Google Getting Into the Solar Mirror Business

adeelarshad82 writes with this excerpt from a Reuters report: "Google is disappointed with the lack of breakthrough investment ideas in the green technology sector, but the company is working to develop its own new mirror technology that could reduce the cost of building solar thermal plants by [25%] or more. The company's engineers have been focused on solar thermal technology, in which the sun's energy is used to heat up a substance that produces steam to turn a turbine. Mirrors focus the sun's rays on the heated substance. ... Google hopes to have a viable technology to show internally in a couple of months, Bill Weihl said. It will need to do accelerated testing to show the impact of decades of wear on the new mirrors in desert conditions."

Comments

Power?

[Coren22]

I guess they figured out thier electric bills were too high.

Re:Power?

[Fred_A]

I guess they figured out thier electric bills were too high.

Is it just me that's annoyed that in most power plants we actually still use glorified steam engines ?

I know that it's the best way we currently have to convert heat (which is the only type of energy we manage to recover) into electricity, but it still feels kludgy. I hope we'll figure out something else eventually.

Re:Power?

[emilper]

Glorified ? How about "highly sophisticated" ? Even a nuclear submarine is powered by a "glorified steam engine".

Re:Power?

[FooAtWFU]

Not really. It's good, proven technology. It is simple, with just a few moving parts that all move continuously in the same direction. It scales up very well: you get one big expensive steam turbine and you can point a boatload of cheap mirrors (/heat sources) at it. It takes advantage of some of the exotic properties of one of the most fascinating chemicals out there: Water. It produces no toxic waste to dispose of (not from the steam-engine part, at least... maybe a few lubricants you'll need to recycle, but that's pretty trivial). It doesn't distribute well (if you're piping hot working fluids around from one site to another, the heat tends to leak). Photovoltaics have it beat there, but they can't use all the spectrum. I suppose it doesn't scale down spectacularly well either; you might have better luck with a Stirling engine (more moving parts, though).

I don't see the big "kludge", myself. Is it the part where you hook it up to a bundle of wire and spin it around in a magnetic field to make electricity? I think that's pretty awesome too; you can move a whooole lot of electrons that way.

Re:Power?

[jfengel]

The kludgy bit is that you go from heat->motion->electricity, with losses at each step. Maybe that's just good reuse, having already debugged both steps pretty thoroughly.

But after a century or so of power plants, it's starting to feel like optimization is no longer premature. The power plant is the very center of a tight loop, and worth optimizing.

Unfortunately, any time you replace a well-understood legacy system with a new one you get bugs, and the whole heat->electricity thing isn't yet anywhere near well library-quality code. It actually turns out to be less efficient, not more. But as a programmer you look at the inefficiencies and figure there's got to be a better way.

Re:Power?

[Yvan256]

Isn't it extremely dangerous? It could escape the power plant by pretending to be a cop and then go on a killing rampage.

An interim solution

[CRCulver]

It will need to do accelerated testing to show the impact of decades of wear on the new mirrors in desert conditions.

Solar panels don't have to last too long when fusion is only thirty years away, am i rite?

If Google would run candidates....

[jnmontario]

I'd vote for them. They (corporate entity) seem to have a better head for good governance and forward thinking than any politician I've had the 'pleasure' of running in my province.

meanwhile....

[Luke_22]

...Italy just dropped all economical support to solar-termal energy.
photovoltaic still has subsides, but no more for solar-thermal.
and we were the 3rd country with most solar thermal in europe untill now.
...

Google and Govt talk:

[140Mandak262Jamuna]

From the article:

Weihl said Google had not intended to invest much more in early years, but that there was little to buy. "I would say it's reasonable to be a little bit discouraged there and from my point of view, it's not right to be seriously discouraged," he said. "There isn't enough investment going into the early stages of investment pipeline before the venture funds come into the play." The U.S. government needs to provide more funds to develop ideas at the laboratory stage, he said. "I'd like to see $20 billion or $30 billion for 10 yrs (for the sector)," Weihl said. "That would be fabulous. It's pretty clear what we have seen isn't enough."

Google: "Government, please throw in some 20 or 30 billion dollars to into solar energy research"

Govt: Nah, deficits are high. We dont have money. It should be done by the private sector. 20 or 30 billion dollars is too much way too much we cant afford it It is not a trivial sum like 780 billion dollars to clean up after wall street greedy moneybags. Tell you what? Grow too big to fail. Then come back asking for a couple of trillion dollars. Then we will be able to do it. OK?

Talked to a friend at Google about this

[Thagg]

A friend of mine who worked at Google at the time had clearly been involved in this project (although he didn't tell me...exactly) We were discussing alternative, sustainable power, and I've always been a fan of solar thermal -- he described in way more detail and depth than I thought possible the resource limits we'd run into if we tried to power America by solar thermal -- in particular the current mirrors in the prototype plants use a huge amount of aluminum, and scaling those plants up to make more than a rounding-error of our energy needs would take way more aluminum than we could forsee having. Plus, of course, it takes a ridiculous amount of electricity to refine the aluminum in the first place.

I was rather surprised, and checked his math...which was pretty accurate. I do think that other alternatives to aluminum are practical, and Google's going there.

Thad

Use More

[Wardish]

Since I'm fond of flights of fancy...

Beam splitter, Fresnel lens, simple prisms, whatever works to separate different parts of the spectrum. Thermal energy going to thermal power generation, the rest going to solar cells that efficiently utilized that particular part of the spectrum.

The rest of course is the engineering.

Did Google misinterpret

[Anonymous Coward]

Did Google misinterpret the reason that Oracle bought Sun?

Re:Refining Aluminum?

[Savior_on_a_Stick]

Aluminum is refined from bauxite and takes a huge amount of energy to produce initially.

It is extremely rare to find it in free form.

Re:Where did they get the people?

[AK Marc]

They hire professional problem solvers. They just redirected a small portion of problem solving power to this.