Re: power-grid physics

["James R. Frysinger" <frysingerj@COFC.EDU>]

Based on my experience with bringing turbine generators on line, Michael's
comments are pretty much on the money. The governor's characteristics are
where the speed droop, that I mentioned earlier, is set. Purposely, these are
made to have a "negative droop" for reasons of system stability. Our
generators used the classic pantogrammed, spinning ball linkage that
controlled oil pressure to the hydraullically lifted throttles. If the
turbine slowed down, the balls came closer to the axle of the governor and
ported more oil into a constantly bled plenum. A device sensitive to the
generator's load included in the system bled off some amount of this pressure
in proportion to the load, thus providing the slope of the speed droop curve.
This causes the machine to rotate more slowly under full load than under zero
load --- the scale here is much less than 1 % of rated speed over its full
range.

I must take small exception to what Michael says in the portion below and I
will insert comments accordingly.

On Thursday 2003 August 21 08:45, you wrote:
....
> Contrary to what another respondent stated, the newly added generator
> will not slow down.  When connected in this manner it neither supplies
> energy to the grid nor takes energy from the grid.

        This is an inherently unstable (or at best, metastable) condition. The risks
of motorizing a turbine generator are so great that, at least on the electric
plants I operated, the oncoming generator is brought on line while spinning
very slightly faster than the system's phase rotation. This is to  ensure
positive loading of the generator. Usually, I observed only a few percent of
the load being picked up immediately upon closure of the generator's breaker,
but that was sufficient. With computerized controls, modern generating
stations can probaby do this within parts per million and so the load picked
up would not be visible on the meters. But  I would imagine that motorizing
even a large, massive, expensive, commerical turbine generator would be risky
business and one that companies try to avoid for economic reasons in addition
to safety reasons.

        Imagine a Cartesian graph with speed or frequency on the y axis and loads on
the x axis. The left portion of the x axis represents the load of one machine
and the right hand portion represents the load of the other machine. Now,
draw two straight lines from the y axis, one in each direction, but both with
some downward slope (the droop rate). One draws a horizontal line, equal in
length to the system load between the two droop lines. The height at which
this line fits is the frequency of the system powered by these two
generators. If one raises the no-load speed setting of one machine, it's
droop curve moves upward on the y axis, more of the load line is moved to its
side of the y axis (indicating movement of load to this machine from the
other), and the load line moves up a bit in frequency. The slopes of the load
lines are very small ---- less than 1 % difference in frequency between no
load and full load --- so the rise in frequency is very, very small. But it
is real and that is essential for system stability. WIth many generators on
line, the rise in frequency is even less; a human operator would not notice
it and in fact it was hard to discern on our submarines unless doing a full
load shift from one generator to another.

        At least on our submarine plants, we used the voltage settings of the
machines (controlled by adjusting field current as Michael described) to
control in essence the sharing of reactive loads. Imagine one machine putting
out enough voltage to just "cover" the real portion of the voltage phasor and
the other machine putting out enough voltage to "cover" the total voltage of
the phasor. That puts all the reactive loads on the second machine.

> .... After the connection has
> been made we can slowly raise the excitation current to raise the
> voltage of the new on-line generator.  This is when the new generator
> begins to supply power to the grid.  This is done slowly so that the
> governor-throttle can maintain the frequency and phase lock.

        I cannot imagine this working well. Again, on submarines, we controlled
relative loading of generators by moving their no-load speed settings up or
down, thus shifting the speed droop curve of one machine relative to another.
This controls the load sharing since these machines are running in parallel
and therefore are forced to be operating at the same frequency. (Power
stations do not beat generators against each other by forcing them to run at
different frequencies as we do with audio oscillators to demonstrate beat
frequencies; that would kill their system and their customers' equipment.)

> There is an instrument used by power companies to more easily allow this
> matching.  It is called a synchroscope.  If you enter this word into a
> Google search you will get about 1000 hits.  Some of them explain how
> the synchroscope works.

        That's the device I described in an email last night. Ours was a dual input
sychro receiver and the needle represented the phase difference between the
two inputs. This device came out of the old synchro-servo analog computer
designs and replaced the three synchronization lights that had been used
earlier.

> This one has a Java applet to show a synchroscope in action...
>  http://www.eece.maine.edu/Power/Java/synchro/synchro.html
>
> This one explains how it works and how it is used...
> http://www.tpub.com/fireman/121.htm

        Thanks for those links, Michael!

regards,
Jim

--

James R. Frysinger
Lifetime Certified Advanced Metrication Specialist
Senior Member, IEEE

http://www.cofc.edu/~frysingj
frysingerj@cofc.edu
j.frysinger@ieee.org

Office:
  Physics Lab Manager, Lecturer
  Dept. of Physics and Astronomy
  University/College of Charleston
  66 George Street
  Charleston, SC 29424
  843.953.7644 (phone)
  843.953.4824 (FAX)

Home:
  10 Captiva Row
  Charleston, SC 29407
  843.225.0805