Synchronising at Transmission Level

[mgtrp]

Hi all,

I was curious as to what various utilities were doing as standard practice for synchronising in transmission systems.

At our older substations, we would typically have a single synchroniser with relays switching in the appropriate VT's and close coils. SCADA would issue a brief pulse, latching in the synchronising circuit for a predetermined time, with the circuit being interrupted when the circuit breaker closed, or if a trip command was issued from any source.

In our new substations, we are getting designs from consultants that require a latched close command from the SCADA system, and simply have a permissive in one of the protection relays which will allow the close command through once the two systems are near enough to synchronism. The SCADA command is latched for several minutes, and isn't unlatched by anything other than the timer.

What are people's views on the use of a single device for synchronising? (one blown VT fuse is all it would take for a disaster "dead-bus" close, for example). In generating stations (where I have more experience), the standard we would apply is minimum two devices for synchronising, taken from different phases. This would generally be either the auto-synchroniser OR the operator (when manual closing), in conjunction with suitable sync check relays.

Any thoughts on the latched close signal from SCADA? I don't like the idea of a signal remaining latched in, and I especially don't like the idea of it not being interrupted by a trip command, other than by the CB anti-pump circuit.

 

[bacon4life]

Only unlatching by a timer after several minutes would leave me nervous as well. At a minimum, a close should unlatch the timer independent of the anti-pump circuit. At our generation stations we use lockout contacts to interrupt the close circuit if a trip is received, but it does not directly unlatch the synchronizer relay. The GCB also only closes for live line-live line conditions, so only 1 phase with two devices is used.

We do not sync at transmission sites, although for reclosing we do sometimes have a "sync check" or a live line/dead bus check. In our case it is really a wide allowable angle but zero frequency error allowable. In your case you may need this type of setting to allow normal closing, and a second settings group that has very narrow allowable angle and a reasonable slip frequency for actually synchronizing two systems. Our setup is mostly a backup to the transfer trip, so our risk of a dead bus with a blown fuse is an N-2 contingency.

If the sync check were only being implemented for Aurora prevention per the link, having it self contained seems reasonable. The NERC website appears to have more information about this as well, but is restricted access.

http://www.selinc.com/WorkArea/DownloadAsset.aspx?id=8504

Perhaps you need two separate SCADA commands, one for a normal close without a latching timer, and a second one with a timer. It seems like synchronizing two large islands via telephone to multiple plants could take a while and would be complicated by a timer with too short of a duration.

 

[mgtrp]

We require transmission level synchronising for a variety of situations, generally to do with maintenance or line outages between generation areas, when black starting the grid (this happens more often that you might expect around here...), etc. Certainly not for "Aurora Attacks", which I'd never heard of. To be honest, I'd thought when I first read your post that you meant disconnecting parts of the system during periods of increased solar activity and resulting GIC, which is something we theoretically may need to do, but haven't done yet.

We have the provision for closing onto a dead bus or dead line, as well as the narrow angle synch-check for live line/live bus synchronising, done through either of the two systems described in my first post.

The synchronising can take several minutes - we are basically just waiting for the two systems to drift into something close to synchronism before the synchroniser issues the close. We are not trying to issue a close from SCADA at just the right instant over communications channels! (hence the latched signal).

I've also been starting to think about using a relay input in parallel with the trip coil to detect trip commands from all possible trip devices (and therefore interrupt the sych sequence), but had thought that this may cause issues with the trip circuit monitoring input to the relay.

 

[slavag]

Hi.
Why latched SCADA signal, from SCADA you need only send pulse "Close command" to
control device of CB with synchrocheck functionality.
Control device ( see control terminal or control and protective terminal) is included all needed infomation about CB ( statuses of objects CB/DS/ES and MCB VT/PT indication, trips, some another monitorings).
Such terminals is included a tested synchrochek functinality with lot of options, for example , for connection two asynchronius systems with autorecloser and with command mode and timer.

 

[mgtrp]

The "close" signal is not an input to a synchronising relay, it is passed through an output in a protection relay which is which is closed only when the relay detects synchronism. This is fairly standard in designs that we receive from consultants for our newer sites. It means that, until we have a suitable outage to modify the system at each site, we either have to sync manually on site or latch the close from SCADA until the synch-check relay permissive allows the signal to the close coil.

Which is one reason why I was curious as to how other utilities were performing this synchronising?