Do the Math..

[SeanB]

Many power stations as well have no cold start capability, many of the larger hydro plants did eventually install some cold start capability after a few large blackouts in the world. In many cases the switching stations have no cold start capability if the blackout lasts longer than the local battery backup can operate the SCADA systems, and there is air to power the switch yard. If the blackout is too long they have to bring in a generator and do a cold start to at least close the incoming breakers for local plant power.

[technix]

Many power stations as well have no cold start capability, many of the larger hydro plants did eventually install some cold start capability after a few large blackouts in the world. In many cases the switching stations have no cold start capability if the blackout lasts longer than the local battery backup can operate the SCADA systems, and there is air to power the switch yard. If the blackout is too long they have to bring in a generator and do a cold start to at least close the incoming breakers for local plant power.

For the city of Shanghai the cold start is this HVDC conversion station. Three Gorges can cold start itself.

[grochomark]

Just a couple of items, HV/DC is produced from AC and step up transformers and then rectified into DC for transmission (so Edison was still wrong). at the other end there is a giant inverter to produce 3 phase, 60Hz AC for distribution (similar to the pictures in this thread) see ink below...

http://www.abb.com/cawp/seitp202/5685eb71f1afdf3085257ef20049e5b0.aspx

[rstofer]

4000 MW / 765 kV => 5200 Amps

I'm not sure what utilities use for ampacity but, for industrial use, 1000 MCM is only good to 600 Amps so we would need 9 strands on each side of the DC circuit.

The resistance of bare 1000 MCM is 0.011 Ohms/1000 feet and we have 9 strands to about 0.001 Ohms/1000 feet.  750 miles would be about 8 Ohms and the loss would be 200 MW or about 5% (down and back)

My guess is that they will use more than the equivalent of 9 strands of 1000 MCM.  It comes down to cost of wire versus cost of MWh.

Transmission losses are already in the 4% range:
http://blog.schneider-electric.com/energy-management-energy-efficiency/2013/03/25/how-big-are-power-line-losses/

HVDC transmission has been done for many years.  There is a 2000 MW link between France and the UK that runs at 200 kV and was installed in 1986.  This is really old news!

https://en.wikipedia.org/wiki/List_of_HVDC_projects

http://patelnishant5616.blogspot.com/2014/02/advantages-disadvantages-of-hvdc.html

[rstofer]

These projects were all economically viable at the time they were built or they wouldn't have been built.  The utilities have a pretty good handle on the losses of transmission for AC and DC circuits.  They certainly wouldn't go to the expense of rectifying and commutating DC if they could deliver the power cheaper with AC.


BTW, that 4000 MW is larger than the Palo Verde Nuclear Power Plant (3.3 GW or 3,300 MW) in Arizona.
https://en.wikipedia.org/wiki/Palo_Verde_Nuclear_Generating_Station

Palo Verde is the largest nuclear power plant in the US based on generation.

[commie]

I haven't read all the posts but the reason ac transmission is prefered over dc is that ac will go through a transformer and so, at the power station, they step up the voltage to 11kV or so using a transformer whose output approximates an ac current source.What this means is the current in the power transmission lines is tiny and relatively constant producing very low copper losses.When the electricity arrives at the houses,etc they simply transform it  from 11kV back to 240V ac(U.K.) using nothing more than a transformer. So you see, dc transmission isn't viable.

[Monkeh]

So you see, dc transmission isn't viable.

 

At the time you posted that we were receiving around 1.3 GIGAWATT via DC transmission.

[Richard Crowley]

...So you see, dc transmission isn't viable.

For "retail" local distribution and short- and medium-distance transmission that is currently true*.  However, DC is provably superior for special conditions like long-distance (where the length of the transmission line approaches the AC wavelength) or where the frequency (or even phase) of the AC at the two ends are different, etc.

* One can imagine a world in the future where high-power semiconductor technology will be cheaper than all that steel and iron and copper hanging around on all those power poles, etc.

[helius]

I haven't read all the posts but the reason ac transmission is prefered over dc is that ac will go through a transformer and so, at the power station, they step up the voltage to 11kV ... they simply transform it  from 11kV back to 240V ac(U.K.) using nothing more than a transformer. So you see, dc transmission isn't viable.

