[EE] Power grid synchronization
Charles Craft
1. Saw a blurb today about a power cable between countries and got me
thinking -
how do they synchronize all the power stations on a grid?
If it's 60HZ (US) then each cycle is 16.67ms and depending where you are
in the cycle
the slope is pretty steep so potential for voltage mis-match is large?
2. http://en.wikipedia.org/wiki/Submarine_power_cable
How do they convert the HVDC back to AC?
DC motors turning AC generators?
thanks
chuckc
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
Herbert Graf
> I had one question that turned into two after Googling around. :-)
>
> 1. Saw a blurb today about a power cable between countries and got me
> thinking -
> how do they synchronize all the power stations on a grid?
> If it's 60HZ (US) then each cycle is 16.67ms and depending where you are
> in the cycle
> the slope is pretty steep so potential for voltage mis-match is large?
The "old" way was you had a light bulb across the grid and your
generator. You manually adjusted your generation until the bulb stayed
"off" (it would blink on and off if you were off in frequency and it was
remain illuminated with a steady brightness when your frequency was
correct but your phase was off), which meant you were both phase and
frequency matched. Then you flip the knife switch and boom, you're on
the grid. After that point the generator stays synced.
We did this procedure in a power electronics lab in university, it's
pretty tricky, and God help you if you flip that switch without it being
properly synced (some students got impatient, very dumb...).
These days I'm sure the equivalent procedure is done electronically with
most generation sources.
TTYL
Bob Ammerman
Olin Lathrop
> 1. Saw a blurb today about a power cable between countries and got me
> thinking -
> how do they synchronize all the power stations on a grid?
Very carefully. As for long distance feeds, they are usually DC to avoid
this issue. For example, the feed from Hydro Quebec that ends just a mile
or so from my house is DC. When you look at the towers, yup there are only
two lines instead of the usual 3.
********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014. Gold level PIC consultants since 2000.
Carl Denk
with DC was it couldn't be transported very far before the losses left
little power available, and the AC was necessary to get the power some
distance from the generating station. This goes back to Edison's days.
Herbert Graf
> I thought that way back in the early electricity days, that the issue
> with DC was it couldn't be transported very far before the losses left
> little power available, and the AC was necessary to get the power some
> distance from the generating station. This goes back to Edison's days.
The "problem" with DC back in the day is you couldn't step it up/down
cheaply.
Stepping it up was necessary due to I2R losses in the transmission
lines. AC got around the problem by using cheap transformers to step it
up, transmit it and then step it back down. For DC you'd have had to use
rotary converters, very expensive and a maintenance horror (imagine
every pole pig was a rotary converter!).
On the other side of things, AC has it's issues as well. For REALLY long
distances the skin effect starts to have a bearing, along with the
bigger issue of power loss (due to coupling) to the ground. DC doesn't
have these issues.
As a result of advances in high power solid state electronics DC is now
a viable option for long runs, as evidenced by the line Olin described.
TTYL
Bob Ammerman
-- Bob Ammerman
RAm Systems
Olin Lathrop
> I thought that way back in the early electricity days, that the issue
> with DC was it couldn't be transported very far before the losses left
> little power available, and the AC was necessary to get the power some
> distance from the generating station. This goes back to Edison's days.
There is nothing inherent to DC that prevents pushing power long distances.
In fact, DC is more efficient due to no skin effect and no inductive or
capacitive or radiation losses. The farmer stealing power with a long fence
next to the power line isn't possible with DC, for example. In the early
days it was probably difficult to get high voltage DC. That's easier with
AC as you can run it thru a transformer.
It's pretty much a voltage versus current game. The longer the distance,
the more it makes sense to pay for high voltage infrastructure. At higher
voltages it takes less current to transfer the same power, and the
limitations and losses are related to current (mostly).
Charles Craft
At about 100 lbs/meter, that's some beefy cable.
Cable data
Voltage 450 kV DC
Power 1,000 MW
Insulation MI – oil impregnated paper
Weight 44 kg/meter
Length 494 km submarine cable
16 km underground cable
Conductor 1 x 1430 mm2 Cu
Customer BritNed Development Ltd
National Grid (U.K.) and
TenneT (the Netherlands)
Year 2007-2010
>> http://www.piclist.com PIC/SX FAQ& list archive
Steve Smith
converter both ends (it's the only topology that is reversible) It arrives
in England at Sellinge converter station at 500Kv DC if my memory is
correct...
Steve
-----Original Message-----
From: piclist...@mit.edu [mailto:piclist...@mit.edu] On Behalf Of
Olin Lathrop
Sent: 08 June 2010 22:36
To: Microcontroller discussion list - Public.
Subject: Re: [EE] Power grid synchronization
zip...@comcast.net
Olin Lathrop
> Voltage 450 kV DC
> Power 1,000 MW
That's 2.2KAmps.
