Whose boosters are you talking about? Certainly not SpaceX; the entire first stage (the majority of the rocket) gets reused. Are you thinking of the Shuttle SRBs? Technically the "majority" was reused, as the casing was, but because they're solids, it's not a simple matter of refilling; you have to tear down the booster and recast the propellant mixture, which is a ton of work. Indeed, there weren't a fixed "set" of boosters; a given launch may involve segments that had never been part of the same booster previously. The SRBs also landed hard and faced saltwater corrosion every time, which significantly increased their maintenance.
To clarify what catmistake means by "just like last time": MDMA was a psychotherapy drug used for, among other things, PTSD. That's how it got its start, before breaking out into the recreational scene. When the FDA considered banning it, there was a court hearing on the topic, which turned into a constant stream of psychiatrists stepping up and saying, "Don't do this!". The FDA at the time was unaware that it was used in psychotherapy. The judge ruled that it should be classed as a Schedule III drug, aka something with an established medical use but also the potential for abuse. However, the DEA administrator overrode him and classified it as a Schedule I drug. The DEA was sued by a Harvard psychiatrist for misclassification, and he won; the court stripped the DEA's Schedule I classification. The DEA responded by simply reclassifying it yet again as Schedule I.
The scheduling has made research difficult over the years, but the widespread attestment to its effectiveness is compelling. Research in other regards has shown that the act of recalling a memory also involves, to some degree, writing it back; there's been treatment researched for trauma wherein the patient recalls memories while on drugs that induce mild amnesia. I wouldn't be surprised if something similar is at work here.
If you're talking about clouds shading areas around you but not your panels themselves, sure, that's possible. But not when your panels themselves are shaded.
The temperature dependence of solar cells is small. The light difference from shading is huge.
Yes, the efficiency does increase. When calculating the short circuit voltage of a solar cell, you start with the AM0 voltage and then scale it by a correction factor based on the logarithm of the short circuit current. The short circuit current also modifies the temperature dependence factor.
The efficiency difference is very small, about 0.5% per degree celsius. The difference in light between cloudy and sunny days, however, is most definitely not small. Also, for a given cell temperature, solar panels become more efficient the more concentrated the light they're receiving is.
This person and the GP are talking bollocks. There's no meaningful difference in efficiency between sunny and cloudy days, but a big net generation difference due to the vastly reduced light availability on cloudy days. Graphs of PV power output make this abundantly clear - check them out for yourself.
Actually, even that's wrong. Solar cells do get greater efficiencies at lower temperatures, but they also get greater efficiencies at higher light intensities. The highest efficiency solar cells in operation use a combination of concentrators, splitters, and cooling so that as much light as possible, at cell-optimized frequencies, falls on as little area as possible with that area being kept as cool as practical. The world record using that approach is 46% efficiency.
Also, you want your light coming from as dead-on as possible, not scattered and coming in from all angles. Panels are tilted to be optimal relative to the expected angle of the sun (and ideally tracking it). The steeper the angle, the more light is lost to reflections.
I'm sorry, but solar panels do dramatically reduce in power on cloudy days. The absolutely do not "operate surprisingly close to the same capacity as on a non-cloudy day." Here's what a daily generation profile looks like on a day with scattered clouds. Here you can see a mixture of cloudy, sunny, and partly cloudy days.
Your statement was simply wrong.
Literally nothing you wrote in your post was correct. UV is a nearly irrelevant source of energy at the surface. Clouds do provide some UV blocking, and they're nearly opaque to IR, not just "a bit". Normal solar panels can't run on IR, and are either highly inefficient with or can't use UV at all. And no, solar panels do not "operate surprisingly close to the same capacity as on a non-cloudy day"
US solar capacity factors are vastly higher than 14,5%. And you don't need to explain what capacity factor and nameplate capacity are, people aren't idiots.
We don't know what the capacity factor of this plant is, but at around $1/W nameplate for almost-no-operations-costs power produced at peak consumption hours, it will be quite cost effective.
