Slashdot Mirror
Google Files First Solar Patent, Builds R&D Team
bizwriter writes "Google has moved beyond investing and using solar power and has started on serious R&D work in the field. Its first patent application in solar energy technology just became public, and the company is staffing a new R&D group 'to develop electricity from renewable energy sources at a cost less than coal' at 'utility scale.'"
The Sun has a limited supply of hydrogen fuel. If we start depending on solar, in a few measly billion years we'll be depending on hydrogen imports from undemocratic planets. And the chance of a meltdown within 5 billion years or so is pretty much 100%.
all patent are not evil and this is exactly the kind of patent that the system was designed to encourage.
to develop electricity from renewable energy sources at a cost less than coal' at 'utility scale.'
This is not a good example of evil.
Jehovah be praised, Oracle was not selected
IP patents may be an oxymoron, I agree. But what they do with a patent is the salient part. Squash competition, or donate it to some patent freedom pool? I'll await further details.
"Tongue tied and twisted, just an Earth bound misfit
For the love of GOD, Slashdot, fix the login popup to STAY ON THE ARTICLE BEING READ.
What's the point of having a fancy Ajax Web 2.0 "popup" login if it just redirects you to the main page afterward???
What's the point of having an account if you post as AC?
Google's also been really involved in enhanced geothermal, one of my favorite techs. For those not familiar, here's a good rundown of its promise and pitfalls. Namely, it's baseload, works basically anywhere on the planet (all that changes is the required depth of the borehole), is renewable with virtually no environmental impact, and can provide thousands of times more power than we currently consume. At the same time, it's not widespread currently for one main reason -- not that it doesn't work, but that it doesn't work *reliably*. When you fracture the rock to pump in water to heat, the fractures go wherever the heck *they* want, and in many cases your water just seeps away (also, the fraccing can cause minor -- up to just over mag. 5 in theory in most places, lower in practice -- earthquakes). Here's one of the latest ways around those pitfalls, using a closed-loop system with an underground heat exchanger instead of fraccing a new reservoir. That also has the advantage (or potential disadvantage, depending on how you look at it) of not bringing minerals back up with the water.
I just invaded Grammar Czechoslovakia and duped Grammar Neville Chamberlain; now it's on to Grammar Poland.
I just hope they can maintain interest longer than they did with their Power Meter API, which was just deprecated.
Solar is usually about an order of magnitude more land-dense than hydroelectric (when you include the area taken up by the reservoir), and about on par with coal (when you include the land taken up by the coal mines required to fuel the plant and the few decades it takes life to regrow on them after an exhausted mine is abandoned)
Daily intermittence is readily countered by a wide range of factors.
* Thermal storage
* Pumped hydro energy storage (works with any type of power; already widespread in China for day/night demand averaging) (does not require a river or a large impounded area!)
* Integrated peaking (you already have a thermal power plant; adding a supplemental source of heat for when demand exceeds supply costs you almost nothing)
* The natural correlation between solar intensity and power consumption (night is off-peak, sunny days have more AC load, etc - -it's not perfect, but it's a nice start)
* Generation-source diversity (wind, solar, tide, wave, etc do not all line up with each other in terms of what generates when)
* Long-distance HVDC power transmission lets you take advantage of the fact that the sun doesn't set in all places at the same time.
* Smart grids and demand-flexible industry allow to shift when power is drawn to when it's abundant.
I just invaded Grammar Czechoslovakia and duped Grammar Neville Chamberlain; now it's on to Grammar Poland.
Unlike copyright, patents actually expire. In the extremely unlikely event Google coming up with something good, they get a short term monopoly on it. Good for them, and anyone else doing something in the physical world. A generation later, this is as good as public domain and anyone can implement it.
If they're locking away uber tech, it still doesn't matter. We miss out now, but our kids will have access. Unless you believe "my uncle's friend came up with a way to save fuel consumption but got bought out by $OIL_COMPANY" conspiracies.
Except that if you read the patent application, it should be shot down. The patent essentially claims "use a camera protected from heat and some image processing software to feed a control system with inputs to control heliostat mirrors to get an optimal image."
There is absolutely nothing novel about that concept, unless they are using a novel method of image processing (which the claims do not appear to indicate; they talk about "measuring bright spots" which is all a camera can do in the first place) or a novel method of keeping the camera cool (which the claims also do not indicate).
Linking image processing to a control system has already been done, and just because it hasn't been done "for a heliostat" doesn't make it novel. So I would argue that this is indeed just the type of patent that should not be allowed.
"There are a dozen opinions on a matter until you know the truth. Then there is only one." - CS Lewis (paraprhase)
Patenting physical devices that actually work is what the system was designed for.
Patenting software which physically is a long string of ones and zeros on paper is not what was intended. Symbols on paper are covered by copyrights.