> Conductor 1
??
> 1430 mm2 Cu
That's the same area as a 1.7 inch (43mm) diameter circle.
********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014. Gold level PIC consultants since 2000.
--
Bob Ammerman
----- Original Message -----
From: "Olin Lathrop" <olin_p...@embedinc.com>
To: "Microcontroller discussion list - Public." <pic...@mit.edu>
Sent: Wednesday, June 09, 2010 7:41 AM
Subject: Re: [EE] Power grid synchronization
> Charles Craft wrote:
>> Voltage 450 kV DC
>> Power 1,000 MW
>
> That's 2.2KAmps.
>
>> Conductor 1
>
> ??
>
>> 1430 mm2 Cu
>
> That's the same area as a 1.7 inch (43mm) diameter circle.
Don't bet on power distribution using such large wire. They'll optimize the
wire size for cost of lost power vs. initial cost of wire constrained by
maximum (free air) temperature rise. The temp rise they'll allow is a lot
higher than e.g. NEC allows.
As an example, in my old house (built around 1850) the electric service was
100A (before I upgraded it to 200A). The power company used about 100+ feet
of #6 AL wire to bring the power to the house. (After I upgraded to 200A
they replaced it with 1/0 cable). When I went to install a 100A sub-feeder
to an out building I was required to use 2/0 AL IIRC.
-- Bob Ammerman
RAm Systems
--
Mark E. Skeels
What does a 500KV DC converter look like?
How do you even begin to create a solid state device to do that?
Mark
Olin Lathrop
>>> 1430 mm2 Cu
>>
>> That's the same area as a 1.7 inch (43mm) diameter circle.
>
Since the cross section area of copper was specified, it has to be at least
that size in this case. Actually I expect the cable to made of a bunch of
smaller strands, each small enough to not suffer too much skin effect and to
be more flexible. That's why I carefully said 1430 mm2 is the same area as
a 1.7 in diameter circle, not that the copper cable was a 1.7 in diameter
cylinder.
RussellMc
>> Voltage 450 kV DC
>> Power 1,000 MW
Olin said:
> That's 2.2KAmps.
>> 1430 mm2 Cu
> That's the same area as a 1.7 inch (43mm) diameter circle.
About half the current density allowed here
http://www.powerstream.com/Wire_Size.htm
they prorate about linearly BUT for butied cables the figures will be
lower due to temperature rise.
That mass per meter is about 3.5 x the mass of copper involved.
Seems like there is a lot of other material there as well. Maybe armour ? :-).
Russell
Dave Lagzdin
clocks were co-ordinated and arranged via telex the precise time of
slightly shifting the HZ.
They worried about long term consistancy and had a 3'rd party track
and report across
North America. No doubt methods have progressed :)
On 8 June 2010 15:50, Charles Craft <chuc...@mindspring.com> wrote:
> I had one question that turned into two after Googling around. :-)
>
> 1. Saw a blurb today about a power cable between countries and got me
> thinking -
> how do they synchronize all the power stations on a grid?
Robin D. Bussell
developed a technique of accurately dating a recording by analysing the
frequency of the mains hum on the recording, seeing how it varies with
time and then comparing this signature with a database of line frequency
variations. This sort of thing was impossible before digital recordings
as tape speed would vary too much to track the minute variations in line
frequency, apparently.
Interesting!
Cheers,
Robin.
-----Original Message-----
From: piclist...@mit.edu [mailto:piclist...@mit.edu] On Behalf
Of Dave Lagzdin
Sent: 09 June 2010 14:49
To: Microcontroller discussion list - Public.
Subject: Re: [EE] Power grid synchronization
enk...@gmail.com
> Bob Ammerman wrote:
>
>>>> 1430 mm2 Cu
>>>>
>>> That's the same area as a 1.7 inch (43mm) diameter circle.
>>>
>> Don't bet on power distribution using such large wire.
>>
> Since the cross section area of copper was specified, it has to be at least
> that size in this case. Actually I expect the cable to made of a bunch of
> smaller strands, each small enough to not suffer too much skin effect and to
> be more flexible. That's why I carefully said 1430 mm2 is the same area as
> a 1.7 in diameter circle, not that the copper cable was a 1.7 in diameter
> cylinder.
>
Skin effect at DC???
MJ
Olin Lathrop
>> Since the cross section area of copper was specified, it has to be
>> at least that size in this case. Actually I expect the cable to
>> made of a bunch of smaller strands, each small enough to not suffer
>> too much skin effect and to be more flexible. That's why I
>> circle, not that the copper cable was a 1.7 in diameter cylinder.
>
> Skin effect at DC???
Oops. Yes we were talking about DC, so there is no skin effect. Sorry for
the confusion.
RussellMc
Very long cable ?
