Slashdot Mirror
MIT Solar Towers Beat Solar Panels By Up To 20x
An anonymous reader writes "A team of MIT researchers has come up with a very different approach to solar collectors: building cubes and towers that extend solar cells upward in three-dimensional configurations. The results from the structures they've tested show power output ranging from double to more than 20 times that of fixed flat panels with the same base area (abstract, full pre-print). The biggest boosts in power were seen in the situations where improvements are most needed: in locations far from the equator, in winter months and on cloudier days."
Picture available here. It's a solar pancake!
Use 50 times as many solar cells, and OF COURSE you'll get more power out.
So, MIT has basically recreated what a 7th grader has previously done.
"National Security is the chief cause of national insecurity." - Celine's First Law
Big surprise that structures in volumetric configurations ended up being more efficient at gathering energy... considering plants have known this since they left the seas hundreds of millions of years ago.
while(1) attack(People.Sandy);
As seems depressingly common in science journalism, they vaguely mentioned the existence of a paper, but don't actually give the title or (dare we hope) a hyperlink to the paper. At least they did mention the name of the journal it was published in.
In any case, the paper is "Solar energy generation in three dimensions." If you're at a university with a subscription the official version (not open-access) is here. There is also an open-access preprint version at the arXiv.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Most people use solar panels because they can be comfortably put on rooftops. If someone has enough room for these 3D structures they could just install a Sun tracking system that's even more efficient.
The cost/watt is higher, this is DOA I dare say.
They're simultaneously saying that it's most beneficial for northern/southern areas where daylight is diminished and that it's a more compact arrangement of cells.
Those two don't go together well... Most northern and southern areas have very large open areas due to having low overall population density.
Cost/Watt is all that matters in most areas for solar panels, Watt/weight in the rest. I can't see this being of use except in powering small devices
20x output (compared to a flat panel with the same footprint).
Not really news. This is like excitedly proclaiming that a 20 story building has nearly 20 times the floorspace of a single story building with the same footprint. Uh, no shit? (Or that a 20 story building receives more insolation than a 1-story building; hmm, you think maybe it has a lot more surface area?) I also like that they hand-wave away the fact that it costs significantly more per unit output by saying that cells are getting cheaper. Great.
Not that there aren't uses - it absolutely makes sense to go this route where you have limited footprint space - but it just doesn't seem at all revolutionary. I guess if you tack the letters M-I-T onto a press release it instantly becomes newsworthy.
Solar Towers are actually pretty well defined.. nor is this a Solar Power Tower, nor a Solar Furnance, or anything else.. this is just a stack of Solar Panels..
looking at their configuration all I think of is how much of a mess it will make when it blows of my roof. there's a reason people install panels flat to their roofs. not to mention the added weight and live load reaking havoc on a standard truss roof not designed for that load configuration.
Someone figured out a long time ago how to build the optimal structure for turning sunlight into energy. The next step is to figure out how to get these solar arrays to utilize their own power generation to improve their performance. To grow....
This is like saying "OMG, new invention allows many people to live on the same plot of land while still having their own space and most of the privacy they would have if they all owned their own houses... it's called... an APARTMENT BUILDING!!!"
This does not constitute news. This will not revolutionize shit. You want efficiency? Build a transparent plastic radome, (a geodesic dome transparent at the wavelength you're interested in collecting) and put a parabaloid dish inside, focus the sun's energy to the highest degree your energy collecting material of choice can tolerate, and place that where the subreflector would go, or use a Cassegrainian or pseudo (offset) Cassegranian system, with a proper subreflector, and auto-track the sun. Automatic tracking is almost trivially easy, but if you don't want to do that, you can always use math and solar astronomy to point the dish where you know the sun is going to be. The druids did it, surely we can... and that's more efficient use of PV cells (or whatever) than building a wavy-looking tower out of them.
By the way, if you don't want to move the collector, you can always create a mirror array that will track the sun, and use the array's clever layout of cells to ensure the energy always hits the stationary, ground-mounted collector, in the event you're using water instead of PV, and it turns out to be heavy.