Your calculation is not just wrong, but stupid. First off, hint, check your units in your divisor: where is seconds per hour coming from? You have nothing in seconds in that formula. The "hours" in "hours per year" is supposed to cancel with the hours in kilowatt hours, the MW and kW are supposed to cancel out watts, leaving you with $/years. Instead your denominator has an additional seconds per hour in it. You should have had 1000kW/MW there. But beyond that, that's not how power markets work. There's no "constant value of power", and even if it was, that value would not be the same as the incremental residential rate. Power varies by a number of factors, such as time of day and responsiveness to demand. At low penetration, solar power is worth more than baseload as it compensates for demand peaks. At high penetration, solar is worth less than baseload because of its complete lack of response to demand. Lastly, simple payback periods are not how you determine whether an investment is an economically appropriate decision as they don't take into account the time value to money. You calculate an X-year ROI based on what sort of financing rates you can get on the project (which in turn are largely based on risk) and compare it to other ROIs you could get from other projects. I'll save you the time: this plant easily makes economic sense.
I've actually seen arguments for just that. And it probably will eventually happen. But smaller nationwide / continent-wide grids obviously come first; you walk before you run. Likewise, superlong undersea connects are yet to be proven. Should the Iceland/Scotland link go through, that'd be a nice demonstrator.
The problem is that nuclear plants have been getting more expensive over time, not less.
Also, they don't have to pay for catastrophic liability.
And even with government-provided catastrophic liability coverage (which nobody in the private sector would provide - a $200B payout for a Fukushima-style event would bankrupt anybody), there's few takers. Nuclear plants underwent a two decade lull when the last generation of nuclear plants turned out to be more expensive than expected before undergoing a "nuclear renaissance" with a new generation of "cheaper" plants... that turned out to be even more expensive. For example: Hinkley Point in the UK, when all is said and done, is expected to cost about $11,5M per MW. And that's not counting operations nor decommissioning. Hence investment in new nuclear plants has plunged yet again.
Why are you acting shocked that the plant's power rating is nameplate (aka peak) rather than average? Power plants are always reported by nameplate capacity. If you want to know the capacity factor, that's a different statistic: capacity factor.
Re, India's power consumption: India consumes 1106 TWh/year. Assuming a capacity factor of 0.22 here then this plant would generate 1,25TWh/year, or 0,11% of India's consumption, not 0,0007%. 0.00015% of India's land for 0,11% of its consumption, aka 0,13% of India's land for 100% of its consumption. In terms of wildlife health and agricultural output effects relative to generating power from polluting sources (pollution hurts animals and reduces crop yields), that's a no-brainer - all issues of climate change aside. It's also worth noting that solar plants tend to be more energy dense sources of energy than hydroelectricity (when the reservoir is counted), sometimes by large margins, and many orders of magnitude more energy dense than growing plants for biofuels, per unit energy therein. PV plants also require no cooling water, meaning huge benefits for rivers, and more water availability for agriculture. Lastly, PV plants can be built on marginal lands unsuitable for agriculture on their own - and the shade they provide reduces evaporation from the underlying soil, increasing water availability downstream.
1) Solar panels *do* drastically reduce power output on cloudy days. 2) Clouds do *not* only block visible spectrum light "and a bit of IR". They're highly effective blockers of IR and UV as well. They do block IR and visible light better than UV (and there are relatively rare situations where they can actually enhance UV light via reflections), but they absolutely block all three, and do so well. See chart c in figure S1 / figure 6. 3) Only 3-5% of the sun's energy at Earth's surface is UV. IR is 52-55% and visible 42-43%. 4) Standard solar panels can't use infrared (too low photon energy), and are very wasteful with UV (still just one electron per photon regardless of photon energy) if they can use it at all (glass coated panels = UV blocking).
Not the same thing. Standard AC distribution lines are not a cost effective means to distribute huge amounts of power from one side of a large country to the other. For that you need high power HVDC lines.
