Stirling Engines vs Solar Panels

[wx7]

Hi there.

I've been searching about Stirling engines recently, and I'd like to ask you guys something.

What do you think is more efficient, Stirling engines or solar panels?

Do you think it's more practical to build fields with Stirling engines with parabolic reflectors as a source of heat instead of solar panels?

And what about when it's rainy/cloudy, would solar panels do a better job?

Thank you.

[Shpaget]

If it's too cloudy for parabolic reflectors, it's just as cloudy for photovoltaics.

Concentrated solar thermal is a great thing, not with sterling but a proper steam turbine.

There are some environmental issues with it though. Everything that enters into the solar rays near the tower, gets vaporized.

[magnemoe]

Thermal solar makes lots of sense, more so if construction costs are low.

However you use steam turbines, for smaller installations you use PV.

[cfds]

It depends on scale:

You want to power your smart phone: Photovoltaics

You want to power bigger electric devices: Stirling engine

You want to power an industrial plant: steam turbine

[SciMan]

Why not line the mirrors with multi-junction photovoltaics, and apply a coating to them that reflects long wavelengths like thermal infrared?

The heat goes to a Stirling engine or steam turbine, and the photovoltaics convert many of the visible wavelengths.

If it was practical, most applications for technology would be highly hybridized.

Some already are.

Take a high-power radio broadcast transmittier, for example.

Vaccuum tubes, transistors, and ICs all have a place in it.

Tubes in the main RF amplifier, transistors in the pre-amp, and ICs for signal generation and audio processing.

Each one doing what it does best.

It's also why hybrid cars are still such a good option. The energy density and quick refueling of gasoline and diesel engine powered vehicles currently can't be beat, but an electric motor is much better for stop-and-go traffic or short trips.

Heck, even the multi-junction photovoltaics I suggest using are a hybrid technology.

The more specialized the application, the less different technologies will apply to it. Most technology applications have much more than one solution available.

For something as generic as grid-tied electric power generation, you could write a one hundred page list and still not have room to type all the different ways to skin that particular cat.

[noparse]TL:DR[/noparse] I'm with Zoidberg. Why not both?

Edited September 25, 2015 by SciMan

[Mattasmack]

Hi there.

I've been searching about Stirling engines recently, and I'd like to ask you guys something.

What do you think is more efficient, Stirling engines or solar panels?

It depends on what kind of solar panels you are comparing to. Typical panels used on homes or in slightly larger-scale installations (e.g. over parking lots or solar gardens) have efficiency in the 10-20% range, depending on whether the panels are thin-film, polycrystalline silicon or monocrystalline silicon. Concentrating photovoltaics (CPV) can have higher efficiency, around ~30%. These use high-efficiency multijunction solar cells (efficiency 40%+), but lose some energy in the concentrating optics. I don't know what the state-of-the-art of solar Stirling systems is, but my understanding is that their actual demonstrated efficiency are right in that same range of ~30%. So a solar Stirling system can be more efficient than run-of-the-mill solar panels, but if you compare like-for-like concentrating systems, it's about a wash between Stirling and CPV.

Do you think it's more practical to build fields with Stirling engines with parabolic reflectors as a source of heat instead of solar panels?

Better to look at cost as a figure of merit rather than efficiency. The plain old solar panels have lower cost and are therefore more practical -- assuming you have the space for them. The problem with highly concentrating systems (Stirling or CPV) is that concentration requires mirrors or other optics. These must be aimed precisely towards the sun and must be very rigid structures so that they aren't deflected or deformed by wind. The structure, the pointing mechanism, and the mirror surface or lenses add cost. (Solar Stirling systems that I've seen are pretty big, because there's a minimum practical size for the Stirling engine. That means the parabolic mirror structure is large and tall, which puts it further off the ground and exposed to stronger winds, so it has to be built stronger. CPV systems can be made smaller to avoid some of this problem.) The simple PV panels can just be plunked down on a fairly low, fixed structure so that they're tilted a bit to the south (or to the north, if you're in the southern hemisphere). If they get deflected a bit by the wind, or if the installation isn't precise, it doesn't really matter.

If you have limited space, then a higher-efficiency concentrating system might make sense.

And what about when it's rainy/cloudy, would solar panels do a better job?

Thank you.

No solar system will do much when it's cloudy. But, solar panels accept diffuse light and will produce a little power as long as there's some light. Concentrating systems use only direct sunlight, so in cloudy weather I wouldn't expect them to produce any power at all.

[Shpaget]

Mirrors, even the parabolic ones, are cheaper than PVs of the same surface area.

Parabolic trough design requires Sun tracking only on one axis, which can be done very cheaply, while maintaining structural integrity.

[Nibb31]

Sterling engines require moving parts. Space engineering typically doesn't like moving parts, because you get into all sorts of problems with lubricants, wear, and thermal tolerance. When a mechanical part fails, repairs are a pain.

[SargeRho]

Stirling engines are fairly simple, though. And since they mostly consist of a few large parts, they're one of the things that 3d printing could produce in Space.