Anyway, this is not news, please stop pretending it is.
Quick, someone alert all of the major energy companies so they can buy up the patents and sit on them for eternity!
For large scale solar to electric, the best setup is a big array of mirrors bouncing light up to a tower filled with molten salt. Excess amounts of molten salt can be stored underground. The salt is stored at very high temperatures, way above the boiling point of water. Then that heat is used to boil water and generate electricity 24 hours a day, it keeps running all night, and even for 3 or 4 cloudy days. All the technology is invented and tested. People just need to build the things out in the American southwest and start generating electricity already!
With the zigzag tower configuration, it's just more nooks and crannies for snow to collect in and block even more sunlight.
I'll wait for Zero Point Energy.
In other news, a high rise building holds more people than a one story house!
I live against a hill with a small wooded area. My front property is open and there's a space I could put a tower like this where the sun will hit it better then on my roof.
~~ Behold the flying cow with a rail gun! ~~
Of course it will produce more power for the same base area since there are more solar cells. Sadly it doesn't improve solar cell efficiency at all. It is more like curlers (the kind worn in a woman's hair) for a house (IMHO). I wonder how it compares with the same number of solar cells in a sun tracker setup?
Bad form to reply to oneself, but I found the discussion of the methods I believe the article was referencing in this comment on the Watts Up With That article.
Python coder | PyQt Applications | Writer
You're still limited by physics, and ultimately even with an advanced 3D layout only so much sunlight hit's every square meter. Even if we could magically capture 100% efficiency it will never touch other forms of power generation for the same density, and will require large tracts of land for the same effect.
This is pretty neat, but a far cry from ever solving our energy crisis.
Tree designs, especially those requiring vast amounts of sunlight for photosynthesis, as opposed to say conifers, would seem to make a logical design template here.
Is anyone testing anything like this?
The article is a bit obtuse about where its coming from, these configurations don't make the panels themselves more efficient but instead are much more standardized and have a greater energy density per unit of land/roof they take up. There are definite applications, while your cost per panel watt goes up (because you need more panels) these configurations, mass produced, could bring down installation costs considerably. Instead of having to have a roofing crew come outspend a whole day custom installing the panels on your roof, two guys stop by on one day, pour a base and run some conduit. A week later another guy or two come with the solar package, bolt it down and connect it to the base, and expand it. 8-12 hours and 6 guys reduced to 2-3 guys and 3 hours. I like the idea of the car charging stations as well, no cabling, no customized mounts, just drive up to one of these things, grab the cord and plug it into your vehicle.
I affixed high-efficiency monocrystalline silicon PV cells to the aluminum cans and used pizza boxes strewn about in my yard and now my trash is generating electricity! Electricity from trash, wow!
That was the turning point of my life--I went from negative zero to positive zero.
In the place where this would presumably be most useful, where horizontal space is at more of a premium than vertical space, it could well be illegal due to solar access laws. Here in Denver, it has led to some odd-looking asymetric second-stories when they are added to existing bungalows -- where, say, the left half of the A-shaped roof has a shallow or near-flat slope and the right half has a steep pitch.
Is there some interesting physics going on, or is this just taping a bunch of cells vertically to intercept more light at a low incidence angle? Surely,that can't be all there is to it, right?
The problem of this approach is that it does not scale.
In real size photoelectric systems, the cells are in series. And this arrangement cannot cope with different angles, and partial panel shadows. The whole output of the string decreases to the weakest in the chain. Fail.
aaaaaaa
Solar shingles have been around for quite some time.
http://www.google.com/search?&q=solar+shingles
If our conversation is going to follow the pattern of a typical slashdot discussion thread, you will now need to retroactively define the terms "major", "mass produced", and "smaller" in such a way that you can insist that I am not only wrong, but also a smelly hippy that likes Hitler.
Am I the only one thinking of the next posting "Dysfunction In Modern Science?" on /. today right after this one?
Or the one a while back about a child who made a TREE with solar leaves that performed better but it turned out he had it all wrong and the media hyped the BS?
