Economic prospects for solar photovoltaic power

[firejuggler]

Why does every time I hear of 'storing power' I have a Matrix flashback?
(and yes I know i'm either a clown or a village idiot)

[kladner]

Originally Posted by chalsall

It would have to be ***VERY*** high DC voltage, and is only more efficient than AC when the cable is routed through water (or some other inductive medium).

Aren't there reactive losses even with wire in air?
Losses:
http://en.wikipedia.org/wiki/Electri...mission#Losses

As of 1980, the longest cost-effective distance for Direct Current transmission was determined to be 7,000 km (4,300 mi). For Alternating Current it was 4,000 km (2,500 mi), though all transmission lines in use today are substantially shorter than this.

Note that these are 1980 figures, though I'm not sure that there have been big changes since then.

Distance finder, example: Tunis to Berlin, crow flight and land route:
http://www.distance.to/Tunis/Berlin

I chose Tunis as an easy marker at the edge of a short ocean crossing, near substantial amounts of desert. Berlin is a pretty good marker for northern Europe. Getting the power to more southern locations would be easier, but even Berlin is in reasonable range for the transmission line lengths given. I'm only looking for rough estimates to show possibilities.

In fact, here is a rather gloomy report on the actual project.

http://www.nature.com/news/sahara-so...-shine-1.11684

It sure looked good in the beginning-
http://en.wikipedia.org/wiki/Desertec

[chris2be8]

Originally Posted by cheesehead

... or enough storage capacity ...

Storing several hundred GigaWatt-hours of electricity would be expensive. Work out how much 100 GigaWatt-hours of battery capacity would cost.

Also how many countries in north Africa would you think stable enough to rely on to provide your electricity supply?

I used to work in the electricity industry so I know quite a lot about the subject. The point I was trying to make is that solar power is much more cost effective in some parts of the world than others. And Germany is one of the places where it does not make economic sense today.

Chris

[chappy]

Originally Posted by chris2be8

Storing several hundred GigaWatt-hours of electricity would be expensive. Work out how much 100 GigaWatt-hours of battery capacity would cost.

Also how many countries in north Africa would you think stable enough to rely on to provide your electricity supply?

I used to work in the electricity industry so I know quite a lot about the subject. The point I was trying to make is that solar power is much more cost effective in some parts of the world than others. And Germany is one of the places where it does not make economic sense today.

Chris

I agree that it doesn't make sense in Germany, but 100 million euros has made it feasible. As I said in my earlier discussion Germany has taken a huge economic hit for very little gain, but the plus side is that the rest of the world can benefit from that gain. (In fact it only compounds the problem if the rest of the world doesn't take advantage!)

The article about the US military heading that direction is also of economic benefit to the rest of the world. So long term we have at least one working model of a large and industrialized country implementing (however flawed) solar power. And we have a huge industrial-complex of the military pushing for lighter weight and more robust materials. There is no downside now that the money is already spent. (I've already covered the downsides as I see them here, and elsewhere in this thread).

We've seen some discussion here about the real problems that renewables face: 1) Energy storage for when the sun isn't shining or the wind dies down. (my anecdotal knowledge of solar power is that even a partially cloudy day has a significant impact (25-40%) on total power generated. The company I work for has test banks of several of the competing technologies 2) cost effective base load power. 3) variabilities in the various grids that don't allow power to be transfered across regions. 4) line loss and other problems related to transporting power.

When we fix any one of those issues we will be ready to talk about replacing the current bad system with something better.

[kladner]

Originally Posted by chris2be8

Storing several hundred GigaWatt-hours of electricity would be expensive. Work out how much 100 GigaWatt-hours of battery capacity would cost.

Also how many countries in north Africa would you think stable enough to rely on to provide your electricity supply?

Chris

Thermal storage probably makes more sense than storing the electricity itself.

Instability is a problem in many regions, many of which have good solar potential. However, there is a great deal of dependence on North Africa for energy supplies already. To what other regions and sources would you turn?

[chalsall]

Originally Posted by chappy

High Voltage DC has much less line loss than AC, and creates less heat which equals less sagging in the lines. Also means less insulation is needed which reduces cost. The greatest advantage from a transmission system operations standpoint is that there is no need to synchronize the grids.

