MIT stacks solar panels like pancakes, increases their power output by up to 20x
extremetech.com
extremetech.com
(2) any geometric arrangement of solar cells that includes the 'third dimension' will shade an area behind it that is proportional to the height of the stack, that dead zone should we taken into account when calculating efficiency. In other words, the total surface area of the stack is what counts, not just the footprint. And placing these panels at a 45 degree angle actually decreases their efficiency. The total output will go up but the cost per Watt will go up as well (and it will go up more than the power output).
Why is it that every announcement about solar panels is always accompanied by either gross exaggeration or inaccuracies? This really gives a bad rap to the whole renewables industry, if I had a 1% increase of efficiency on PV panels for every 10x announcement I'd be at 300% efficiency by now...
MIT really hurts their reputation with announcements like these.
If you want to do a really nice job in measuring panel efficiency you set up equal area arrays (equal areas of cells with the same packing density on the cells themselves) and measure them side-by-side over the course of a year.
The 'control' should be a non-tracking array that you correct for elevation once every month and that points due South on the Northern hemisphere and vice versa.
It can be measured that way, and for many practical applications it is. Solar tracking a small surface area is easier than tracking a larger one, water management of six solar panel mounting penetrations through a roof is more efficient than for sixty penetrations.
Engineering efficiency shouldn't just be measured in a lab.
However, based on the article one standard measure of efficiency does seem to be addressed, kw-hrs/panel/day because of the claims of better performance at low incident angles.
Mounting solar panels using aluminum stand-offs is very easy and tried-and-true tech, setting up a tower like this up on your roof is not going to be nearly as easy as you think because there will be a lot of force trying to pull the whole assembly off the roof whereas conventional installations have to deal mostly with compressive forces (which is a whole lot easier).
And that's with the wind directly on the panels. The arm of force from such an assembly will be proportional to the height of the total stack and will be quite impressive. You'll need - at a minimum - braces and guys behind the stack if you don't want it to take off in the first storm.
Those sixty penetrations have a nice side effect of spreading the load evenly over the roof so that you don't need to design a structure behind it to deal with the force of the whole array.
http://www.freepatentsonline.com/7045702.html
The windmill scene is even more outlandish than the solar panel people, and the funny thing is all the weird ideas come full circle every decade or so.
Though part of me is tempted to build it to see what it would do.
Gravity loads are not typically the primary concern for conventional solar arrays, uplift is. Deadload is an effective means of counteracting uplift. Lateral force per unit area on a roof mounted object is significantly less than uplift (and lest we forget, the length of the lever is mid height for a uniformly distributed load.
Roof design is like computer security, every penetration in the membrane is a potential site of infiltration. More holes = more probability of bulk water in the building. One of the worst places to direct bulk water is around electrical equipment.
Flat layouts create additional problems in regards to maintenance. Access to penetrations and the the membrane will be more largely dependent on the ease with which panels can be demounted, in addition access to panels in the center of the array may also be dependent upon demounting panels - one probably does not want to walk across an energized electrical grid and solar panels tend to be energized when the sun is shining. Hail and wind borne debris are banes of roof mounted arrays and flat panels are generally at significantly greater risk to both.
One ignores that solar panels are a part of a building's electrical system at their own peril, quite literally. A flat scheme not only requires significantly more conductors, it also increases the area over which the potential for a lethal shock is high quite significantly - talk to a fire fighter sometime about how one plans for operations around solar arrays during daylight hours.
It may be reasonable to consider that a longer diurnal period in which significant power is generated will reduce the need for storage capacity - lead acid batteries, hydrogen tanks, and other standard storage methods all come with significant life safety concerns.
Finally, packaged technical solutions tend to provide better quality, lower cost, and a higher degree of safety compared to custom site built solutions. One can assemble a tower in an enclosed factory using industrial Q.C. processes. Doing so on a pitched surface ten meters above grade as light fades at the end of a cold winter day is a bit more unlikely.
This is not to say that the claims of the article are accurate. But from a building science standpoint, a vertical array is a significantly better option than a flat one.
Contrast that with the freestanding arrangements that I've seen that were literally nothing but trouble. Especially on trackers (see picture above, and that was (and is) definitely one of the better trackers on the market).
Most rooftop solar installations are retrofits, so the building is not designed with a solar installation in mind. The general way of doing this is to mount an aluminum subframe to the roof using stand-offs and stainless steel bolts. Copious butyl sealant will take care of any water worries you might have, inspections are typically done every five years. Bigger worries are having to - in North America at least - reshingle a roof.
The mounting is done in such a way that it meets up with the studs underneath the plywood in the roof. If you wanted to do that with too few attachment points they'd have to be so strong that the joist underneath would be the limiting factor.
Electical wiring is run along the back of the array, then passes to the aluminum frame and slips under the eaves into a conduit into the premises (which is then foamed up to avoid moisture and air from getting into / out of the building).
The aluminum subframe is then grounded to avoid lightning troubles. (if a panel or the subframe gets a direct hit it can be very interesting if you forgot this step).
