Hi,

Ok, quick intro to the frequency steering.

There still remains rules that say that network frequency should be on
average 60 Hz on the US grid. (Yes, there is proposals to remove it, but
it is still effective.)

Since the generation (let's talk traditional here not to make things
more complex than they need to be for the first overview) is from
generators, essentially big rotating lumps of iron, the balance between
load and generation causes the frequency change. If you have more load
than generation, the frequency will lower while if you have more
generation than load the frequency will go up. Essentially, if you
undergenerate, you would need the rotating energy of the lumps of load
to deliver, but that reduces their speed and if you underconsume the
energy goes into the rotation of the lumps.

Now, by monitoring the frequency you can steer the balance, askning
hydropower to increase or decrease production to balance the shift of
load. The operators have a fair clue on how the day will proceed as
people wake up, industry starts, workday, industry closes down, people
get home etc, so there is a basic pattern there to give a clue, but they
monitor it and balance it.

By also balance the phase, you can know how much you lag behind and
needs to run up by running the frequency high. This require spending
energy by increasing production compared to the load. Now, by being
smart you do that when you have low load, so that you don't have to
spend as much energy to achieve it, but never the less.

Then you have to manage your reactive energy, the VAr, which is a
different matter.

Breakers have several form of catastrophic protections in them, among
those if the frequency goes bad. Turns out that the frequency monitoring
of breakers gives so diverse readings such that for post mortem
analysis, they provide bogus values. They learned this the hard way
after the North-Eastern Blackout. When they threw out all the
traditional frequency readings, the PMU data that remained painted a
consistent picture.

The detailed monitoring of PMU gives much more data, also illustrates
forced oscillation, inter-area-oscillations etc. which makes the phase
wobble in interesting ways, and when analyzed gives good clues about
problems in the network.

An even more "fun" scenario is when the network runs into islanding,
since the link between areas is to weak to keep the frequency at the
same rate, i.e. the link is to weak to support the load, so one part has
overload and goes down in frequency while the other have overproduction
and gos high in frequency, which you can see by the way that phase
starts to deviate between the networks, and that before you have the de
facto islanding.

The islanding illustrates the need of the links to be strong enough so
that generators synchronize, or should we say syntonize to be correct
with terms, that is, they have the same rate.

The four islands that you identified do their own independent frequency
steering, but they exchange power. The generation-load thing still
happens, but phase/frequency decoupled. HVDC cables achieve the same thing.

Anyway, phase monitoring has become a very good tool for so many of
these measurements, and that requires a common "reference" phase and
that is GPS. That helps to monitor the phase and frequency of the grid
so that it can be controlled.

A peculiarity of the field is the ROCOF - Rate Of Change OF Frequency.
This is what we call linear frequency drift. Looking on those numbers
give you a good hint where you are going.

Until recently, photoelectric would not provide any of the rotating iron
properties, but the increase popularity of it now requires it to start
to have such properties for the stability of the system.

Cheers,
Magnus

On 04/04/2017 11:28 PM, Thomas D. Erb wrote:

Think of it as an ocean liner trying to keep a dead straight course to
it's destination.
It weighs many tons and wind and waves may drive it off it's path but
the captain
can correct for this.  It eventually arrives at it's destination and is
only a few feet
from the dock.

The total rotating mass of all the generators in a network is many times
the mass of an
ocean liner.  The operators do their best to keep them running at the
correct frequency.
Unexpected load changes can cause some divergence, but over time the
average is
dead on.

When I installed power plants in the 1970's they has a special "clock"
that showed the
cumulative error in terms of clock time.  The clock had two inputs, one
from the utility
power and the other from some reference, possibly WWV.  Normally the
"clock" was
pointing up at zero and not moving.

If the generator ran a little too fast the clock would move forward.
As the operator
observed the clock moving away from zero he would reduce the plant's
power and the
clock would move backward toward zero.  His goal was to keep the clock
at zero and
not moving.  Thus, your bedside clock was always on time even if there
were temporary
excursions fast or slow.

Pete.

On 4/4/2017 5:28 PM, Thomas D. Erb wrote:

On 4/4/17 2:28 PM, Thomas D. Erb wrote:

The line frequency is adjusted, for the most part, by adjusting the
prime power (steam valves, dam penstocks, etc.) on the generators at
power stations. That changes the speed, slightly, although as generator
1 of N starts to get ahead, the electrical load increases, and it slows
down.

It's actually a pretty complex system, since there are a whole raft of
"spring constants" in between the multiple generators in a system,
there's phase shifts due to transmission line inductance and capacitance.

"Stabilizing" a system in the face of changing demand is a non-trivial task.

I have seen some proposals to require VAR capability in photoelectric
installations but how feasible is that?  I cannot imagine utility
customers being pleased with having to pay extra for such a nebulous
to them capability.

