« H2 Nanostorage | Main | Simulating Culture and the Ethics of the Off Switch »


pvconcentrators.jpgWe're all familiar to some degree with traditional photovoltaics, the flat dark-gray panels that generate electricity from light. Improvements in photovoltaic technology keep popping up, and solar will undoubtedly be a big part of the bright green future. But less well-known are photovoltaic concentrators, systems that use lenses and/or mirrors to boost the amount of light hitting a given patch of photovoltaic material, with the goal of increasing the overall output. Photovoltaic concentrators are commonly used on satellites to maximize the power-to-weight ratio, but have had more limited success for Earthly energy production due to the cost.

That may soon change.

The US National Renewable Energy Laboratory just held a conference on photovoltaic concentration technology in Scottsdale, Arizona. Among the announcements: the availability of PV concentration systems with efficiencies close to 40% at concentrated sunlight levels.

At the conference, NREL announced a new record efficiency of 37.9 percent at 10 suns, a measure of concentrated sunlight. Soon thereafter Boeing-Spectrolab, under contract to NREL and the Department of Energy, surpassed the NREL record with 39.0 percent at 236 suns announced at the European photovoltaic conference in Barcelona, Spain.

This should allow PV concentration system makers to reach their near-term goal of a $3/watt installed price, roughly competitive with other forms of power generation.

Most PV concentration systems are big, 20-35 kilowatt units able to track the motion of the Sun across the sky. The latest Wired, however, has an article about the "Sunflower Solar Concentrator," small enough to be carried by two people, and operating at 200 watt peak production. Rather than shifting the entire unit to track the sun, the Sunflower has a multitude of small mirrors that track together, reflecting the light into the PV system.

The head of the company making the Sunflower, Bill Gross, is better-known for founding a run of successful and semi-successful dot com companies, including Idealab. His goal is to make renewable power sufficiently affordable that rooftop installations are the obvious economic choice even for companies that aren't otherwise concerned about the environment.

It will be interesting to see if the ~40% efficiency photovoltaic systems announced at the NREL conference would work as well in the Sunflower design.


Listed below are links to weblogs that reference Concentration:

» NEW ENERGY CURRENTS: 2005-08-04 from Winds of Change.NET
Much like the thank-God-it's-finally-over Energy Bill, New Energy Currents for July is a little late. Hey, it's summer. New Energy Currents is a broad, monthly roundup of new development... [Read More]

» New Energy Currents: 2005-08-05 from Winds of Change.NET
Much like the thank-God-it's-finally-over Energy Bill, New Energy Currents for July is a little late. Hey, it's summer. New Energy Currents is a broad, monthly roundup of new development... [Read More]

» New Energy Currents: 2005-08-05 from Winds of Change.NET
Much like the thank-God-it's-finally-over Energy Bill, New Energy Currents for July is a little late. Hey, it's summer. New Energy Currents is a broad, monthly roundup of new development... [Read More]

Comments (24)

$3/watt is competitive? I thought big wind turbines are now $1/watt installed (eg $3M for a 3Mw machine). Anyone know?


It's important to note that the high efficiency rates achieved by NREL et al. are for TINY, one-off cells -- a few inches square. Translating this efficiency into production-scale modules is quite a ways off. Most experts believe that efficiencies will remain in the low 20% range for quite a while.

And yes: $3-per-watt solar IS competitive with $1-per-watt wind. Solar's $3 is a retail price delivered directly to the end-user on-site; wind's $1 is a wholesale price at the turbine, which must then be distributed through transmission lines before being sold to end-users (typically through one or more middlemen).

Yes, thanks Joel. But for apples-to-apples, wouldn't the solar figure have to include the control electronics and inverter? Plus the grid tie or battery storage interconnect? IIRC, domestic PV is about $4/W retail but more like $8/W installed with the electronics.

The NREL release claims a target of $3/W installed.


The free piston stirling engine being developed by NASA and DARPA has over 35% thermal-electric conversion efficiency and delivers AC power at any desired voltage. It is intrisically inexpensive, being nothing but steel and copper, and has enough life to be considered for powering satellites going to Pluto.

Day 4 Energy (http://www.day4energy.com/) is another company working on PV concentrator technology. They are claiming generation costs "competitive with fossil fuel based technologies"

Day 4 Energy (http://www.day4energy.com/) is another company working on PV concentrator technology. They are claiming generation costs "competitive with fossil fuel based technologies"

Re: NREL's $3/price installed. I'll personally pay the full cost of anyone who can get that cell installed in the next year at that price. (I'm tempted to say two or even three years, but I'm not THAT crazy.)

Point is, you can quote whatever price you want -- but until it's commercially available, it's all vaporware.

wimbi, that Stirling engine is great.  Now tell me where I can buy one for anything like $5/watt (perferably $2.50).

I'm serious about this.  I've got several great applications for an engine like that, and the best I can find is plans with partial kit parts available and lots and lots of machining to do.  $1000 for a 40-watt engine is not going to do it either.


Thanks, Engineer-Poet. Would that I could tell you where to buy one of those great NASA free piston stirlings for anything like $3/watt. I called up Sunpower.com and asked if they would sell me one, and they said-Sure, glad to, a mere $90K a piece!

