Re: electrical power

["John S. Denker" <jsd@MONMOUTH.COM>]

Rick Tarara wrote:
> How is the voltage regulated with changing loads?

Multiple schemes reinforcing each other:

1) To maintain roughly-constant voltage in the face
of local, short-term changes in load, they rely
on _saturable transformers_.  Think about the
sigmoid shape of the saturation curve (!!without!!
hysteresis, just saturation).  Drive it on the
X-axis hard enough to get into the flat regions.
Then if the driving voltage (on the primary)
changes a little bit, the voltage on the secondary
basically doesn't.  Of course this means you're
putting out something that doesn't even vaguely
resemble a sine wave, but that's OK.

2) Over longer timescales, and/or to respond to
larger changes in load, they can throw switches
at the local substation.

3) At some point, all this is reflected as a changing
power demand at the terminals of the generating
station.  This is noticed by the feedback-type
regulator mechanisms which adjust the throttles
on whatever is driving the generators.

4) If necessary, they can bring additional
generators online, and/or buy power from neighboring
districts.

> In my understanding, turning up the power really means bringing more
> generators on line

See above.

> and if the demand is less than the capacity of the
> running generators, the excess is dumped back to ground.

I have no idea what that means, but I'm pretty
sure it's not right.  Excess what?  Excess energy?
There's no way they're going to dump energy.  They
just throttle back on the prime mover, and leave the
energy in the fuel tank.

> Another Web site talked about DC transmission being more efficient than
> AC--but I think they were talking about at the same voltage.

OK.  At any particular peak voltage, a sinusoidal waveform
carries only half the power.  Square waves wouldn't have
this problem, but might have other problems.

AC has the great advantage of having zero-crossings in
the current, which is a really nifty opportunity for
opening switches.  Switching high-voltage high-power
DC is nasty business.

> The particular Canadian DC line was running at 900K!

For any particular power being carried, higher voltage
means lower current meaning lower I^2R losses in the
cable.  Or smaller cable for the same loss.  Copper is
not cheap.

> how [do] the various power
> plants on a grid synchonize frequency and phase as their output is merged?

How do you shift from 3rd to 4th in a car?
If you're clever, you can manipulate the
throttle and/or the timing so that the RPMs
of the front of the clutch match the RPMs
of the back of the clutch, so you can engage
the clutch with no jerk.  If you aren't quite
perfect, there's a small jerk but it isn't
tragic.

Getting the phase right is just more of
the same.  Get the frequency real close and
wait until the phases line up, then flip
the switch.