Solar cell
in [Design]
|
From: Sjouke Burry on 24 Apr 2010 23:11 Greegor wrote: > On Apr 24, 8:50 pm, mpm <mpmill...(a)aol.com> wrote: > MPM > I can also state with certainty, that WV > MPM > has the most (shall we say...) innovative > MPM > way to distribute food to the poor. > > Are there any squirrels left in the trees? A well prepared squirrel is much better for your health than a Big Mac..... (and tastes much better too....)
From: krw on 24 Apr 2010 23:57 On Sat, 24 Apr 2010 18:54:35 -0700, "Joel Koltner" <zapwireDASHgroups(a)yahoo.com> wrote: ><krw(a)att.bizzzzzzzzzzzz> wrote in message >news:pi67t5pmefeen4jn1i5c96v6rd2jmqt97f(a)4ax.com... >> PVs have had the same time frame *and* technology advances and *still* are a >> couple or orders of magnitude short of break-even. > >Sure, one can't really schedule technology breakthroughs. But they've had free access to every other semi breakthrough and still can't come within two orders of magnitude of interesting. >The history you've >cited there certainly ought to be included as part of whatever policy >considerations fund PV R&D, certainly (or just outright purchases, although I >think Don's made a pretty good case that that isn't really helping anything). "Green" research support academics. That's enough for Obama. >>>It's interesting to consider what technology would become porminent if, >>>somehow, all the gas stations and oil refineries suddenly disappeared over >>>night -- bet we still had all the technological know-how as of 2010. I'm >>>thinking the replacement infrastructure might include a lot more diesel >>>automobiles... >> A silly what if. So what? > >I'm pointing out that what happens to be cheapest today is often as much a >function of history as it is of the actual cost of the technology *today*. >This fact is one argument for government policies that attempt to change >behavior away from the status quo. It's not working. Diesel has been around as long as gasoline. >> You do know that IBM wasn't allowed to bid on the DARPA contract because >> they >> already *had* such a network (several hundred nodes, worldwide). > >I didn't know IBM wasn't allowed to bid. >I actually had a VMNet e-mail address back in 1989 -- several years before I >had an Internet address. You mean VNET? I don't remember when I first got email access but we were sending data around in the mid-'70s. "WideBand" was the first network, but batch oriented (used mainly for sending chip data around the company). VNET came about shortly after, by 1980, certainly. >> After forty years of trying (at least), what makes you believe it's possible >> to jump a couple of orders of magnitude ($/W). > >One order of magnitude would be plenty though, in the case of PV. With subsidies, perhaps. >I suppose that Moore's Law makes me think that many technologies can be >cost-reduced/performance-enhanced by orders of magnitude over time -- it's >just that the time period is sometimes much longer than the 18 months it is >(or at least was) with transistors/IC technology. But PV has already had the benefit of Moore's law. You can't count on squaring Moore.
From: Paul Keinanen on 25 Apr 2010 06:15 On Sat, 24 Apr 2010 18:54:35 -0700, "Joel Koltner" <zapwireDASHgroups(a)yahoo.com> wrote: ><krw(a)att.bizzzzzzzzzzzz> wrote in message >news:pi67t5pmefeen4jn1i5c96v6rd2jmqt97f(a)4ax.com... >> After forty years of trying (at least), what makes you believe it's possible >> to jump a couple of orders of magnitude ($/W). > >One order of magnitude would be plenty though, in the case of PV. As long as PV cells are made by cutting wafers out of huge pure silicon crystals, the price is not going to go down significantly, no matter what subsidies are used. Perhaps someone invents how to grow the cells on a flexible base material that can be processed roll to roll just like paper processing, the price might drop one or two orders of magnitude. Just cover the roof and walls of a building with sheets of flexible solar cells. >I suppose that Moore's Law makes me think that many technologies can be >cost-reduced/performance-enhanced by orders of magnitude over time -- it's >just that the time period is sometimes much longer than the 18 months it is >(or at least was) with transistors/IC technology. Much of Moore's Law has simply been the line width reduction of individual components. Much more functionality has been obtained over the years from a single wafer and hence the cost of a single chip has been reduced significantly. In a solar cell, you are interested in having a large surface area as possible and there is no benefit from using smaller line widths. Ultimately the solar cell cost depends on the wafer processing costs. Cutting the wafer into multiple chips, reducing the chip cost, but at the same the power output also drops proportionally :-), thus, there is no point in cutting the wafer and hence the Moore's Law does not apply.
