Clock Losing Time....

The master clock is in one location. All generators in an area are frequency locked by virtue of them all being connected to the same grid. The error signal between "real time" and "clock time", which is part of something called ACE (area control error) gets turned into raise and lower commands for individual generators. When a generator has its power set point increased, it raises the system frequency microscopically, which then eventually minimizes the accumulated time error (assuming you were running slow).

At each generator, it's actually easy. When you bring a synchronous generator online to a big bus, you first adjust the turbine steam valve to spin it up to match the bus frequency, or a little high, then adjust the field to match the bus voltage. At the instant the phases match, close the breaker. If you do it right, the breaker stays closed!

Then open the turbine steam throttle more to get the power transfer that you want, and simultaneously adjust the generator field to tweak the power factor.

You can do all that by hand; I've done it bringing up a turbogenerator on a ship, and it's easy once somebody shows you how.

If there's no motive drive available (no steam) all you can do (after you somehow spin it up) is tweak the field current. The generator free-wheels on the bus and becomes a "rotating capacitor", handy for system PF correction. Makes sense from a COE standpoint: no mechanical power in/out means that the machine must look mostly reactive to the bus.

The big problem is to trim the entire grid to be on-time longterm, which must be a cooperative effort. I don't know how that is done.

Why does that seem so impossible. I think that is a lot easier than a flock of birds turning in unison or a school of fish swimming in lock step.

Think about that. If every station is following the grid, then the grid will vary all over the map. There has to be a way to set each station from the reference, not the grid.

Each station will monitory the grid to determine how well they are matching it and I expect they will accommodate short term variations. But the station has to exert some influence on the grid, meaning it will have some residual delta in phase as the grid varies from the reference.

TL;DR - I can talk long enough to put any listener to sleep even if I don't really understand the things I'm talking about.

a synchronous motor works, and the only difference between a generator and a synchronous motor is the direction of power flow.

given a stroboscope, or other phase compoarison method, I can't see it being any harder than merging with freeway traffic,

Do you have a point?

The rules of physics will lock all generators in sync (barring extreme events). The total power injected to the grid, minus the load being pulled out, determines how quickly the grid will accelerate or decelerate. Each generator has a governor (regulator) which has a setting called "droop". As the system frequency goes up, the governor will reduce the generator's power output, and vice versa. This could be considered a P term in a PID controller. The I term comes from comparing the "master clock" (WWVB or GPS) to the one that has been running from the line frequency (note that time is the integral of frequency). When either the frequency or time error exceeds a dead band, a central facility sends out raise or lower pulses to certain generators (those on AGC, or automatic generation control), and those units gradually push the system frequency up or down to slowly null out the time or frequency error.

Are you saying this is how the power system works or how it could work? I'd like to see a reference as I have not been able to find one.

BTW, you didn't explain what "droop" is or how it is used.

This IS how it works. Droop is a gain setting for how much the power set point varies as the frequency varies. Typical droop settings are around

For references, Google "generator droop control", "AGC generator" or "area control error".

Check this out...

Nice. That gives a good overview of all the major AC interconnections and also points out (by omission) the non-interconnected areas.

Okay, thanks for the antiquated technology reminiscence but the really big stuff is now required by law to feed into these super high voltage DC transmission corridors to be distributed, electronically chopped and stepped down for local distribution.

If it wasn't for all that "antiquated technology", you'd be freezing in the dark. Don't take infrastructure lightly.

Most individual generators still work just like I described, connected and disconnected to an AC bus to other generators. They mostly have automatic controls, and if you are competent to click a mouse and have a controller do all the understanding for you, enjoy.

I think you need to use transparent shipping tape.

If you use duct tape you can't read the display when you are setting the oven controls.

Jeff

Maybe that is what led to the one I have that is programmed for DST. I have another, and older, one that is just like it but sets itself from the time code. It has a bug so that on the evening of the DST change it moves back 1 hour then corrects itself over night. I figured that whoever designed it took the quick fix way and just programmed it in.

Bill

Ok, everyone seems to think it is a breeze to "get" almost 19,023 AC generators in phase. I think Larkin may have the answer in his response, I don't have a full understanding, but it sounds like if you get them close, they will sync up on their own. I think, my mistake was in thinking the rotation speed depended on the input drive to the generator and the load. It seems the the input drive is not so important to proper phase, although very important to generating power output. Thus you don't really "get" them in phase, they get themselves in phase because of the field input. Am I close? Mikek