Letters to the Editor
-
Silicon vs. cadmium, etc.
While there may be a choke point in high-grade silicon production, it is worth noting that silicon is made from one of the most abundant substances on the earth (silica), while the new materials mentioned in the article are rare heavy metals.
Stick with silicon, and have faith in supply and demand. If you wave money at it, it will come.
-
Simple Figurin'
It occurred to me that all places on Earth get equal times of daylight over the course of a year. So... Sun Angle and Cloudiness are the determining factors.
If you combine daylight duration and sun angles, at midsummer, Minneapolis gets 91% as much sun as Houston. At midwinter, Minneapolis gets 58% as much. So, overall, I guess Minneapolis splits the difference:
Minneapolis gets just about 75% as much sun as Houston. I am not sure how much we can equalize this just by changing the angles of the solar collectors.
Cloudiness will be a huge aditional factor, obviously.
-
100 miles squared
The point is being missed. The area needed for one central station plant to power the whole country is manageable (100 x 100 miles or whatever it is) but this is simply illustrative.
You need diversity of technology, geography, and so on. The point of this image is to illustrate the possible. Deploying solar power this way is terribly inefficient.
PV is a great technology as a distributed generation asset because it cleanly, silently converts sunlight to kWh at a scale from sub-1 Watt to multi-MW. Rooftop installations are not just possible, but actually fairly common at 1 kW to 1 MW scale (residential through "big box") and peak generation is timed to peak loads (air conditioning), displacing expensive and inefficient peaker plant generation. The specific PV technology (crystalline, thin film, CPV) that is used is not a function of which is "better" but which is more cost effective for a given application; this varies based on the technology's efficiency, the building load profile and rate tariff, and available roof space.
Wind in contrast is great for displacing base-load, especially at night. Solar thermal electric is also good for central-station peaking and there is the possibility for storage. However both wind and solar thermal electric suffer a few disadvantages as compared to PV. First and foremost, they have to be installed in a central station model so they must compete with wholesale electricity costs, which are much less than retail peak rates the distributed PV displaces. Second they suffer from the same transmission and distribution losses as traditional generation. Third, they require much more ongoing maintenance (especially solar thermal electric) and active management. Fourth, they take a long time to permit and deploy, much longer than PV. Wind is also less predictable or "firm" than solar which utilities must take into account when managing their generation mix.
I should also address some misconceptions about thin-film. First thin-film is not "new". There was a significant push in the 80's and 90's to bring these technologies to market by big players (Arco / Siemens / Shell, BP, and others) which fell utterly on its face due to reliability problems. Unisolar has successfully sold a niche (low efficiency, but flexible) a-Si thin film product for at least 10 years. The jury is out on the reliability the thim-film technologies being deployed more recently (First Solar, Nanosolar). Also it should be noted that these technologies are far from "being sprayed on buildings" and so forth. The form factor is traditional, glass mdoules as the PV materials are subject to degradation by the elements and must be protected. "Spray On" PV tech has been made in the lab but lasts a very short period of time outdoors, far too short to be useful. Maybe one day, but it has been "right around the corner" for 10 years at least. The closest commercialized tech is Konarka "power plastic" - but read the fine print - it will only last for a couple of years!
-- A solar engineer
-
April Fool's joke?
"The principal problem with solar energy, as the column correctly points out, is that the sun shines less than half the time, on average, at any location on earth."
No, that's not it. The big problem is that the total system cost of the panels, conversion equipment, etc., is still rather high for the energy recovered, compared to other technologies.
"However, a satellite correctly placed in geosynchronous orbit can receive direct sunlight 99% of the time."
That's true. But it's 27,000 miles or so away from where it's needed.
"a network of solar power collector satellites could collect enough energy to power all current uses on our planet, with no adverse environmental consequences."
I call shenanigans. Almost everything we do has adverse environmental consequences.
"The energy collected by space-based solar power satellites can be transmitted wirelessly to the earth's surface as microwaves, and fed directly into the power grid with minimal processing."
Not really. Let's think about this a moment....
First we have to build the satellites, which must be capable of enduring the space environment for decades. They have to contain not only the panels but the microwave conversion equipment and antennas. Plus they must beam the microwaves to a precise location on earth where the collectors are, and be absolutely safe against misalignment (or they will cook whatever the microwaves hit, if the energy density is high enough to be useful.)
How much would it cost per watt to build such satellites? Granted, they could be mass-produced, but even with such economies we're talking about a lot of very expensive hardware.
Then we have to launch the satellites and get them into orbit. These are huge satellites, too, if they're going to collect and process lots of energy.
How much would it cost per watt to launch them?
There will also be needed a system of ground collectors to intercept the microwaves and convert them to useful electricity, and feed it into the grid. How much will all that cost?
Consider all the losses in the system beyond just the PV panels, using wildly optimistic numbers:
- Conversion of DC from panels to microwaves: 50%
- Transmission from orbit to earth: 50%
- Conversion from microwaves to useful energy: 40%
0.5 x 0.5 x 0.4 means that for every 10 watts of electricity from the orbiting panels, we get maybe 1 watt of useful electricity on earth. Maybe. The advantage of 99% sunlight is lost in all the inefficiencies of getting the electricity to where it's needed. And those are wildly optimistic numbers; transmission of useful energy over long distances by microwaves is totally unproven technology. The reality might be 100 to 1 rather than 10 to 1.
Now add in all the energy needed to build the satellites, launch them, and keep them in orbit. Think of how many would be needed just to power one typical American city, given all the inefficiencies.
Then add up how much it would all cost. Remember too that we'd probably need a manned spaceflight system that could reach geosynchronous orbit to repair them. What would that cost?
Earth based systems start looking really good...
"Why doesn't the author of the article mention this option?"
Because it's not even close to being realistic compared to ground-based systems.
"It deserves to be discussed."
Yes, it does, to determine just how much it would really cost to do, and all the ramifications.
How much are *you* willing to pay per kilowatt-hour? That's the bell-the-cat question.
There's an old saying:
"Engineering is doing for a penny what anyone can do for a pound".