I'm sitting here laughing that you used "transmission" and 11kV in the same sentence. 11kV is used for local distribution, transmission uses 50kV and up. Transmission \$ \neq \$ distribution.

[IanB]

I haven't read all the posts but the reason ac transmission is prefered over dc is that ac will go through a transformer and so, at the power station, they step up the voltage to 11kV or so using a transformer whose output approximates an ac current source.What this means is the current in the power transmission lines is tiny and relatively constant producing very low copper losses.When the electricity arrives at the houses,etc they simply transform it  from 11kV back to 240V ac(U.K.) using nothing more than a transformer. So you see, dc transmission isn't viable.

A few things that you didn't get right about the UK electrical grid:
- Power is typically generated at a higher voltage than 11 kV (for example around 25 kV)
- It is stepped up for transmission across the national grid at voltages of 275 kV or 400 kV
- The output of generators and transformers is a voltage source--the voltage is regulated, the current varies with power demand
- The current in transmission lines is not that tiny; it could be thousands of amps
- Losses in the transmission grid are "low", maybe 2 - 3%, but this is still more than 1000 MW!
- The UK has several HVDC transmission links, to other grids in France, the Netherlands and Ireland
- Voltages in in the UK grid are based on multiples of 11 kV:
-> Transmission: 400 kV, 275 kV, 132 kV
-> Distribution: 132 kV, 66 kV, 33 kV, 11 kV
-> 11 kV is usually the last link in the chain before being stepped down to 240 V

The above is only a simplified summary of a large and complex system.

[commie]

Jeez guys, gimme a break...

If dc transmission is to be used then how exactly are you going to step up/down, oh wait, you are going to use bjt's and opamps yeah?

Right, the fact that they step it up is for what reason?, to play tiddlywinks yeah?, no its to make the current (during transmission) as small as possible and why do they do that?, because power = I^2.R so making the current small i.e., stepping up the ac voltage reduces the transmission losses.

[rsjsouza]

Back in my university days, one very knowledgeable power distribution teacher told us that the only application where HVDC transmission lines were king was in underwater links - much less overall losses when compared to AC. However the calculations at that time (mid 90's) also involved the efficiency of the inverters. This single factor may have changed since them.

[dom0]

If dc transmission is to be used then how exactly are you going to step up/down, oh wait, you are going to use bjt's and opamps yeah?

HVDC is older than op amps, or transistors. One of the first viable systems was built in Nazi-Germany and later moved to the Sovjet Union. It was based on mercury arc valves.

I'm surprised how many people here don't seem to understand that HVDC has been used for decades at scale, and how little it has to do with the 230 or 115 V coming from the sockets in your homes. It appears many here are not at all aware that there are power electronics beyond TO-220?

[SeanB]

Your typical power line runs at full load at around 1kA per conductor, and a typical voltage drop at that load is around 100V per kilometer. Want to increase power capability you either go up in voltage, or you add extra conductors to the existing, making a bundle that can carry a higher current, but with the individual cores being separated with spacers so they can be cooled with natural convection, to keep the temperature rise at full load within the sag allowance. You can run the conductors at a much higher current, but the heating and expansion means the conductors might get low enough to arc over tolocal vegetation, which has caused at least one major US blackout from the cascading trips after the one heavily loaded line tripped out after flashing over to a tree at full load.

The main advantage of HVDC is that the power transferred is simply V times I, not like AC where the voltage is limited to 70.7% of the peak ( RMS voltage versus the peak half cycle voltage) and thus the power transferred is 70% of the capacity, all else being the same.

[Monkeh]

Jeez guys, gimme a break...

If dc transmission is to be used then how exactly are you going to step up/down, oh wait, you are going to use bjt's and opamps yeah?

Right, the fact that they step it up is for what reason?, to play tiddlywinks yeah?, no its to make the current (during transmission) as small as possible and why do they do that?, because power = I^2.R so making the current small i.e., stepping up the ac voltage reduces the transmission losses.

'is to be used'? HVDC transmission IS used. In multi-gigawatt scale all over the world.