A nice thing about HVDC is that unlike AC, it also works well under seawater. Also, it shares power between disjoint AC grids (since it's always converted to the local waveform) and improves power quality on the distribution end. And prevents the cascading power failures that AC is prone to (aka, one part of the grid going out of sync with others). The lines themselves are vastly cheaper vs. how much power they carry, and the losses tiny, even over great distances. The main downside is that the terminals are quite expensive.
Exactly my reaction. If they say the project cost $679m to build, then it means just that: the project cost $679m to build. Not "one aspect of it" cost $679m.
A price of just over $1 a watt is superb. That's about what it costs to build a typical fossil plant - except that the cost to build a fossil plant is dwarfed by the cost of running it. Now, a fossil plant will have a 3x higher capacity factor, but still, this is highly competitive power. To put it in perspective, some of the new nuclear plants they're building in Europe cost over $10 per watt. Just to build, not counting operations and decommissioning.
Now, up to a given level of penetration, solar aides the grid by boosting the supply curve when demand is highest (the middle of bright sunny summer days). So up to that point, the baseload vs. intermittent supply argument is moot, and even the capacity factor doesn't play in (in a sunny location, at least), because the only capacity you need is to fill in those daytime peaks. At high levels of penetration however you start having to factor in increasing levels of peaking and/or storage. This can be somewhat offset by geographic smoothing and diversity of energy sources (solar + wind + others), but nonetheless your cost effectiveness will decline once your market penetration becomes large. Still, these are some superb numbers that bode very well for the future of solar.
It's best not to base it on "hours per day", and instead just look at capacity factors. Capacity factors on commercial scale solar plants range from under 15% to over 30%, depending on the tracking tech (none, single axis, dual axis) and plant design (as well as the most critical aspect, of course - location).
A nice thing about solar is that it tends to align pretty well with the demand curve, so up to a point adding actually makes grid operators' jobs easier, not harder. It also runs contrary to wind, which tends to blow stronger at night, and periods of low sun tend to most often be high wind and vice versa.
Tesla X + big boat = huge aero drag problem, not weight problem. And even concerning weight, freight truck tires are much more efficient than normal car tires.
2kWh/mi is what the electric cargo crate haulers at the Port of Los Angeles get. Maybe reduce the efficiency some for higher average speeds, but it's ballpark.
Slashdot Mirror
Intelligent car-damaging flesh-eating parrots?
Hopefully not "Never Mind The Buzzcocks" ;)
Hmm, then what's the name of his social media feed? "Buzzwords"? "The Daily Buzzbomb"?
Bad voters! Sad.
Where else are they supposed to stage the first manned landing on Europa if not Antarctica?
Or, for that matter, the POTUS's two adult sons. I always forget their names... Uday and Qusay?
You can officially go F*** yourself, 2016.
Whose boosters are you talking about? Certainly not SpaceX; the entire first stage (the majority of the rocket) gets reused. Are you thinking of the Shuttle SRBs? Technically the "majority" was reused, as the casing was, but because they're solids, it's not a simple matter of refilling; you have to tear down the booster and recast the propellant mixture, which is a ton of work. Indeed, there weren't a fixed "set" of boosters; a given launch may involve segments that had never been part of the same booster previously. The SRBs also landed hard and faced saltwater corrosion every time, which significantly increased their maintenance.
To clarify what catmistake means by "just like last time": MDMA was a psychotherapy drug used for, among other things, PTSD. That's how it got its start, before breaking out into the recreational scene. When the FDA considered banning it, there was a court hearing on the topic, which turned into a constant stream of psychiatrists stepping up and saying, "Don't do this!". The FDA at the time was unaware that it was used in psychotherapy. The judge ruled that it should be classed as a Schedule III drug, aka something with an established medical use but also the potential for abuse. However, the DEA administrator overrode him and classified it as a Schedule I drug. The DEA was sued by a Harvard psychiatrist for misclassification, and he won; the court stripped the DEA's Schedule I classification. The DEA responded by simply reclassifying it yet again as Schedule I.
