Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[10ebbor10]

That's not the cost of a solar only grid. Neither is it the cost of storage.

It's the cost of a smart  grid, which is more secure and more advanced in allocating power. It could aid solar integration, but it won't carry solar on it's own.

http://www.nytimes.com/cwire/2011/05/25/25climatewire-smart-grid-costs-are-massive-but-benefits-wi-48403.html?pagewanted=all

On a side note, a smart grid wouldn't really help with solar at all. The entire point of the smart grid is to balance load, such as to reduce required load changes and peak/baseload difference.

Seems like Nuclear would be a perfect fit.

[GavJ]

That article confuses me. It barely says what a "smart grid" is in the first place, and then it says completely conflicting things like:
"It will save money" but
"the average consumer will pay $9-$12 more per electric bill"


Who exactly is saving money, then? Or is that supposed to be after like 3 years, and then the next 17 are gravy? Or what? No details, no explanation.

[mainiac]

1) What does that even mean? A grid costs initial and ongoing annual costs. One number doesn't make sense.
2) Citation of that number, whatever it is, for solar-only + storage?

Why store when we can transmit?

Look we are talking about taking something that is very expensive (electricity geographically isolated from markets) and making it really really cheap.  That means the old delivery paradigm should be rethought.  Right now it makes sense to generate electricity at the source because power doesn't get cheap enough far from cities to support long range transmission.  So you make it near the market.  But in 20 or 30 years we might be looking at a situation where it costs 10 times as much to make energy continuously than to make it sporadically.  That means that there is a opportunity to save money by long range transmission, sending power worldwide.

The Sun is always shining somewhere.  And if we are talking 20 or 30 years in the future, transmission looks very attractive.

[GavJ]

That means that there is a opportunity to save money by long range transmission, sending power worldwide.

Yes and I'm waiting on a citation for a feasible system for beaming half the entire planet's electricity 12,000 miles, in a way that allows for the point A and the point B to constantly shift, and then somehow getting it from there into each individual home reliably, all of which costs demonstrably less than local nuclear.

[miauw62]

How do you know it will look attractive in the future? Long-range transmission will make you lose a lot of your power, and you'll be transporting power from around the equator toward the poles, and around more than a quarter of the world so people can turn on the lights at night. And often even both.

We don't have ANY superconductors above even liquid nitrogen temperature. Even with technology 20 years into the future, you'll probably have to pump liquid nitrogen along thousands of kilometers of power cable to transmit efficiently.

[10ebbor10]

HDVC, which is about the best tech we have for now, has 3.5% losses per 1000 km, and a 5 million per GW per km cost.

So if you're transporting power further than 1000 km, it's cheaper to go Nuclear.

Your argument, is very, very hypothetical, relying cost of solar to crash, on cost of storage to crash or on cost of transmission to crash.

[GavJ]

By comparison, one of the longest DC power lines in existence is the pacific intertie: http://en.wikipedia.org/wiki/Pacific_DC_Intertie
Which says it carries 3 gigawatts maximum.
It's also built overland, whereas a world transit system would need to be mostly (and much more expensiuvely) underwater. But I'll completely ignore that.

World energy consumption is 21~ terawatts. Let's say 1/3 needs to be used at night. That's 7 terawatts = 2,000x the most ambitious existing lines' capacity. Add on quite a bit to that for additional average losses per mile for higher distances, probably more like 3,000+ (being VERY generous and futuristic here)

Now add on the extra length (so far we've been talkign about cross-sectional capacity costs, not how much longer it would be). It needs to ring the entire world, either 2 or 3 times, or go up and down each continent chain as well. Something in the neighborhood of 50,000 miles, let's say. That's 50x longer than the intertie, so we're up to 150,000x the cost.

I can't seem to find a cost for the intertie's initial construction, but some recent maintenance and upgrades to it were estimated at $428 million.  So just for upgrade/maintenance of your cable, as a scaled comparison, you'd be looking at 428,000,000 * 150,000 = $64 trillion with today's technology...

[Sergarr]

I doubt that the cost saved by using superconductors will be greater than the cost of supporting all that liquid nitrogen... so why even bother =\

[mainiac]

Yes and I'm waiting on a citation for a feasible system for beaming half the entire planet's electricity 12,000 miles, in a way that allows for the point A and the point B to constantly shift, and then somehow getting it from there into each individual home reliably, all of which costs demonstrably less than local nuclear.

Well we already have projects that go 10% of the way.  It's not exactly rocket science to figure out how to make things go farther, there just isn't a need.

How do you know it will look attractive in the future? Long-range transmission will make you lose a lot of your power, and you'll be transporting power from around the equator toward the poles, and around more than a quarter of the world so people can turn on the lights at night. And often even both.

We don't have ANY superconductors above even liquid nitrogen temperature. Even with technology 20 years into the future, you'll probably have to pump liquid nitrogen along thousands of kilometers of power cable to transmit efficiently.