For me, in winter I have a 78 degree perpendicular with the sun-- that is nearly vertical in which case a bunch of staggered 45/-45 degree panels would work and the lower ones would get a lot of sun considering they are supposed to work fine with 10 degrees off center and the snow would reflect light towards them as well....
Problem is the sun is near 0 degrees in the summer. so half the panels would get jack.
CIGS panels handle diffuse light better; get those cheaper as many of us have more of a 2/3 cloudy year. Or IR light since most the IR light is not impacted by clouds. Or various coatings I've read about that keep more light from bouncing off the panels...
Or how about a small scale low pressure steam turbine? heat still beats out PV by a huge factor if you can extract energy from it effectively enough you can beat PV not to mention that heat storage works better than batteries.
Democracy Now! - uncensored, anti-establishment news
Oriented or tracking panels produce only around 20-30% more energy than flat horizontal panels, when averaged over a year over most of the USA. This because much of the insolation is diffuse. NREL has maps that show the measurements at http://rredc.nrel.gov/solar/old_data/nsrdb/1961-1990/redbook/atlas/
Put some white space to the left of the text on your page. It's a real pain to read words that are jammed right against the edge of a monitor.
It's a stupid concept. If I had been asked to review their paper I would have recommended not publishing it. Here's a link to their abstract:
http://pubs.rsc.org/en/content/articlelanding/2012/ee/c2ee21170j
"We demonstrate that absorbers and reflectors can be combined in the absence of sun tracking to build three-dimensional photovoltaic (3DPV) structures that can generate measured energy densities (energy per base area, kWh/m2) higher by a factor of 2–20 than stationary flat PV panels for the structures considered here, compared to an increase by a factor of 1.3–1.8 for a flat panel with dual-axis sun tracking."
Yet they admit the following: "The increased energy density is countered by a larger solar cell area per generated energy for 3DPV compared to flat panels (by a factor of 1.5–4 in our conditions)...." IOW, they need a larger cell area by a factor of 1.5 to 4 to generate a given amount of energy than would be needed by a flat panel! So they admit their concept is a factor of 1.5 to 4 less efficient than a flat panel. But how can that be true, if their design generates significantly higher energy densities?! Probably they're merely generating higher peak energy densities at certain times and in certain localized regions of their solar panels, while their average energy density (averaged over their entire solar-panel area and over the entire operation time) is lower by a factor of 1.5 to 4.
All they're really claiming is that they can reduce the variability of the power generated by stationary solar PV panels:
"3DPV structures can mitigate some of the variability inherent to solar PV as they provide a more even source of solar energy generation at all latitudes: they can double the number of peak power generation hours and dramatically reduce the seasonal, latitude and weather variations of solar energy generation compared to a flat panel design."
But is a reduction of variability worth sacrificing total energy production by a factor of 1.5 to 4? Not likely.
If you have one solar cell and want to install it at a particular latitude, there is a specific orientation that will produce the most energy over a day. All cells should be in that orientation to maximize their individual energy production. This is what leads to fixed installations being large flat arrays with every cell in the same orientation. If you have 100 cells it is true that you can put 50 of them facing more to the easy and 50 more to the west, and you may be able to do this with less land area than all 100 laid out facing south. It will also be true that it produces LESS energy than the large flat array of 100. We can also add a tracker to the flat array and gather more total energy than EITHER fixed structure through the day.
By far, the cost of the cells is more than any other component, so each one needs to produce the maximum energy and that leads to all of them being in the same orientation. A sun tracker can be cheaper than adding additional cells, so that's a good way to increase output. Building upward is an obvious way to get more cells per land area, but this is both obvious and silly. This is the second time I've seen an article talking about power per land area instead of power per cell area (actual cell efficiency) or watts per dollar.
Beyond that, the control/inverter is another big cost. Each cell has an optimal Voltage/Current operating point given the angle and intensity of light falling on it. As soon as you put 2 strings of cells in different orientations, they need to operate at different voltages or currents for optimal conversion efficiency which will require multiple inverters rather than just a bigger capacity one. This is a second way these non-optimal systems cost more.