OK, let's think (and talk) about this.

Let's say you have 120,000 volts of AC. How do you down convert it to 120 V AC?

A simple ferric ring with windings will do. Also known as a transformer. It will give off some heat.

Let's say you have 120,000 volts of DC. How do you down convert it to 120 V DC?

A pulsed modulation impulse into a capacitor might do, with a resister downstream, but that control might be rather expensive. And, it will also give off heat.

Just thinking out loud....

[kladner]

I'm a little tied up fixing dinner, but here's a basic rundown on conversion-
http://en.wikipedia.org/wiki/HVDC_converter_station

and on transmission lines. Note that typical voltage in AC setups is above 110KV, but there are limits related to corona discharge leakage.
http://en.wikipedia.org/wiki/High_vo...nsmission_line

A big reason for HVDC transmission is efficiency. Losses a substantially lower.
http://en.wikipedia.org/wiki/High-vo...direct_current

There is a link in an earlier post in this thread which addresses voltage limits and relative efficiency of DC and AC over long distances.

[chalsall]

Originally Posted by kladner

I'm a little tied up fixing dinner, but here's a basic rundown on conversion...

I hope you appreciate that I'm being playful here, and I might be completely wrong.

My understanding it is very easy to convert AC to DC (a rectifier (four diodes), and maybe a few capacitors and resistors), not so easy to convert back.

Further, it is very easy to convert voltages when the power is presented as AC; rather difficult to convert voltages when it is presented as DC.

But, then, things might have changed since I went to school....

[kladner]

I am happy with discussion, don't worry.

My impression from reading some of the articles is that the efficiencies of conversion and transmission trade off against each other. I don't remember the exact figures posted earlier, but I think DC transmission was somewhere in the area of twice as efficient. One way this was stated in practical terms was the distances over which the two approaches could be used.

As to conversion, there are several kinds of devices. Solid state tends to dominate now with thyristors playing a role. I would also expect diacs and triacs (SCRs) to be involved. However, I will have to read more to be sure of efficiencies between solid state and older technology. An older device mentioned is the mercury vapor valve, which is, of course, the British term for an electron tube.

Having grown up fascinated with electronics when tubes still ruled, I remember reading about very high power rectifiers which contained mercury vapor. Given the conductive nature of ionized metal vapor, I would not want to bet that solid state devices are more efficient. They are, however, more compact and robust by a long shot. Diesel-electric locomotives changed over from DC to AC traction with the advent of high-power solid state devices which could provide frequency control of the motors. It was technically possible before, but utterly impractical to have a tube rig to control your locomotive.

[chappy]

It's hard (or impractical) to step up and step down DC at the level we are talking about. The current (watts up with that word?) practice is stacked thyristors to convert the power back to AC for stepping back down.

http://www.hydro.mb.ca/corporate/fac...s_nelson.shtml

another less efficient but much more fun way it the motor-generator, which can be used to transfer power from AC to DC or step up and step down any conversion you want. The downside is a loss of efficiency inverse to the efficiency of the mechanical torque, but even that loss of efficiency is less than line loss in HVAC lines over about 600miles (1000km).

Again this isn't exactly my area of expertise. I deal with 125vDC and 24vAC. And what I can tell you from personal experience is that DC hurts when it bites you. But don't those Thyristor stacks (DC voltage Valves!) look awesome?

http://www.ptd.siemens.de/B4-203.pdf

[chappy]

Originally Posted by kladner

I am happy with discussion, don't worry.

My impression from reading some of the articles is that the efficiencies of conversion and transmission trade off against each other. I don't remember the exact figures posted earlier, but I think DC transmission was somewhere in the area of twice as efficient. One way this was stated in practical terms was the distances over which the two approaches could be used.
.

I think this is a pro-HVDC way to phrase it "in the area of twice as efficient" but yes, I'm reliably informed that AC line losses CAN reach up to 24% of total load. And DC line losses over the same long distances CAN reach up to 12-15%. So it's power cost x loss - any additional steps (like electronics and inversion) + cost savings on insulation. HVDC (I'm told by the EE's I work with) becomes cost effective at around 600miles (1000km) that I referenced above, when AC line losses cost more power than the resulting transmission costs.