The whole installation sits only a few inches away from the roof itself, very little space for the wind to get a grip and you can easily limit that further with a skirt around the edge.
Anybody that messes with a solar system when the sun is out will be able to tell you tales of arc welding with the sun.
I agree with the access issue, that is one of the reasons why the more clever arrangements include a hinge at the lower end so the panel array can be stood up in case there is every any trouble.
Commercial buildings are not just scaled up houses in the same way that Github isn't a scaled up version of what someone created for Geocities. Thus, the reason globs of magic goop are not a viable solution to my bulk water management concerns. Bad things can happen when you have to put buckets in a server room or surgical suite.
I suspect you might be surprised at the pathways by which bulk water can enter a building without causing structural failure when a hurricane passes by.
I will add that your understanding of roof uplift forces is a bit more intuitive than scientific. Objects are sucked off, allowing air to move beneath them has the potential to reduce uplift because the pressure differential is reduced - something one is allowed to consider when designing tile roofs.
Of course, in that context you are right.
And yes objects are sucked off due to the pressure differential but objects are also blown over and this structure looks like a very efficient wind-trap to me which means it will incur significant loads at high wind speeds, far higher than panels mounted in a regular fashion with those forces concentrated in a much smaller area.
Btw, there are millions of roof mounted panels in Germany because of subsidies, the biggest problem seems to be that the panels live substantially longer than the roofs they are mounted on...
Why does 45-degree panels reduce efficiency? I mean, it might reduce it when the sun is overhead, but solar power is generally measured in the total wattage it produces over a long period.
So how can they make this claim today? Besides that, that's pretty easy to calculate, all you need is one of these: http://www.solarpathfinder.com/index
It will show you quite precisely what the effect is of the object you place in the path of the suns rays.
As for the 45 degree reducing effeciency: solar panels produce power that varies with the angle of the incident light. So if you place one panel at exactly 90 degrees to the incident light (for both 'tilt' and 'pan' or elevation and azimuth if you want to be precise) then you have a panel that performs at maximum efficiency (anywhere from 15 to 30% depending on how much you were willing to pay for your cells). If you use two panels on a 45 degree angle relative to each other the amount of output per panel drops but since you have twice the active surface you can now combine the output of both panels. Since one panel is 100% and the output varies as the co-sine of the angle you now get 0.7 times the output per panel. This gives you 1.4 times the output of a regular panel for a cost-per-watt of 2.0 times the regular panel.
So if you have a square meter of space, you get 'x' Watts/m^2 by laying a flat panel on that space and you can get 'some number higher * x' watts/m^2 by stacking a bunch of panels up above that square meter of space. Forget cost of panels, that's not mentioned. All that is mentioned is you can get more Watts per square meter by stacking more panels. You may be paying 200x more money to get 20x more power.
Each individual panel will be less efficient because it is not normal to the incident angle of the sun's rays, however you make up for it by having so many panels that you end up with net more energy. (Most of those panels are wasted though, by not pointing directly at the sun.)
It's kind of a dumb "discovery". As a thought experiment, you could even stack your accordion of panels up several miles if cost is no concern. You'd even be able to generate power after sunset (at sea level) :)
I know: http://pics.ww.com/d/90257-1/IM000384.JPG ;)
But even that is not the most cost effective, and it is somewhat maintenance intensive compared to a static array.
This was in Northern Ontario.
They're ~20% percent efficient in terms of the light that hits the ground they're mounted on.
But part of what makes a difference here according to the article is that it captures more light at times where traditional cell arrangements does really poorly because their arrangement means less light actually hit them.
E.g. early mornings, late nights, when most of the light won't hit a traditional arrangement.
So this only works if you are the only person in the area who uses solar cells, and if you only have a single row of them.
In certain applications it makes sense - remote isolated installations. But if you are trying to cover your entire roof it makes no sense (you can only use these in a single row, not an entire roof), and if your neighbor wants some light for his garden he won't let you add so much shade.
On stuff like this, though, it's hard to tell if it's the "journalist" screwing it up, or if the original data is flawed.
Edit: to be clear, I think it's because MIT's PR departement is hyper-inflatory, which you can hardly blame them for. But maybe due to it's reputation, the regular press picks up on like it's golden.
Lets say the current design takes up 1 cubic meter. (I can't tell from the photo).
Lets say you have a 10,000 square meter roof, 100x100 meters.
Does this mean you can put 10,000 of them up there or a number far lower since they need space in between them to let the light in?
The article isn't clear whether an arrangement of 20 of them is more efficient then the space required, but it does hit at it when it says: Then, to confirm their model’s predictions, they built and tested three different arrangements of solar cells on the roof of an MIT laboratory building for several weeks.
> So far, the team has modeled individual 3-D modules. A next step is to study a collection of such towers, accounting for the shadows that one tower would cast on others at different times of day. In general, 3-D shapes could have a big advantage in any location where space is limited, such as flat-rooftop installations or in urban environments, they say. Such shapes could also be used in larger-scale applications, such as solar farms, once shading effects between towers are carefully minimized.
Which it appears now has been debunked.