There is an active 270kVDC 2GW link between the UK and France.
There is an active 250kVDC 250MW link between Scotland and Northern Ireland.
There is an active 450kVDC 1GW link between the UK and the Netherlands.
There is an active ±200kVDC 200MW link between the UK and Ireland.

And to put all those to shame, there is an active 800kVDC 8GW link in China. Coming online next year, a 1.1MVDC (yes, MEGAVOLT) 10GW link in China. That one will be 3,333km long.

Obviously DC transmission isn't viable, that's why China's spending more money than you can comprehend building a 10GW DC transmission system.

And no, you don't use "bjt's and opamps", you use massive fucking thyristors.

Here's some thyristor arrays from a ±350kVDC 1.2GW link in New Zealand:

[SeanB]

Building each one of those stacks must take the entire month's supply from the manufacturer of those high voltage SCR units. They build them very conservatively, and test them very thoroughly, so that you have a very reliable block, and you always keep a spare locally so you can quickly swap in a good spare if one of them starts to show degradation.

[dannyf]

Does the USA know Ohm's Law?

you can benefit from reading a little bit more and talking a lot less.

If dc transmission is to be used then how exactly are you going to step up/down, oh wait, you are going to use bjt's and opamps yeah?

Having read that, i'm pretty sure that time machines were operational before the 1800s.

[commie]

still nobody has answered my question as to how you go about raising and lowering dc electricity with 90% efficiency?.Remember chaps we need to go from 240Vdc to 1000000Vdc and back again, what do have in mind?

Also, I need a vacuum cleaner that works of dc, anyone?

[Monkeh]

still nobody has answered my question as to how you go about raising and lowering dc electricity with 90% efficiency?.Remember chaps we need to go from 240Vdc to 1000000Vdc and back again, what do have in mind?

Do, we don't. And you should take a look at the efficiency figures on modern switch mode supplies some time.

ps. 90% is hilariously low for distribution, let alone transmission.

Also, I need a vacuum cleaner that works of dc, anyone?

You are once again confusing consumption, distribution, and transmission..

[Richard Crowley]

Wow, the level of ignorance around here is sometimes astounding. HVDC transmission of utility electrical power was developed back before any of us were born (1930s) in Sweden and Germany.  There are scores of HVDC transmission lines, both land and undersea all over the planet.  There is a 3.1GW line not far from where I live now (Portland, Oregon) which links to where I used to live (Los Angeles, California).

I am equally astounded at people who repeatedly flaunt their gross ignorance in public and never bother to do the most trivial research. Apparently they enjoy looking foolish in public.

You could start here:  https://en.wikipedia.org/wiki/High-voltage_direct_current

[technix]

Jeez guys, gimme a break...

If dc transmission is to be used then how exactly are you going to step up/down, oh wait, you are going to use bjt's and opamps yeah?

Right, the fact that they step it up is for what reason?, to play tiddlywinks yeah?, no its to make the current (during transmission) as small as possible and why do they do that?, because power = I^2.R so making the current small i.e., stepping up the ac voltage reduces the transmission losses.

'is to be used'? HVDC transmission IS used. In multi-gigawatt scale all over the world.

There is an active 270kVDC 2GW link between the UK and France.
There is an active 250kVDC 250MW link between Scotland and Northern Ireland.
There is an active 450kVDC 1GW link between the UK and the Netherlands.
There is an active ±200kVDC 200MW link between the UK and Ireland.

And to put all those to shame, there is an active 800kVDC 8GW link in China. Coming online next year, a 1.1MVDC (yes, MEGAVOLT) 10GW link in China. That one will be 3,333km long.

Obviously DC transmission isn't viable, that's why China's spending more money than you can comprehend building a 10GW DC transmission system.

And no, you don't use "bjt's and opamps", you use massive fucking thyristors.

Here's some thyristor arrays from a ±350kVDC 1.2GW link in New Zealand:

That 1.1MVDC link runs from Three Gorges to Shanghai, the city I live in. It is in the news for quite a while already. HVDC is much, MUCH more efficient than AC. (For those that is worried about black starting, Shanghai's entire grid can be started using the HVDC conversion station since it have PWM communication and GTO switching support.)

[rstofer]

Jeez guys, gimme a break...