[AngelLestat]

First I dont know if this question is for space or earth, but I guess it does not matter much.

Thermal solar can convert 90% (or a bit more) of the energy received into heat, which is great if you need heat. But convert that heat into electricity is not easy or cheap. You need to concentrate the heat to work with higher temperatures so you increase the efficiency of the cycle, to do that you need parallel parabolics with a host carring Supercritical CO2 in a brayton cycle or circular parabolics using stirlings.

These work better if you also produce heat water as cogeneration for some other use.

The problem with heat is that elements involve needs much more maintenance, the overall efficiency (thermal-electric) would be between 30% to 40%.

Now PV of 25% of efficiency are super cheap, you can make huge farms of these which does not need much maintenance.

You can reach even 45% of efficiency with high tech PV.

The pros of thermal energy is that the same heat can help you to storage that energy, so you dont stop to produce on night. But mostly the biggest energy consumption from cities is in day times.

Mirrors are not cheap either, less if you need parabolic structures.

Maybe for space if we find a good way to produce a reflective thin surface that keeps its parabolic shape (in thermal energy you need to concentrate heat or you lose efficiency) with almost not structure (microgravity), then you might have a good system, but PV is not behind, each time are more light, cheap and efficient. So if I need to bet for one of these two technologies..

In mostly all general cases, PV will be better.

[Mattasmack]

Mirrors, even the parabolic ones, are cheaper than PVs of the same surface area.

Parabolic trough design requires Sun tracking only on one axis, which can be done very cheaply, while maintaining structural integrity.

Sure, mirrors are cheaper than PV cells. But the slumped glass mirrors typically used in parabolic trough receivers are not so cheap. Neither are the structure holding the mirrors, the drive turning the structure, the foundations, the receivers, or the labor required to align all of these components precisely. And all of this typically accounts for half (or less) of the total plant cost.

Let's look at some recent numbers:

The most recent parabolic trough plant completed in the U.S. that I could find is the Mojave Solar Project, completed at the end of 2014: 250 MW net power output for $1.6B in construction costs, or $6.40 per installed Watt. The plant's capacity factor is 28%, suggesting that it has little or no thermal storage. Perhaps it uses natural gas cofiring like earlier parabolic trough plants in California did. (See its Wikipedia article, and this letter citing capacity factor.)

In contrast, as of 2013 the average cost of utility-scale PV projects was $3.7 per installed Watt, with an average capacity factor of 27.5%. (report here)

This comparison suggests that parabolic trough plants are about 70% more expensive than PV ones. Which is consistent with PV being installed at a much faster rate than solar thermal.

Edited September 26, 2015 by Mattasmack

[SomeGuy12]

Another factor that kills the solar-thermal is really simple.

The sun moves across the sky if the array is on earth. Mirrors have a focal point that depends on the position of both the light source and the mirror angle.

So you have to install electric motors on every single mirror and sensors so they can be moved under computer control. This is expensive and complicated and the motors will break. PV panels, you just screw down onto a rack and forget about it. If you use microinverters, failures and shading of individual panels don't affect the rest of the panels, and microinverters now cost $0.50 a watt, the same cost as the larger string inverters now. (for a few years, microinverters were more expensive).

So it's basically foolproof if you use PV. The same consideration applies to solar PV in space. Early satellites and even some modern satellites, they just slap solar cells onto every face of the satellite. No matter how it's oriented, it gets some power, keeping it alive and able to accept commands. Even the big arrays that have motors to keep them oriented are much simpler than solar thermal would be.

I think the OP has a point, actually. If you were doing large scale space based solar, solar thermal might be the way to go. The reason is that per every kilogram you launch, sterling + solar thermal is possibly more power per kilogram as mirrors are far lighter than solar pv, and sterling heat engines can be pretty efficient. Then again, the sterling engine component may be far heavier than just using efficient PV panels everywhere - they make high end double and triple junction PV that are 31-44% efficient. That's probably lighter and more efficient than a sterling engine of the same mass.

No, what you'd do for space based solar is you might use mirrors to concentrate sunlight onto a smaller array of double/triple PV. This might end up being lighter as the mirrors could be very thin and light. Also easier to shed heat from a concentrated panel as blackbody radiation is proportional to T^4.

[zonova]

Just a quick question, I was wondering, why would you want to use a Steam Turbine instead of a Sterling Engine? How do these things compare in terms of efficiency and energy production? Thanks!

[Streetwind]

Brayton cycle turbines are the highest efficiency heat engines that humanity knows how to build - they convert the largest percentage of the heat input into shaft power. A stirling engine doesn't come close (although in its defense, it is also simpler and cheaper).

We use these turbines everywhere nowadays. Coal power plants use steam turbines. Nuclear power plants use steam turbines. Thermal solar power plants use steam turbines. Oil power plants use gas turbines with additional heat harvesting systems that feed into steam turbines for good measure. If efficiency is what you're after, you go turbine