Solar heat engine’s 20% efficiency may trump solar panel’s 15% for some apps

I wonder how installation costs compare between solar Stirling and photovoltaics. I imagine a hefty amount of concrete needs to be poured beneath one of those Infinia dishes. I wonder how many tons and I wonder if it's included in the $15K price.

Ok, let's do a quick back-of-the-envelope ROI on this gizmo. Here in the Midwest (MO), my electric bills average under $100/month (higher in the summer, half of that in the winter with gas heat). Let's say, for sake of argument, that this gizmo would supply roughly one third of my electricity for the year (daylight hours only and being generous for production on partly-cloudy days and loss of output on cloudy days). Even if I figure that I back-feed excess power into the utility when my load is light and the sun is shining, I come up with $400 per year in direct savings (the electricity I use that I generate myself instead of buying) and maybe another $400 per year in electricity that I sell to the utility. $800 per year takes nearly 20 years to pay off the gizmo. Anyone want to bet that it will last anywhere near that long? No thanks. For consumers with access to a reliable and reasonably-priced grid, it doesn't make sense. Now, for an isolated installation where you don't HAVE the grid, these could be very interesting.

Everything old is new again. There were devices very much like this operating printing presses and similar equipment 120 years ago. It's a nice package, and looks enough like a satellite dish to intrigue some people (such as investors), but I'm not sure it's really viable except maybe in certain niches.

The headline on this article is worthless - as is much of the article. Photocells can be more efficient than 20% and they are being "produced". The ONLY thing that matter is the dollars per watt over the lifespan of whatever is doing the generation.

I'm not seeing any advantage. Solar arrays can also be made to track and can still extract power on an overcast day. The stirling unit turns into a boat anchor. Then, and most importantly, what is the ratio of $/W to MTBF? Until they come up with a heat based unit that is cheap and has no moving parts but the tracker, forget it. This thing is a big dish that you throw money into and it makes it disappear.

Stirling Energy Systems have just announced a record of 31.25% efficiency. This is an overall efficiency, net electrical power fed to the grid over total solar power intercepted by the dish and includes all pumps, motors, controllers etc. The efficiency records of stacked junctions is just the DC power over the concentrated solar power at the cell. It does not include losses in the lens or mirror used to concentrate the light, light missing the cell at the edges of the illuminated area, inverter losses and tracker power. I doubt the overall efficiency of these cells exceeds Stirling Energy Systems figure.

@Bob: in Germany we pay 2500? for 1kW(peak) of installed solar power (everything included, roof-top mounting, solar panels, cables, inverters, labour, etc.). That would be some $3.4/W. I guess we will be down to roundabout 2000?/kWp by the end of this year. In my region we get in the order of 900kWh/kWp of electrical energy per year, which means we are close to grid parity (as seen from the end user). If the Watt of a Stirling engine is comparable to the Watt (peak) of a PV installation, then 5$/W is too expensive. This is especially true, when taking into account that the Stirling engine has moving parts and will probably not sustain the 25 years continuous operation, which is expected from PV (granted, the capacitors inside the inverters will probably have to be replaced after 10 years, but that is a minor issue). Michael

Since Infinia uses concentration and tracking, comparison to CPV systems is more reasonable than comparison to flat panels or thin film. CPV systems already hit 25%-28% efficiency. What is impressive is $2B of orders.

Whoo hoo - I'm the first commenter. Like it matters....