has been a thing for a while and seem to be more expensive than normal PV panels.
If this design really focuses the sun "10000 times", then it'll need complicated cooling to avoid killing the solar cell. I wonder how that is supposed to work.
But are usable Stirling engines available? Dean Kamen's been promising a refined version for over a decade and has yet to release anything to production.
No idea about Stirling engines but you could always use other kinds of steam engines or turbines, albeit with lower theoretical efficiency. There's a very niche hobby for this kind of stuff, see https://www.youtube.com/watch?v=uJGpbvvJA2I for example.
The one thing that is problematic overall probably is the noise that any steam-based solution will make.
Stirling engines are used in some niche applications. In particular, several Swedish and Japanese submarines are powered by Stirling engines running on diesel and liquid oxygen. The ability to run on a separate oxygen supply from breathing air is apparently a main advantage.
NASA and DOE did some tests with Stirling powered cars in the 70's and 80's. They got up to 50 MPG. In the end, retooling, cheaper fuel and different UX when driving was the negatives that prevented mainstream adoption.
You could definitely use moonlight to heat up an engine, if you have a big enough concentrator. You just can't reach a temperature higher than the surface of the moon itself. Stirling engines are capable of operating with relatively small temperature differences.
> You just can't reach a temperature higher than the surface of the moon itself.
False. You can't reach a temperature hotter than the black body which emitted the light. The moon is a reflector which is as cold as -153 C, not a black body emitter.
Try reflecting sunlight with a large mirror cooled by air to 23 C through a magnifying glass onto an object in the shade. You can still light things on fire.
A better analogy is a piece of paper. Like a mirror, it reflects rather than thermally emits the light, but unlike a mirror, it scatters the reflected light, limiting how much it can be focused.
I'm a huge Randall Munroe fan, and so I was hugely disappointed when he posted that "What if", because it is 100% wrong.
Moonlight is pretty much partially reflected sunlight. The moon is acting more like a dirty mirror than like a blackbody emitter. This is easily apparent through two simple facts:
1) If the moon was emitting light due to its temperature rather than reflecting light from a higher temperature object, we couldn't see it all. Objects at 120 C don't really emit light in the visible range.
2) The color temperature of moonlight is around ~4000K, about 1000 K lower than sunlight. It's not like we haven't measured this stuff.
So, yes, you could start a fire with moonlight. I leave this as an exercise for the Mythbusters-inspired reader.
Im not following your reasoning. Which I admit has more to do with my lack of knowledge than anything else, so don't read that as "I think you are wrong". I did find other places with the question asked... some have similar answers to the what-if... others came to a different conclusion as you did. Either way even if it is on some boundary of what is possible, it seems weird for this company to mention it as a feature considering it would be many many times less effective than using it with sunlight and even then only on a few nights out of the month if that..
EDIT: After giving TFA another look, I see that they do mention using PV and not just steam engines... so I guess you could use moonlight sometimes to charge a PV, then switch it to the steam engine during the day or something.
Another way that you could look at it, is as LeifCarrotson said below, that the moon is like a mirror. If RM's idea that the sun's light reflected from the moon could not start a fire was true, what makes the situation different if you are reflecting the sun's light from a cold mirror? Can the light from the mirror not cause a higher temperature than the temperature of the mirror? We know that mirrors (and even lenses made of ice) can be used to create temperatures at their focus high enough to melt them. So, what is it about the moon that is different from a mirror, besides a lower reflectivity?
I suspect that RM just got a little lost in the weeds on this one. I remember making a similar mistake in undergrad when I was thinking about a problem with two light sources with an optical filter in between. The key insight I needed to make was that the filter was a physical object with its own temperature.
Thinking about things like this is how you develop what physicists call "physical intuition".
Like many "advances" in solar energy, nobody is going to care about the efficiency gains if they are overshadowed by cost increases.
The article tries to make it sound like a miracle technology that concentrates light "up to 10,000x", but that doesn't mean you get 10,000x more energy out of it, it just means that you can use a much smaller panel since the light is concentrated on a smaller spot by a comparatively large, bulky piece of hardware.