((very) arguably skin effect appears as length approaches infinity
when f=0 :-) )
The fact that total mass per metre is about 3 x copper mass made me
wonder if it is 3 phase DC *
Russell
* :-)
:-) :-)
maybe:
* :-)
|
/ \
:-) :-)
Sean Breheny
You must be a very interesting guy to be around in person! :)
Sean
RussellMc
> > wonder if it is 3 phase DC *
> Russell,
> You must be a very interesting guy to be around in person! :)
Some think so :-).
Some are just stunned.
Ask James Newton. (Or my wife :-) ).
My wife and I and a friend stayed with him and Maria for several days in 2003.
He found the experience "interesting".
(I was (even) more freneticised than usual as that was the start of a
9 week world tour and I was still desperately trying to get my in-car
PC and camera power and backup systems working properly. Nowadays it
just seems to work. Then it was a bit harder. Many technical issues
now but a dim and smouldering memory. ).
Russell
Electron
>On Tue, 2010-06-08 at 15:50 -0400, Charles Craft wrote:
>> I had one question that turned into two after Googling around. :-)
>>
>> 1. Saw a blurb today about a power cable between countries and got me
>> thinking -
>> how do they synchronize all the power stations on a grid?
>> If it's 60HZ (US) then each cycle is 16.67ms and depending where you are
>> in the cycle
>> the slope is pretty steep so potential for voltage mis-match is large?
>
>The "old" way was you had a light bulb across the grid and your
>generator. You manually adjusted your generation until the bulb stayed
>"off" (it would blink on and off if you were off in frequency and it was
>remain illuminated with a steady brightness when your frequency was
>correct but your phase was off), which meant you were both phase and
>frequency matched. Then you flip the knife switch and boom, you're on
>the grid. After that point the generator stays synced.
How does this AC system work relatively to the DC version (i.e. diodes)?
I mean, will the generator only contribute to the power grid, or will it
work like a engine itself, if the (mechanical) energy is not strong enough?
On a DC system diodes take care of it.. but in the AC system?
Herbert Graf
> >The "old" way was you had a light bulb across the grid and your
> >generator. You manually adjusted your generation until the bulb stayed
> >"off" (it would blink on and off if you were off in frequency and it was
> >remain illuminated with a steady brightness when your frequency was
> >correct but your phase was off), which meant you were both phase and
> >frequency matched. Then you flip the knife switch and boom, you're on
> >the grid. After that point the generator stays synced.
>
>
> I mean, will the generator only contribute to the power grid, or will it
> work like a engine itself, if the (mechanical) energy is not strong enough?
It will "work as an engine" if there isn't enough mechanical power (i.e.
if you disconnect the generation source the "generator" will be driven
by the grid).
RussellMc
not.
> I mean, will the generator only contribute to the power grid, or will it
>
> work like a engine itself, if the (mechanical) energy is not strong
enough?
>
> On a DC system diodes take care of it.. but in the AC system?
The power station can progressively "put its shoulder to the load". You can
make voltage and no power and even run as a motor BUT usually you drive
energy into the alternator and it TRIES to accelerate the alternator and
instead phase shifts it ever so slightly and adds power. If the alternator
makes a significant percentage of the total network power available you can
start to influence the frequency or voltage but usually the network is
substantially more capable and any one alternator "gets in behind". If
desired the alternator can be run out of phase with the network and make
VARs* but no power, to balance net reactive imbalance. VARs cost $ to send
and make no power at the other end so killing them locally makes sense. It
may be economic to have a small power station for this purpose - eg a gas
turbine container unit installed in an industrial area. (* Gargoyle knows).
DC links driving AC converters to small locales where link energy is a large
% of total can cause major issues. Without adequate control the link can
become unstable and drive voltage and frequency to interesting values. Long
ago Sardinia (I think) (just maybe Sicily) had this issue when mainland
fed.
Re NZ DC link across Cook Strait (among the worlds most treacherous
waterways) - voltage AFAIR is 500 kV DC wrt ground and cable are run in +/-
pairs for 1 MV per pair. Memory may misserve - may be half that each. Still
'a lot of voltage'. Fishing boats are not overly welcome :-).
Russell
Xiaofan Chen
> 2. http://en.wikipedia.org/wiki/Submarine_power_cable
> How do they convert the HVDC back to AC?
> DC motors turning AC generators?
http://en.wikipedia.org/wiki/High-voltage_direct_current
I heard about this long time ago when I was a secondary school
student after visiting various power stations during a summer camp.
I heard it in the university as well.
But in real life, I might only came across this one since an
adjunct professor from Southern California Edision gave
some lecturers about Power Generation.
http://en.wikipedia.org/wiki/Pacific_DC_Intertie
I heard SCE was considering setting up Electric
Charging Stations for the Electric Vehicles across
Southern California. That must be a vaporware...
--
Xiaofan http://mcuee.blogspot.com