It will look attractive if the cost of electricity falls enough.  You dont need high efficiency if electricity is cheap enough.

So if you're transporting power further than 1000 km, it's cheaper to go Nuclear.

How many times must I repeat: I AM TALKING ABOUT INVESTMENTS WE WOULD BE CONSIDERING 20 OR 30 YEARS IN THE FUTURE.

[GavJ]

Well we already have projects that go 10% of the way.  It's not exactly rocket science to figure out how to make things go farther, there just isn't a need.

1) 10%? Erm... I'd love to see your map that somehow draws a line between every country in the world using only 12,000 miles of cable. This should be amusing. MS Paint will be sufficient, thanks.
2) http://new.abb.com/systems/hvdc/references/xiangjiaba---shanghai It only has a 7200 MW capacity.  That's a THOUSAND TIMES less than what you need. In case you are not aware, a wire that can carry 1,000x as much electricity... costs more... to say the least.

[mainiac]

Well we already have projects that go 10% of the way.  It's not exactly rocket science to figure out how to make things go farther, there just isn't a need.

1) 10%? Erm... I'd love to see your map that somehow draws a line between every country in the world using only 12,000 miles of cable. This should be amusing. MS Paint will be sufficient, thanks.
2) http://new.abb.com/systems/hvdc/references/xiangjiaba---shanghai It only has a 7200 MW capacity.  That's a THOUSAND TIMES less than what you need. In case you are not aware, a wire that can carry 1,000x as much electricity... costs more...

Why would you connect every country in the world?
You seemed determined to design the most inefficient system possible.

[10ebbor10]

Yes and I'm waiting on a citation for a feasible system for beaming half the entire planet's electricity 12,000 miles, in a way that allows for the point A and the point B to constantly shift, and then somehow getting it from there into each individual home reliably, all of which costs demonstrably less than local nuclear.

Well we already have projects that go 10% of the way.  It's not exactly rocket science to figure out how to make things go farther, there just isn't a need.

You continue to loose efficiency. With a 3.5% loss per 1000 km, you're looking at a 80% efficiency for 5000 km. Resolving said efficiency loss will require upping voltage to silly levels, resulting in very spectacular failurs.

In addition, you need to maintain grid reliability, so you will need a lot of back-up capacity in your network.

How do you know it will look attractive in the future? Long-range transmission will make you lose a lot of your power, and you'll be transporting power from around the equator toward the poles, and around more than a quarter of the world so people can turn on the lights at night. And often even both.

We don't have ANY superconductors above even liquid nitrogen temperature. Even with technology 20 years into the future, you'll probably have to pump liquid nitrogen along thousands of kilometers of power cable to transmit efficiently.

It will look attractive if the cost of electricity falls enough.  You don't need high efficiency if electricity is cheap enough.

You're stacking up a lot of speculative thingies though.

How many times must I repeat: I AM TALKING ABOUT INVESTMENTS WE WOULD BE CONSIDERING 20 OR 30 YEARS IN THE FUTURE.

That does not give you a free pass on any considerations surrounding technical feasibility. You have given no indication whatsoever that this project might be even remotely feasible, Let alone desirable.

Just waving hands, and saying the future will resolve problems won't solve anything.

I mean, obviously in 15 years there will be a cold fusion energy breakthrough and everyone will have infinite energy forever.

Edit: Just really, the amount of speculative technologies you've thrown in here is getting a bit silly.

[Gentlefish]

Plus the redundancies if the SINGLE power plant somehow loses power would be !!fun!! to try out the first time.

[10ebbor10]

Plus the redundancies if the SINGLE power plant somehow loses power would be !!fun!! to try out the first time.

It's solar, square kilometers and square kilometers of solar. If there's one thing this system doesn't have, it's a single point of failure.

((Provided there's enough transmission back-up, that is.))

[GavJ]

Why would you connect every country in the world?

Because all the major population centers need power...? They need to be hooked into the world day/night cables if that's your plan.

And if you're judging your costs based on 7,200MW (which you are), you'll need to actually hit every country.

For example, South Africa alone uses 3x as much as that, so at 1/3 night energy use, they need an entire dedicated line the size of the Shanghai cable to provide them sufficient long distance nighttime energy.

The U.S. uses 18x as much as that and needs the equivalent of 18 Shanghai HVDC lines to cover it

etc.



ALL of those need to be plugged into either a consistent world grid, or at the very least, their own "sister cluster" of nations on the other side of the world from them with different day/night.

And I haven't even gotten into the fact that you have much less than 12 hours of sunlight away from the equator at certain times of year. (Thus near-polar countries need cables thick enough to give them nearly 100% of their energy for their winter accommodations AND they need North-South connectivity in addition to or instead of East-West, hugely complicating your whole cable layout. Summer doesn't help you, because the cable still needs to be there for winter.)