Electrical engineers in the audience?

[C_eh_N_eh_D_eh]

I'm in Ottawa, and one of our radio talking heads was going on about why power is so slow to come back up. He said that the engineers need to make sure that all the AC power grids "pulse" (his word) perfectly in phase. I know a bit about electical engineering, so to me this sounds just plausible enough to be true. Is it?

And don't worry about us here in Ottawa. We've weathered storms like this before.

[Atlant]

1. Yes.

Yes, that's actually just about right.

(I don't know what your level of understanding is here, so
please forgive me if I "undershoot" you; I'll try to explain
for everybody who might read this.)

While electricity appears to be continuous, in North America,
it's actually switching off and on 120 times per second. In fact,
the flow of electrons actually reverses (alternates) 120 times a
second! So this gives us a total of 60 "Cycles" of power per
second, known by the honorary name of Hertz (Hz). North
American power is "60 Hz" power.

     (In contrast, and for peculiar historical reasons, most
    of the rest of the world uses "50 Hz" power, where the
    flow of power reverses 100 times per second.)

The exact timing of these reversals of power flow is critical;
If a new generator (really, an "alternator") tries to connect to
the grid but the timing of its cycles isn't exactly matched to
the timing of the power already flowing on the grid, really
bad things happen as the alternator connects to the grid.
Really dramatically bad things happen, not unlike the sparks
and explosions in ancient sci-fi movies.

So the engineers bringing the grid back up have to precisely
time the arrival of every new alternator on the grid, making
sure that the "phase" of its alternations exactly matches the
phase of the alternations already present on the grid. The timing
has to be much closer than 1/120th of a second. And this has to
be done with each and every arriving new alternator.

In addition, the overall power flow has to be managed. One
moment, you've got a big building full of steam boilers just
sort of idling along, spinning the alternator at just the
right speed but not really doing much work. A moment later,
that alternator's "on the grid" and might be struggling to
power up all of New York State! There are a million details
to be managed to make sure that doesn't occur.

---

One of the more interesting question is "How does it all get
started? Who goes first (onto the grid)?"

In my own state of New Hampshire, it turns out that, after
a total and complete blackout, the first power that is generated
comes from a little Caterpillar diesel generator at the hydro
dam up in Manchester, NH. That dam is a "black start" plant; it
can come up with no external power, depending solely on this
little Cat. After the Cat diesel starts, they fire up the control
systems of the hydro dam. Then the dam starts making real power.
Then, using that power, PSNH can start the three fossil-fueled
power plants in the state. Then they can re-start the nuke at
Seabrook station.

I asked one of the engineers "What if the 12 volt battery on the
Cat diesel is dead?" He pointed and said to me "You see that
pickup truck over there?" :-)

Atlant

[Scottie72]

7. Yes it is correct,

and I have experienced it first hand. That is a very good analysis. A few weeks ago, our electrical power at the lab went "went out of phase". The results of this caused our lab to "down" for the entire weekend.

[rock]

2. Yes, that's correct

If the power were DC, phase wouldn't be a problem but polarity would (which is simpler to handle).

[Atlant]

3. Voltage, too.

> If the power were DC, phase wouldn't be a problem but polarity would
> (which is simpler to handle).

If the power were DC, voltage (aside from the absolute value
of its polarity) would be a factor as it would primarily be
the voltage that would determine the power flow.

On the AC circuits, it's the instantaneous voltage that determines
the power flow, so it's the vector quantity of the RMS voltage
combined with the phase. And, of course, over typical transmission
distances, the speed of light comes into play to complicate things.
The nation is, in fact, more than one 60Hz cycle wide!

Atlant

[ender]

4. if power were DC, then...

we'd have power generating stations every couple of blocks in a city.

[Atlant]

5. No, that's a 19th century answer.

> if power were DC, then we'd have power generating stations
> every couple of blocks in a city

No, that's a 19th century answer. The problem with DC was
that, in the old days, you couldn't "transform" it from low
voltage/high current to high voltage/low current for long
distance transmission. And that's why you would have needed
a power station on every other block; otherwise, the IR=V
voltage drop would have killed the economics.

Enter the thyristor (SCR), and more importantly, the Insulated
Gate Bipolar Transistor (IGBT).

Now you can build amazing low-loss devices that "transform"
dc to dc.

And Voila! Suddenly, you can put up a MEGAVOLT
(Ultra High Voltage) dc transmission line from (for example)
James Bay, Quebec to Ayer, MA.

But wait, there's more!

Because it's dc (instead of three phase ac), you only need
two wires (the source and the return). One runs at +500,000
volts and the other runs at -500,000 volts. And if one wire
fails, you can run on just wire at half-power, sending
the return current through the earth.

And that's not all!

Because it's dc, there's no issue with "phase". The source
and destination ac grids can be running at completely
different phase relationships or even different frequencies!
And the solid-state power coverters can even control just
how much power they'll push through the circuit, which
gives you FAR MORE FLEXIBLE control than with
ac.

Overall, the control logic is vastly simplified while the
network stability is vastly increased.

DC is the wave of the future, and it was exactly
this sort of "conversion to UHVDC tie lines" that
the Republican-rejected power bill was going to
address.

Atlant

[Atlant]

6. Google this for more info on HVDC transmission...

http://www.google.com/search?q=HVDC+Transmission&ie=UTF-8&oe=UTF-8