Panels are already pretty cheap, so it's not clear if the cost of a giant glass sphere, plus heavy-duty steel frame, plus tracking hardware will beat just blanketing your roof with stationary panels. It looks cool, sure, and maybe it's a bit more efficient at converting sunlight to electrical energy, but the real figure that's going to matter is the $/kW ratio.
Light concentrating lenses and mirrors have been a thing for a very long time. This article doesn't go into any detail about what makes the globe so special, other than it looks kind of cool.
The focal point for light going through a sphere is right at the surface of the sphere on the opposite side of the light source. So the focal point will rotate around the outer surface of the sphere as the sun moves throughout the day (and it also changes slightly each day in a perpendicular direction to the sun's daily movement). Assuming this is using some kind of steam engine, the water tank that gets heated by the focused light needs to follow this focal point.
That's for efficient electrical generation, which I assume is the point of this thing. As far as pure heat, you're right. You could just have a big curved metal water tank wrapped around the bottom half of the sphere and you might get pretty good water heating capabilities without a complicated tracking system.
The focus point changes with time, so the cell/converter portion needs to move or the entire rear side of the sphere would need to be covered in cells.
A water filled glass sphere? Really? A single typical (300 to 400 watt) solar panel has a surface area of around 2 square meters. A glass sphere with the same cross sectional area is 0.8 meters in radius and would weight over 2130 kg or around 4700 pounds.
>"Rather than glass lenses, the consumer models will rely on water-filled acrylic polymer lenses and will also be able to handle thermal energy generation in addition to solar energy generation."
I don't think the weight of the glass is the main problem here. (Though obviously plastic will be lighter and easier to transport.)
Where do you mount this contraption? Solar PV is light enough for any roof - otherwise wasted space. You probably have to mount this on the ground unless you want to spend a ton of money reinforcing everything.
I think there is a bit of a communication breakdown here.
@todd8 is simply stating that the sphere full of water would be too heavy for this to be a serious proposal.
@dghughes is basically replying, "Look, it says so right here. That's their plan."
They're not arguing over the material to construct the sphere. It is merely incidental that there was conflicting information in each of their comments.
Here is eastern Canada pitched rooves are capable of withstanding about a metre of snow. Wet snow supposedly has a mass of close to a tonne per cubic metre, a cube of pure water is one tonne at 4C. I can't imagine these devices would weigh more than a tonne (1,000 kg) per square metre.
Bummer Questions:
- How much does it cost?
- Does it require special expertise to install? I'd hope/assume the sphere can be filled with water on-site
- How often will it need to be inspected/repaired? Will that require a specialist?
- How often will it need to be cleaned?
This wouldn't seem to make much sense for individual installations (unless you're wealthy and want something to show off on your roof), but I wonder if the economics would work out in medium-sized arrays, to fill the gap between small residential/commercial installations and huge solar collector operations. Still though, is a 35% efficiency increase enough to justify this particular design? Is it that much better than other concentration methods?
This is obnoxious, but why do they report the percentage improvement, but not the actual yield?
If it is 35% improvement, it would be nice to know what it was an improvement on. One of the linked sources suggests that current maximums on PV efficiency are around 24% of total sun energy converted to electricity out of a maximum of 33.7%. so, if that is the case then does this get us from 24% to 32%, or very close to the maximum solar electricity extraction per area?
Please double check my math or correct me where I'm wrong, but putting everything in marginal terms is very confusing.
Beautiful art, but it's inconceivable that this would be cost effective per unit area if it's only increasing panel efficiency by 35%
That's roughly 500Kg of water, in a glass sphere, and the whole thing would have to cost a couple hundred dollars to beat standard PV, an it STILL needs an active tracking system.
34 comments
[ 3.9 ms ] story [ 148 ms ] threadhas been a thing for a while and seem to be more expensive than normal PV panels.
If this design really focuses the sun "10000 times", then it'll need complicated cooling to avoid killing the solar cell. I wonder how that is supposed to work.
https://en.wikipedia.org/wiki/Concentrated_solar_power
The one thing that is problematic overall probably is the noise that any steam-based solution will make.
NASA and DOE did some tests with Stirling powered cars in the 70's and 80's. They got up to 50 MPG. In the end, retooling, cheaper fuel and different UX when driving was the negatives that prevented mainstream adoption.
https://what-if.xkcd.com/145/
False. You can't reach a temperature hotter than the black body which emitted the light. The moon is a reflector which is as cold as -153 C, not a black body emitter.