Say What?

This is the result of a design to NASA space power specs, and of course, making them one a a time.

What would such a thing cost in the real world? I once worked with Bob Harkness, a very fine engineer and gentleman, who was at the time the chief design engineer at Briggs & Stratton. He told me the best way he had for guessing the mass manufactured cost of anything was simply to look in any catalog for something that had the same type and amount of materials and tolerances, and go with what it quoted as the price.

The Sunpower free piston engines have about the same materials and tolerances as a linear motion refrigerator compressor made by LG in Korea. Maybe a little more stainless steel. Those compressors are made by the millions, and have to cost somethng close to any other compressor or they would not sell. Say roughly$100 as a ballpark number.

That puts the 1kW engine-alternator in about the $1000 bracket. Now that is only the engiine-alternator, of course, not including the burner, or solar concentrator and whatever else makes up the whole system.

But to get there somebody has to have the guts to shell out maybe 20-50 megabucks to make a factory that will punch the engines out by the hundreds of thousands, if not milliions. And there's the rub.

I think there's some fly in the ointment for Stirling engines.  Energy Innovations started with a scheme for a solar-heated Stirling, 200 watts peak at a dollar a watt and cogenerated heat too.  The Stirling was removed from the schedule and is barely mentioned on the site any more; now it's concentrating PV.  And it's not just EI; other concepts fail to get to market too.

Stirlings are obviously more difficult to build to the necessary specs (power, cost, lifespan) than it appears, otherwise we'd see millions of them.


Thanks, again, Engineer-Poet, for your quite sane comment.

There is indeed a great big fly in the ointment of stirling engines, and that is that people do not distinguish between crank types and free piston types. Crank types that move the components by "normal" mechanisms - bearings, con rods, cranks, rod seals and lube oil DO NOT LAST. It is hard to find any honest data on mean time to failure for these machines, but it is in hundreds or maybe low thousands of hours. That is not enough for a real product. My little diesel tractor has about 8000 hours on it and runs fine. Even my old car has 3000 hours, and is still good.

But free piston stirlings have clocked tens of thousands of hours and are still running. They have also taken brutal shake tests, overpressure and such like insults. They have no mechanisms that need oil or wear out. That is why NASA is serious about putting them in satellites.

Trouble is, nobody notices this great big difference, after all, the name is the same, and so it is quite reasonable for casual onlookers to see the miserable record of crank stirlings (especially those designed by novice teams) and conclude that ALL stirlings are junk.

So that is why you don't see millions of them. All the money and all the hype has gone down the wrong road to a junk yard. Ask NASA. Ask DARPA

I still don't understand the problem.

Honda is working with Climate Energy LLC to produce a 1-kW cogenerating furnace system.  The cost premium over a standard furnace is about $4000.  If you could build a 1 kW Stirling for $4000, you could incorporate it into the base unit (no need for a second cabinet) and make the whole thing smaller, better and probably more efficient.

Honda is certainly a big enough company to design and build Stirlings in the volume required to make them economical.  They've got some of the best engine technology in the world, and have brought many sophisticated designs to market at a price consumers snap up.  Yet they aren't making Stirlings; if it really does cost $90k or even $10k for a 1 kW unit because of mechanical issues, it's not going to fit in the mass market.

I'd love to know what those issues are.


Honda has many brilliant internal combustion engineers on their product development group. I would find it hard to believe that such a group would choose a stiring when everybody doing the choosing had wonderful new ideas on how to make better IC engines.

But even if they succeeded in making a very long lived engine, say 20,000 hours, that would be only about 5 years of service before it had to be replaced. I expect my heating systems to last much much longer and they do.

British Gas has been trying to make a stirling home cogeneration system for years. But they have the problem that their experience is in selling gas, not manufacturing things, so their design is unnecessarily expensive and complex.

I still vote for the Harkness rule- look at the amount and type of material, and the number of parts and specified fits and tolerances, and that gives you the cost to manufacture. That makes a 1kW freepiston stirling less than $1000.

After all, I can buy a diesel fuel pump, or an air bearing spindle, or a hydraulic motor, or refrigeration compressor, all of which have about the same or closer tolerances than a stirling.

None of those devices have high-temperature sections like a Stirling engine does.  Further, they're all lubricated by liquids (hydraulic fluid and diesel fuel are both pretty good lubricating oils).  If you allow oil into a Stirling engine's working fluid, it will get into the hot section and coke up.  Even if you can tolerate the loss of oil, this will eventually clog the hot section.

I'd love to see Stirlings everywhere (I'd like to see one in every furnace, water heater and even many wood stoves) but dismissing the challenges involved when they are undeniable gets us nowhere.


I've been following Stirling engines for years. Read a book about SunPower a few years ago, dealing with concentrated solar power for electricity production, and a Stirling cooler for chilling computer components. I eventually stumbled across these guys:

Quiet Revolution Motor Company

Unfortunately, their site has become a cobweb (nothing happening, collecting dust). They were originally trying to develop a radial Stirling for use in aircraft. Their design is somewhat complex, and it took a little while to wrap my mind around it, but it is pure genius.