The scheduling has made research difficult over the years, but the widespread attestment to its effectiveness is compelling. Research in other regards has shown that the act of recalling a memory also involves, to some degree, writing it back; there's been treatment researched for trauma wherein the patient recalls memories while on drugs that induce mild amnesia. I wouldn't be surprised if something similar is at work here.
If you're talking about clouds shading areas around you but not your panels themselves, sure, that's possible. But not when your panels themselves are shaded.
The temperature dependence of solar cells is small. The light difference from shading is huge.
Yes, the efficiency does increase. When calculating the short circuit voltage of a solar cell, you start with the AM0 voltage and then scale it by a correction factor based on the logarithm of the short circuit current. The short circuit current also modifies the temperature dependence factor.
The efficiency difference is very small, about 0.5% per degree celsius. The difference in light between cloudy and sunny days, however, is most definitely not small. Also, for a given cell temperature, solar panels become more efficient the more concentrated the light they're receiving is.
This person and the GP are talking bollocks. There's no meaningful difference in efficiency between sunny and cloudy days, but a big net generation difference due to the vastly reduced light availability on cloudy days. Graphs of PV power output make this abundantly clear - check them out for yourself.
Actually, even that's wrong. Solar cells do get greater efficiencies at lower temperatures, but they also get greater efficiencies at higher light intensities. The highest efficiency solar cells in operation use a combination of concentrators, splitters, and cooling so that as much light as possible, at cell-optimized frequencies, falls on as little area as possible with that area being kept as cool as practical. The world record using that approach is 46% efficiency.
Also, you want your light coming from as dead-on as possible, not scattered and coming in from all angles. Panels are tilted to be optimal relative to the expected angle of the sun (and ideally tracking it). The steeper the angle, the more light is lost to reflections.
I'm sorry, but solar panels do dramatically reduce in power on cloudy days. The absolutely do not "operate surprisingly close to the same capacity as on a non-cloudy day." Here's what a daily generation profile looks like on a day with scattered clouds. Here you can see a mixture of cloudy, sunny, and partly cloudy days.
Your statement was simply wrong.
Literally nothing you wrote in your post was correct. UV is a nearly irrelevant source of energy at the surface. Clouds do provide some UV blocking, and they're nearly opaque to IR, not just "a bit". Normal solar panels can't run on IR, and are either highly inefficient with or can't use UV at all. And no, solar panels do not "operate surprisingly close to the same capacity as on a non-cloudy day"
US solar capacity factors are vastly higher than 14,5%. And you don't need to explain what capacity factor and nameplate capacity are, people aren't idiots.
We don't know what the capacity factor of this plant is, but at around $1/W nameplate for almost-no-operations-costs power produced at peak consumption hours, it will be quite cost effective.
Your calculation is not just wrong, but stupid. First off, hint, check your units in your divisor: where is seconds per hour coming from? You have nothing in seconds in that formula. The "hours" in "hours per year" is supposed to cancel with the hours in kilowatt hours, the MW and kW are supposed to cancel out watts, leaving you with $/years. Instead your denominator has an additional seconds per hour in it. You should have had 1000kW/MW there. But beyond that, that's not how power markets work. There's no "constant value of power", and even if it was, that value would not be the same as the incremental residential rate. Power varies by a number of factors, such as time of day and responsiveness to demand. At low penetration, solar power is worth more than baseload as it compensates for demand peaks. At high penetration, solar is worth less than baseload because of its complete lack of response to demand. Lastly, simple payback periods are not how you determine whether an investment is an economically appropriate decision as they don't take into account the time value to money. You calculate an X-year ROI based on what sort of financing rates you can get on the project (which in turn are largely based on risk) and compare it to other ROIs you could get from other projects. I'll save you the time: this plant easily makes economic sense.
I've actually seen arguments for just that. And it probably will eventually happen. But smaller nationwide / continent-wide grids obviously come first; you walk before you run. Likewise, superlong undersea connects are yet to be proven. Should the Iceland/Scotland link go through, that'd be a nice demonstrator.
The problem is that nuclear plants have been getting more expensive over time, not less.
Also, they don't have to pay for catastrophic liability.