Try reflecting sunlight with a large mirror cooled by air to 23 C through a magnifying glass onto an object in the shade. You can still light things on fire.
Moonlight is pretty much partially reflected sunlight. The moon is acting more like a dirty mirror than like a blackbody emitter. This is easily apparent through two simple facts:
1) If the moon was emitting light due to its temperature rather than reflecting light from a higher temperature object, we couldn't see it all. Objects at 120 C don't really emit light in the visible range.
2) The color temperature of moonlight is around ~4000K, about 1000 K lower than sunlight. It's not like we haven't measured this stuff.
So, yes, you could start a fire with moonlight. I leave this as an exercise for the Mythbusters-inspired reader.
[1] https://en.wikipedia.org/wiki/Color_temperature
[2] http://physics.stackexchange.com/questions/244922/why-does-m...
EDIT: After giving TFA another look, I see that they do mention using PV and not just steam engines... so I guess you could use moonlight sometimes to charge a PV, then switch it to the steam engine during the day or something.
I suspect that RM just got a little lost in the weeds on this one. I remember making a similar mistake in undergrad when I was thinking about a problem with two light sources with an optical filter in between. The key insight I needed to make was that the filter was a physical object with its own temperature.
Thinking about things like this is how you develop what physicists call "physical intuition".
The article tries to make it sound like a miracle technology that concentrates light "up to 10,000x", but that doesn't mean you get 10,000x more energy out of it, it just means that you can use a much smaller panel since the light is concentrated on a smaller spot by a comparatively large, bulky piece of hardware.
Panels are already pretty cheap, so it's not clear if the cost of a giant glass sphere, plus heavy-duty steel frame, plus tracking hardware will beat just blanketing your roof with stationary panels. It looks cool, sure, and maybe it's a bit more efficient at converting sunlight to electrical energy, but the real figure that's going to matter is the $/kW ratio.
That's for efficient electrical generation, which I assume is the point of this thing. As far as pure heat, you're right. You could just have a big curved metal water tank wrapped around the bottom half of the sphere and you might get pretty good water heating capabilities without a complicated tracking system.
>"Rather than glass lenses, the consumer models will rely on water-filled acrylic polymer lenses and will also be able to handle thermal energy generation in addition to solar energy generation."
Where do you mount this contraption? Solar PV is light enough for any roof - otherwise wasted space. You probably have to mount this on the ground unless you want to spend a ton of money reinforcing everything.
@todd8 is simply stating that the sphere full of water would be too heavy for this to be a serious proposal.
@dghughes is basically replying, "Look, it says so right here. That's their plan."
They're not arguing over the material to construct the sphere. It is merely incidental that there was conflicting information in each of their comments.
And there are much cheaper and more elegant ways to concentrate light than a massive glass/water-blob:
https://en.wikipedia.org/wiki/Fresnel_lens
Just mount a freakin fresnel lens on the tracking device and you have a system with fractional weight ...
https://www.youtube.com/watch?v=drE54ctrHBY
(My guess is that this is originally an art project or maybe a Joke that was misinterpreted by some not-so-technical media)
Get your fresnel lenses e.g. here:
http://www.ebay.com/sch/i.html?_nkw=fresnel+lens
This wouldn't seem to make much sense for individual installations (unless you're wealthy and want something to show off on your roof), but I wonder if the economics would work out in medium-sized arrays, to fill the gap between small residential/commercial installations and huge solar collector operations. Still though, is a 35% efficiency increase enough to justify this particular design? Is it that much better than other concentration methods?
This feels enormously wasteful in construction, a huge sphere doesn't feel like the most obvious lens shape. Am I missing something?
If it is 35% improvement, it would be nice to know what it was an improvement on. One of the linked sources suggests that current maximums on PV efficiency are around 24% of total sun energy converted to electricity out of a maximum of 33.7%. so, if that is the case then does this get us from 24% to 32%, or very close to the maximum solar electricity extraction per area?
Please double check my math or correct me where I'm wrong, but putting everything in marginal terms is very confusing.
That's roughly 500Kg of water, in a glass sphere, and the whole thing would have to cost a couple hundred dollars to beat standard PV, an it STILL needs an active tracking system.