Just wish they'd start SELLING the blasted things.

If you REALLY want your mind blown, go looking for "thermoaccoustic stirling" engines. They found a way to make a Stirling with NO piston (sound waves moving through the pressurized gas do the work). A variation on the design allows you to separate gasses from a mixture, which could come in QUITE handy if you're try to:

  • remove the CO2 from the air in a spaceship

  • separate the methane from the other components in landfill gas

you get the idea.

Stirlings are wonderful things. If we could just get companies to BUILD the blasted things.


I am pleased to see such good comments on stirlings.

Oil is death on stirlings, exactly for the reasons you gave. That's why it is so hard to make crank type stirlings last. But the freepiston stirlings don't need it. They run on gas bearings just like those spindles for precision grinders do, except the side loads on the fp stirling are way smaller than they are on the grinder.

The fp stirling cryocoolers used by NASA have lasted for years in orbit and were running during launch. That proves they have something right. The load bearing setup for engines is exactly the same as on cryocoolers, except the end of the engine is hot and the end of the cooler is cold. In both cases, any oil at all would be death.

As for the thermoacoustic stirlings, Alas, they do have a moving part, the thing that moves to get power out. So it has the same challenges as the conventional fp stirling. And on top of that, the thermoacoustic has intrinsic irreversibilities tha doom it to lower thermal efficiency than the old piston-displacer stirliing.

Oddly enough, one of my passions is to put a real practical stirling on a wood stove. The stove is there, it has a chamber that runs very hot and is ready for a stirling.

I am making an opposed piston 1kW fp stirling that has no vibration and no noise and, according to standard analysis used by NASA, gets about 25% efficiency, defined as electric power out/ thermal energy to the heater head. This is not good enough for isotope powered satellites but plenty good for me and my wood stove.

To make this thing work for real, I cannot kid myself about ANY challenge, obvious or subtle.

wimbi, if you can make that work, I hope you find a way to parlay the design into many millions of units in the field and millions of dollars for yourself.

I mean that most sincerely.


Thanks for the kind thought. But let's say I do a perfect job, and the thing performs beyond my expectations and lasts forever. That is merely a ticket to a session with a source of money. and any source of money has an infinity of possibilities, so my chance of getting support for my particular wonderful machine is about one over infinity.

Same thing for a solar version (remember the concentrating PV this all started from?), or whatever you like.

We are all awash in great ideas, and even some of them will actually do what we hope for, but the next step is the overwhelming one, and I for one have never been able get past it in a lifetime of tries.

Just for fun, I'll name a few Great Ideas (not necessarily mine)

How about a vacuum tube train that whisks you from NYC to SF between breakfast and lunch, using almost no energy, and with no chance of ending you up dead in a skyscraper?
And what about a farm tractor that handles biomass energy crops using the same rough biomass for fuel?
And what about the simple idea of digging two holes in the ground, one storing winter cool and one storing summer heat, so that your house can be just cosy all year around with no energy usage?
Or a great big algae pond in each city, that eats sewage and produces fuel and chicken feed?

So all of us noble engineers get together and line up a slew of such perfectly feasible brainstorms, then we go give our pitch to a VC, and then we go home and weep for a little while, and then we go back and think up some more really great ideas, and then------.

Oh well, it's better than playing golf-- I think.

wimbi:  You'll find the vacuum-tube train in the back issues of Scientific American, in the 1960's or early 1970's.  (What can I say, I did a lot of reading of old stuff in my youth.)


Right, EP. It is a great idea. I believe it was first proposed by Goddard, the same guy who later became famous for rockets, in something like 1903. And of course, it has become ever so common in science fiction. I would vote for it as pretty close to number 1 priority, if I ever have the chance (fat chance).

There are lots of references to it on the web.

A hydrogen bomb expert first proposed (I think) the holes in the ground method of heating and cooling. Maybe he got the inspiration from looking at great big holes in the ground.

Maybe he got it from looking at how long it took for the temperature of the stuff in those holes to change.

Depending what you mean by "holes in the ground", there have been a number of inventions to exploit this.  Offhand, I can think of:

  • Massive volumes of soil used to store summer heat for winter and winter cold for summer, using nothing but the low surface/volume ratio for insulation.
  • An earth-sheltered house which incorporates a large amount of earth beneath its insulation and runs its ventilation ducts through it, driving its air circulation by convection so that temperatures are automatically moderated.


About 30 years ago I made just such a little hole in the ground house for my mother in law. It was spartan simple and she loved it, and so did I later as a quiet study, and so does my son in law now for the same reason. I recommend holes in the ground very highly, esp. if they have a nice super insulated window facing a peaceful scene.

The hydrogen bomb guy was Ted Taylor, who later recanted and became a strong advocate of sustainable energy.

And so much for concentrated solar energy.


This page contains a single entry from the blog posted on July 15, 2005 5:55 PM.

The previous post in this blog was H2 Nanostorage.

The next post in this blog is Simulating Culture and the Ethics of the Off Switch.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.34