I don't have the sources at the moment, but I remember reading that Germany required the power companies to buy back power at 10X the retail rate as a rather extreme subsidy. At such subsidy levels it might make sense to put solar panels in at the poles, and not just for a station only occupied in the summer...
I don't read AC A human right
So now they're building big solar collectors that cover parking lots so people can park under them.
I read about 'solar roads' a while back. The idea is that they make the roads transparent and put solar panels under them to collect energy, and have light-up road signs and such.
My thought was - Why not cover the roads? Do they have semi-transparent solar panels? Even if not, I'm sure they could stagger the panels so some of the light still gets to the road- while conveniently blocking 'sun in your eyes' type problems and wet, slick roads, not to mention if you're further north the very angle that you end up mounting the panels at for maximum average efficiency should make clearing them a breeze, and better yet, no need for plowing snow! I've read that, for the most part, with a decent slant it's still energy positive to use electric heating to clear panels when necessary. The idea isn't to melt all the snow-just enough that you get a liquid layer between the panel and the snow, causing it to just slide off.
I agree - We're a long ways from needing to increase density in this fashion.
I don't read AC A human right
Yes, I did RTFA. Like I said - we're not limited on space for installs. We DO face the problem that solar energy is barely competitive with grid power even with 50% or higher subsidies, competing against retail price for electricity, not utility price. It's noted that these are even more expensive for the energy, presumably because individual panels aren't ideally placed - but are placed such that they tend to pull more power in alternative conditions because some of the panels have better angles for such times as when the sun is near the horizon.
As Icebike said, these designs are probably no more productive than variable tilt solar panels, and they require a heck of a lot more structure than mounting an array on a house.
I don't read AC A human right
I'm also pretty sure that there are a few 2nd or 3rd year EE students who could solve this. I would think it's just a matter of building some basic logic circuits to bypass unneeded cells.
You will also need controllers and stuff in addition to the $50 electric motor. If you are handy, you build it out of spare bits from the computer junk box, but if you are normal consumer, you will go to solar store and want a ready-made system. That will not be $50. Adding one zero will not be enough, I have been asking around.
A Motor needs servicing and electronics to run it, more mechanics which can break down, etc. When servicing company needs to go on site to fix it, it will cost far more than $50. It only makes sense for a very large system, where single motor can tilt large number of panels connected with rods or such. Panels are around $900 per 5 square meters (approximately 1kW peak power), so the focus is shifting into how to install these cheap in fixed angles, such as going from aluminum structures to plastic molded things etc. It simply becomes cheaper per kWh produced. For example, normal roof material and installation cost is around $1000 per square meter, 5 times the cost of panels, to give some relationship. While roofing is more expensive, panel installations need to carry similar loads, with the difference mostly being not needing to keep rainwater off.
Optimizing for the late evening will not create such a big benefit, as you will loose half of the panel surface when they start shading each other, and atmosphere will start eating into power quickly at low angles (in particular when some scattered clouds are blocking the sun). There are panels with surfacing which optically collects the sunlight from low angles more efficiently, but what really matters is cost and site. You win a lot more by installing at a good site and optimal angle than installing fancy systems to collect the last 10% of power, and getting panel area at the best times of day will buy a lot more kWh per $.
For small sites, tracking systems tend to be very expensive. I asked around for prices, and got tracking system costs supporting 5kW of panels, and the tracking system cost around $4000 with installation and panel supports (which are more expensive in this case). Enough to buy installation supports and panels for at least 3kW more. Group these extra panels into morning and evening sun and you will same or more power, but will not need servicing. Also, If you need to replace the tracking system, say, every 10 years, it will eat into your produced energy quite a bit. This is the same for small wind turbines, get more rotor diameter buys a lot more power than having 5% more efficient blade design.
And install a reflecting pond in front to mirror more sun into the panels, looks nicer, even if the amount of extra energy is negligible :)