I could see the utility companies pushing it as a requirement in lieu
of installing banks of synchronous condensers.  Maybe this could be
integrated with net metering so that users get paid for providing VAR
correction and pay for VARs but I bet it would be politically
infeasible.

Thinking about the stability problems associated with distributed
generation gives me a headache.  I am inclined to believe that any
solution which relies on central management or timing is an invitation
to major failure.

On Wed, 5 Apr 2017 09:52:32 +0200, you wrote:

The response time in a large plant is very slow.  Large steam plants
running at steady
state are running with their steam valves wide open.  A partially
closed valve is an energy
loss and is only used when changes occur.

The power control for a plant running at a steady load is the amount of
fuel thrown into
the boiler.  When you want more power you shoot more gas, oil, or coal
into the boiler.
For a nuke you pull the control rods.  Behind all of this is a lot of
thermal mass.  Things
don't change quickly.

Pete.

On 4/5/2017 9:01 AM, jimlux wrote:

I would guess the tightest control loop is on the generator stator field
windings, with mechanical control being secondary. Definitely a lot of
poles and zeros to worry about.

On Wed, Apr 5, 2017 at 9:01 AM, jimlux jimlux@earthlink.net wrote:

The rotary generators in a system of connected generators are
synchronous
machines. There is no frequency difference between them, only phase
angle,
and not much of that - if the system is stable.

The ocean liner analogy is correct, as there is only one captain
directing
the ship's course. If each plant set its own power levels it would be
very
difficult to maintain stability, due to the springiness of long
transmission
lines.

A set of connected generators is controlled by regional dispatchers, who
tell their plants how much power to generate in order for the day to
average
out to 60.000 cycles per second. They count cycles instead of measuring
the
frequency. You can count cycles with a synchronous clock.

This becomes less tidy when DC tie lines are used, because inverters
have
to be adjusted to get the correct power flow.

Hope I got most of that right.

Bill Hawkins

-----Original Message-----
From: time-nuts [mailto:time-nuts-bounces@febo.com] On Behalf Of Peter
Reilley
Sent: Wednesday, April 05, 2017 7:42 AM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Line Frequeny Stablity

Think of it as an ocean liner trying to keep a dead straight course to
it's destination.
It weighs many tons and wind and waves may drive it off it's path but
the captain
can correct for this.  It eventually arrives at it's destination and is

only a few feet
from the dock.

The total rotating mass of all the generators in a network is many times
the mass of an
ocean liner.  The operators do their best to keep them running at the
correct frequency.
Unexpected load changes can cause some divergence, but over time the
average is dead on.

When I installed power plants in the 1970's they has a special "clock"
that showed the
cumulative error in terms of clock time.  The clock had two inputs, one

from the utility
power and the other from some reference, possibly WWV.  Normally the
"clock" was
pointing up at zero and not moving.

If the generator ran a little too fast the clock would move forward.
As the operator
observed the clock moving away from zero he would reduce the plant's
power and the
clock would move backward toward zero.  His goal was to keep the clock
at zero and
not moving.  Thus, your bedside clock was always on time even if there
were temporary
excursions fast or slow.

Pete.

On 4/4/2017 5:28 PM, Thomas D. Erb wrote:

kept stable ?

On 4/5/17 11:13 AM, Bill Hawkins wrote:

Yes.. basically a bunch of coupled oscillators, and unlike the cool
demos with a bunch of metronomes on a table that self sync, the coupling
factors among oscillators are not all the same, and the damping of each
oscillator is different.

Managed historically by people turning a knob and relying on the large
mass (both literally and figuratively) keeping it from going awry.
If you mess up too much, you get a trip and your generator is offline,
suddenly, with no load.

Long transmission lines (1000s of km) cause real problems because they
have time delay that is a significant fraction of a cycle.
So now you have coupled oscillators connected by a transmission line
(with the characteristics of that transmission line time varying, to a
certain extent).

Computerized Dispatch (which is what the process of coordinating the
generation and load is) has been around since the 1960s, but it's not
perfect.

There is a pretty nice "How it Works" video on steam turbines. As Pete mentions they use valving to control the speed of the turbines, interesting how they reheat the steam for the high/medium/low stages.

https://youtu.be/SPg7hOxFItI

-=Bryan=-

From: time-nuts time-nuts-bounces@febo.com on behalf of Peter Reilley preilley_454@comcast.net
Sent: April 5, 2017 9:34 AM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Line Frequeny Stablity

The response time in a large plant is very slow.  Large steam plants
running at steady
state are running with their steam valves wide open.  A partially
closed valve is an energy
loss and is only used when changes occur.

The power control for a plant running at a steady load is the amount of
fuel thrown into
the boiler.  When you want more power you shoot more gas, oil, or coal
into the boiler.
For a nuke you pull the control rods.  Behind all of this is a lot of
thermal mass.  Things
don't change quickly.

Pete.

On 4/5/2017 9:01 AM, jimlux wrote:

time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.