Star Trek, there maybe hope.
all form of societal organization are evil then since they are all backed with threats of initiation of some kind of violence. It is a valid philosophical position but it is not a pragmatic one.
Jehovah be praised, Oracle was not selected
There is absolutely nothing novel about that concept, unless they are using a novel method of image processing (which the claims do not appear to indicate
See claim 5. The "based upon the determined error". Why patent the image processing to determine the error when it could and should be better maintained as a trade secret.
Patents that attempt to claim what is done are not valid. Patents that attempt to claim what is done, but in a much better, or even in a not-so-much-better but still novel way are patentable.
We hear about the patent system being broken, but my recent patent reviews have asked some good questions about what is being claimed. Since this is an application and not a patent, it is reasonable to assume it will receive some degree of examination related to the obviousness of the invention.
This post comes with a double-your-money-back guarantee!
Any offense taken to this post is at your sole discretion.
So the patent should be on that method. "Based on the determined error" is how every single control system in the universe* works. In fact, claim 23 says "determine the error by comparing images," which is still a "what" and is an obvious "what"; if they want a patent then patent the method they use to compare the images, not stating that they are going to compare them.
While I agree that often people don't actually look at the claims in a patent, this patent still doesn't claim any "how" but merely "what." In fact, even if you go to claims 24-26, you just see a calibration procedure that anyone would know: "put all the mirrors in the desired position to get the reference image, then move them to another position to get a reference undesired position."
There is nothing novel in this particular application, and it makes me ill to think that just because it's Google it will probably get approved.
*This is not the hyperbole for which you are looking.
"There are a dozen opinions on a matter until you know the truth. Then there is only one." - CS Lewis (paraprhase)
Lets say I invent a device that gets you from point A to point B. Someone else invents a machine that gets you from point A to point B in the same time. The fact that both devices do the exact same thing but in different ways ( car and flying car) doesn't make one patent violate the other.
Two pieces of software do the same thing( VOIP). These two programs run on different hardware, do the same thing differently but with the same end result. Even though the two programs have different ones and zeros the first was patented and the second is in violation. In what world does this make sense.
Star Trek, there maybe hope.
That's a nonsensical statement. The amount of power produced relates to both volume *and* head.
Day/night buffers (like those used in China, like those to pair with solar) are several orders of magnitude smaller than those used on the large-head large-scale conventional hydro projects. Which is why conventional hydro projects take months or even years to fill.
You don't need *any* natural body of water with pumped hydro (although it's cheaper if there is one). You simply need an altitude differential.
It's true in about 90% of the First World. And actually you've got it backwards; in most places, it's *warm* areas that use electric heat (since it's not used as often, and is more expensive/less efficient than gas heat)
I just invaded Grammar Czechoslovakia and duped Grammar Neville Chamberlain; now it's on to Grammar Poland.
Let me start you off with this. Tell me when you see the whole US clouded over.
A particular Ohio city is not an island (btw, there has never been a time in recorded history when a city in Ohio has had only five days of sun in 2 months). Ohio is connected on a grid to the rest of the country. The regional grids are increasingly being connected over longer and longer distances by high power runs. It doesn't matter if your particular area is cloudy, because somewhere else isn't.
A single wind or solar plant has a lot of randomness. A large number of them, spread out over a large region, have very little randomness. Also, FYI, but the time a power plant is down for is already built into its cost equation. That's known as the "capacity factor", and is a key element in economics planning for power plants.
Secondly, the grid *already* has to handle fluctuations. Not only fluctuations in supply -- yes, conventional power plants go down too, both for maintenance and for unexpected failures -- but even moreso due to demand. Demand fluctuates wildly, and a demand fluctuation is no different than a supply fluctuation. We deal with this by having "peakers" available. These are power plants that can rapidly scale their production up or down depending on the needs of the grid. One of the great things about solar thermal is that it basically comes with a built-in "peaker"; all you need is a natural gas burner, and you've got your backup at almost no extra charge. The turbines are already there, the transmission, etc.
Beyond all of that, please read the bullet points at the bottom of my last note.
I just invaded Grammar Czechoslovakia and duped Grammar Neville Chamberlain; now it's on to Grammar Poland.
Two other notes: one, the geothermal plant isn't creating the CO2, only providing an alternate route up to the surface. Such areas generally have high rates of natural CO2 seepage on their own. The depth of the reservoir and the strata above it affect how long it will be before the depletion of the subsurface CO2 will have the effect of reducing surface CO2 flows (anywhere from days to millions of years), but in the long run, any CO2 emitted by the plant is CO2 not emitted by other means. And two, some new geothermal plants are looking at using supercritical CO2 as the working fluid -- wherein the plants are actually *sequestering* CO2 in carbonates, as some of the injected CO2 will be lost.
I just invaded Grammar Czechoslovakia and duped Grammar Neville Chamberlain; now it's on to Grammar Poland.