A in one direction and 1A in the other given that the full load range is 4500. But none of the available documentation seems to address that. It does not appear to be designed to detect (and open on) low power flow, just on reverse, but there may be tolerances involved.

 

[SG-1]

It may be that the relay simply needs to have a setting adjusted, the reverse trip setting. Richards may offer an accessory called a "phantom load" or something simular that might help mitigate this issue. Westinghouse called theirs a phantom load. I am not familiar with the Richards product, I did work in the Westinghouse network protector product line. It was the companys oldest product line, started in the late 1800s.

Normally, there would be two or more network protectors feeding the same load. You are using a network protector on a radial line ?

 

[robbietan]

The function of a network protector is to prevent power from flowing in a reverse direction through it. It will open when reverse power flow occurs and will not close under conditions where reverse power flow would happen as soon as the contacts close. It identifies reverse power from a closed condition by comparing the phase angle of the current signal from the Current Transformers to the phase angle of the voltage on the lines. It identifies the potential for reverse power in an open condition by comparing the phase angles of the voltages on either side of the open contacts and the amplitude of both voltages.

If you have something on the load side which is capable of putting energy back into the system for a short period of time, and the "normal" load current is too small, the burst of output power can drive the current/power through the protector negative for long enough to cause it to trip. I think that this condition on just one phase will be enough to trip it.

Some of the things which could supply energy into the circuit from the load side are:

1. PV grid tied inverters.
2. Large motors forced by their loads to rotate faster than synchronous speed.
3. VFDs which have Active Front End / Regenerative breaking circuitry, when the controlled motor is shut down and braked.
4. Possibly large UPSs under certain strange conditions. No idea about that, but if they are the main loads, I would have to suspect them.

Your steps to mitigate this, once you identify the source of the reverse power are to either increase the static load to more than the reverse power or to reduce the reverse power available from whatever source you have identified.

may I add

5. Capacitors that put the power factor to "leading"

 

[GoldDigger]

may I add

5. Capacitors that put the power factor to "leading"

But even with leading power factor instead of lagging power factor, the phase angle should still be on the correct side of the +90--90 degree line.
However, that does bring up the possible inability of the protector to deal with very high harmonic distortion of the kind that may be present if the UPS are feeding mostly power supplies in the data center. Each UPS may include power factor correction for its own charging circuitry but not compensate out load non-linearities.
Depends on whether the UPSs are constantly online dual conversion type or just transfer switch type.

 

[Chrisflhtc]

Thanks for all the replies this protector is part of a three spot that is getting up graded in stages, the tie breaker in the system is currently open till the other board and transformers are replaced. We have a Richards rep coming in today I was just looking for some real world experiance thanks.I will post what ever I learn.
Chris

 

[iceworm]

I looked at the manual, and that did not help at all. ....

Bummer. Looks like one needs the manual on the relays.

ice

 

[Chrisflhtc]

The issue has been solved the we just received this report:
1.

There is high confidence that the power interruption issues that we have had over the last few month were caused by the Eaton UPS during it?s scheduled battery load test.


[FONT=&quot]During this battery load test, The UPS generates an inverse load of 50% capacity pushed on the batteries to order to exercise them; and at 60minutes, it generates an inverse load of 100% capacity. Normally (in theory) , any extra load was then pushed back to the Utility i.e. the power circuits providing power of the UPS currently..Feed 35 and 55. They were able to handle/absorb the inverse power from the UPS which would explain why there were no power outage today; When we had a single power feed (Feed 45) in Feb and March 2013, it was not able to handle the inverse power feedback from the UPS causing the power outage[/FONT].
This feature has been disabled for the time being .
Chris

 

[GoldDigger]

This feature has been disabled for the time being .
Chris

Comforting to hear that everything was individually working as designed. Those pesky interactions keep coming up though. :lol:

 

[controlled]

Bummer. Looks like one needs the manual on the relays.

ice

Thats helpful

 

[iceworm]

Thats helpful

The comment was meant to be consoling - not helpful. Without the factory IOM manual/service literature, one is just guessing - not something I care to do. Unless, I get hired to do a science fair project on obsolete equipment - in which case I am ON-BOARD. Being a capitalist, I am in favor of "fee for service".

Now that we covered that, did you have a point?

ice