If dc transmission is to be used then how exactly are you going to step up/down, oh wait, you are going to use bjt's and opamps yeah?

Right, the fact that they step it up is for what reason?, to play tiddlywinks yeah?, no its to make the current (during transmission) as small as possible and why do they do that?, because power = I^2.R so making the current small i.e., stepping up the ac voltage reduces the transmission losses.

In Reply 31 I posted a link to a document listing DOZENS of operating HVDC links with more coming on line this year and next.  This stuff has been around since 1965 when a HVDC link was built in New Zealand.  So, about 50 years or so!

Old news...

[Richard Crowley]

Remember that back before we had high-power semiconductors (or high-power firebottles, either) there was the "War of the Currents" back when they were scheming how to distribute electricity as a utility.

Thomas Edison was pushing for DC distribution, but that was completely impractical on a large scale back in the 19th century.  OTOH, George Westinghouse was a proponent of AC distribution which allowed economical high-voltage transmission, and easy conversion to lower voltage for distribution and "retail delivery".  Clearly the Westinghouse plan was the practical one and AC distribution is essentially universal around this planet.

Ref: https://en.wikipedia.org/wiki/War_of_Currents

But we are in the 21st century now, and high-power semiconductors are practical for large-scale and widespread utility applications. As I said in a previous post here, it is not difficult to imagine a future where we could see DC distribution in the "last mile" and the end of those big "pole pig" transformers at the top of wooden power utility poles.

[commie]

Do, we don't. And you should take a look at the efficiency figures on modern switch mode supplies some time.

Still waiting for a reply on how we step up/down d.c. voltages up to 1000000V??

ps. 90% is hilariously low for distribution, let alone transmission.

Last time I checked power transformers, both for transmission and distribution were about 90% efficient.

You are once again confusing consumption, distribution, and transmission..

I don't think so, but we will talk about how we convert 1000000V to 240Va.c. a little later but first we need the d.c. equivalent of an a.c.  transformer?

[rstofer]

still nobody has answered my question as to how you go about raising and lowering dc electricity with 90% efficiency?.Remember chaps we need to go from 240Vdc to 1000000Vdc and back again, what do have in mind?

Also, I need a vacuum cleaner that works of dc, anyone?

You don't change the DC voltage, you change the AC voltage before it is rectified.

What's hard about that?  A hydro plant probably generates at 12 kV (I'm just guessing) because higher levels would cause massive dielectric issues at the generator.  It absolutely does not generate 500 kV directly.  The generator output voltage is stepped up to 500 kV by ordinary transformers.  No big deal.  Just a step-down transformer in reverse (more or less).

Now, the 500 kV AC is rectified and, voila', we have 500 kVDC.  Send it on its way.

At the receiving end, there needs to be a commutating setup to convert it back to AC.  Still not a big deal.  It's been done since '65.  Read the list of projects I posted earlier.  This HVDC stuff is over 50 years old and pretty well understood.

Those windmills are going to generate fairly low voltages and it will need to be stepped up quite a bit to get into the 500 kV range.  No big deal.  In fact, I wonder if the windmills generate DC.  Considering variations in wind speed, I would think synchronizing to 60 Hz might be a challenge.

Now we can keep the level at 500 kV AC while we transmit it around the countryside and then step it down to some intermediate voltages like 115 kV.  Finally, the 115 kV will get stepped down to 12 kV at local substations and distributed to neighborhoods where small transformers will step it down to utilitization levels.

Where I worked, we bought our power straight off the 115 kV lines and stepped it down ourselves.  Then we distributed the 12 kV to about 20 unit substations where we dropped it to 277/480V 3 phase.  We used the 480V for motor loads, the 277 for lighting loads and stepped the 480V down to 120/208V 3 phase for receptacle loads.  Pretty common stuff except for buying power at transmission levels.

In New York City, the utility was still distributing DC to end users up until 2007:
http://cityroom.blogs.nytimes.com/2007/11/14/off-goes-the-power-current-started-by-thomas-edison/?_r=0

OT but interesting:  New York City still uses 25 Hz for their Amtrak system.  It is generated at a hydro site as well as being converted from 60 Hz:
https://en.wikipedia.org/wiki/Amtrak%27s_25_Hz_traction_power_system