Try googling "motor bus transfer relay". Beckwith has a dedicated relay, SEL does that application in the SEL-451. We have a couple of the Beckwith relays but, in our system, they're isolated orphans and I'd use the SEL approach in the future simply because they're what we use for all new applications.
I will look at the instruction manual, but is there anything I should know ahead of time regarding the programming? Anything particular settings I should consider?
I don't know; I haven't had an opportunity to try the SEL solution.
A lot of your success or failure will relate to the inertia of your motor loads and the speed of your breakers; breakers with fast tripping speeds may not be as fast to close. The relative phasing of the two sources can also play into the results.
Nearly 10 years ago we installed the Beckwith relay at one of our combined cycle power plants. The normal source off the unit aux transformer lagged the 500kV by 60 degrees while the standby source from the local distribution coop lagged the 500kV by 30 degrees. This meant that the motor bus had to fall back 330 degrees to come into synch with standby source. We were having some critical motors dropping out in the 230-270 degree range and didn't gain a whole lot with the "fast" bus transfer scheme.
We then rewired both sides of the standby transformer to change its vector group from Dy1 to Dy3, placing it 30 degrees behind the unit aux source instead of 30 degrees ahead. The scheme then worked beautifully, taking about 45 degrees to recognized that the normal source had tripped and get the standby source connected. The 15 degrees out of phase was not excessive for the motors.
I have often wondered if we might have been better off going to 150 or 210 degrees between the two sources. With the change that was made, the "oh sh!t" transfers all work perfectly and the return to normal transfers can be scheduled, but may require turning off and restarting motors. This plant has commercial steam customers and an aux boiler so the motor mix is a bit different than most combined cycle plants.
With that 45 degree slip between normal breaker open and standby breaker close, I don't know it the relay would have gone for it if the two sources had started out in phase. If you can start in phase and catch the motor bus on the first time around you then have the luxury of the possibility of a closed transition transfer back to the normal source, or for planned transfers from normal to aux. But if you have to let the motor bus fall back a full 360 degrees there may not be many motors to catch. At that point you have to decide which is more important - closed transition transfer back or an assured catch on loss of normal. Unfortunately, you may not know how many degrees the bus falls back during the transfer time without testing, and for a plant that hasn't been built yet that can be just a bit tricky.
Awesome, I will keep this in mind- certainly the difference in phase angles for fast transfer- as that will come up in the future.
In my current application closed transition from one source to another is by far the dominating factor, so a time delay- even several cycles (or more)- will be tolerated as long as the transfer eventually occurs automatically and can latter happen in closed transition when the preferred feeder is brought back to service.
I am reading the manual and it looks like this relay can easily be configured to do source transfer (even the voltage inputs complement that), and if so I hit the jackpot.