From: krw on 25 Apr 2010 11:46 On Sun, 25 Apr 2010 13:15:35 +0300, Paul Keinanen <keinanen(a)sci.fi> wrote: >On Sat, 24 Apr 2010 18:54:35 -0700, "Joel Koltner" ><zapwireDASHgroups(a)yahoo.com> wrote: > >><krw(a)att.bizzzzzzzzzzzz> wrote in message >>news:pi67t5pmefeen4jn1i5c96v6rd2jmqt97f(a)4ax.com... > >>> After forty years of trying (at least), what makes you believe it's possible >>> to jump a couple of orders of magnitude ($/W). >> >>One order of magnitude would be plenty though, in the case of PV. > >As long as PV cells are made by cutting wafers out of huge pure >silicon crystals, the price is not going to go down significantly, no >matter what subsidies are used. > >Perhaps someone invents how to grow the cells on a flexible base >material that can be processed roll to roll just like paper >processing, the price might drop one or two orders of magnitude. Just >cover the roof and walls of a building with sheets of flexible solar >cells. > >>I suppose that Moore's Law makes me think that many technologies can be >>cost-reduced/performance-enhanced by orders of magnitude over time -- it's >>just that the time period is sometimes much longer than the 18 months it is >>(or at least was) with transistors/IC technology. > >Much of Moore's Law has simply been the line width reduction of >individual components. Much more functionality has been obtained over >the years from a single wafer and hence the cost of a single chip has >been reduced significantly. > >In a solar cell, you are interested in having a large surface area as >possible and there is no benefit from using smaller line widths. Actually, that's were many of the efficiency gains have been made. "Lines" are opaque. >Ultimately the solar cell cost depends on the wafer processing costs. >Cutting the wafer into multiple chips, reducing the chip cost, but at >the same the power output also drops proportionally :-), thus, there >is no point in cutting the wafer and hence the Moore's Law does not >apply.
From: Dirk Bruere at NeoPax on 25 Apr 2010 21:04
On 24/04/2010 00:57, Joel Koltner wrote: > Hi Keith, > > <krw(a)att.bizzzzzzzzzzzz> wrote in message > news:6ba4t5li7gihfbjgou9oi9ftop96dp9d0u(a)4ax.com... >> On Fri, 23 Apr 2010 09:38:54 -0700, "Joel Koltner" >> <zapwireDASHgroups(a)yahoo.com> wrote: >>> Lasers are perhaps a good example of technology that was incredibly >>> nichey for >>> many decades (being fragile and expensive) until someone figured out >>> how to >>> build a semiconductor version of one (making them cheap and >>> reliable), and now >>> the average person likely owns half a dozen. >> Lasers are a special case. The technology's usefulness was known long >> before >> it was practical. Gordon Gould mopped up on that delay. > > Well clearly photovoltaics is incredibly useful as well, with plenty of > immediate application. If someone figured out how to decimate the per-kW > cost of PV panels, their growth rate would immediately jump up into the > triple digits, I expect. Current prices are around $4/W Nanosolar, for example, claims they can manufacture for $0.70/W But since they are selling everything they can produce for a lot more than that, why would they drop the price? -- Dirk http://www.transcendence.me.uk/ - Transcendence UK http://www.blogtalkradio.com/onetribe - Occult Talk Show |