And even with government-provided catastrophic liability coverage (which nobody in the private sector would provide - a $200B payout for a Fukushima-style event would bankrupt anybody), there's few takers. Nuclear plants underwent a two decade lull when the last generation of nuclear plants turned out to be more expensive than expected before undergoing a "nuclear renaissance" with a new generation of "cheaper" plants... that turned out to be even more expensive. For example: Hinkley Point in the UK, when all is said and done, is expected to cost about $11,5M per MW. And that's not counting operations nor decommissioning. Hence investment in new nuclear plants has plunged yet again.
Joules per year = energy over time = power
Your slip is showing.
Why are you acting shocked that the plant's power rating is nameplate (aka peak) rather than average? Power plants are always reported by nameplate capacity. If you want to know the capacity factor, that's a different statistic: capacity factor.
Re, India's power consumption: India consumes 1106 TWh/year. Assuming a capacity factor of 0.22 here then this plant would generate 1,25TWh/year, or 0,11% of India's consumption, not 0,0007%. 0.00015% of India's land for 0,11% of its consumption, aka 0,13% of India's land for 100% of its consumption. In terms of wildlife health and agricultural output effects relative to generating power from polluting sources (pollution hurts animals and reduces crop yields), that's a no-brainer - all issues of climate change aside. It's also worth noting that solar plants tend to be more energy dense sources of energy than hydroelectricity (when the reservoir is counted), sometimes by large margins, and many orders of magnitude more energy dense than growing plants for biofuels, per unit energy therein. PV plants also require no cooling water, meaning huge benefits for rivers, and more water availability for agriculture. Lastly, PV plants can be built on marginal lands unsuitable for agriculture on their own - and the shade they provide reduces evaporation from the underlying soil, increasing water availability downstream.
Okay, that was some grade A class nonsense.
1) Solar panels *do* drastically reduce power output on cloudy days.
2) Clouds do *not* only block visible spectrum light "and a bit of IR". They're highly effective blockers of IR and UV as well. They do block IR and visible light better than UV (and there are relatively rare situations where they can actually enhance UV light via reflections), but they absolutely block all three, and do so well. See chart c in figure S1 / figure 6.
3) Only 3-5% of the sun's energy at Earth's surface is UV. IR is 52-55% and visible 42-43%.
4) Standard solar panels can't use infrared (too low photon energy), and are very wasteful with UV (still just one electron per photon regardless of photon energy) if they can use it at all (glass coated panels = UV blocking).
Literally everything you wrote was wrong.
Not the same thing. Standard AC distribution lines are not a cost effective means to distribute huge amounts of power from one side of a large country to the other. For that you need high power HVDC lines.
A nice thing about HVDC is that unlike AC, it also works well under seawater. Also, it shares power between disjoint AC grids (since it's always converted to the local waveform) and improves power quality on the distribution end. And prevents the cascading power failures that AC is prone to (aka, one part of the grid going out of sync with others). The lines themselves are vastly cheaper vs. how much power they carry, and the losses tiny, even over great distances. The main downside is that the terminals are quite expensive.
Exactly my reaction. If they say the project cost $679m to build, then it means just that: the project cost $679m to build. Not "one aspect of it" cost $679m.
A price of just over $1 a watt is superb. That's about what it costs to build a typical fossil plant - except that the cost to build a fossil plant is dwarfed by the cost of running it. Now, a fossil plant will have a 3x higher capacity factor, but still, this is highly competitive power. To put it in perspective, some of the new nuclear plants they're building in Europe cost over $10 per watt. Just to build, not counting operations and decommissioning.
Now, up to a given level of penetration, solar aides the grid by boosting the supply curve when demand is highest (the middle of bright sunny summer days). So up to that point, the baseload vs. intermittent supply argument is moot, and even the capacity factor doesn't play in (in a sunny location, at least), because the only capacity you need is to fill in those daytime peaks. At high levels of penetration however you start having to factor in increasing levels of peaking and/or storage. This can be somewhat offset by geographic smoothing and diversity of energy sources (solar + wind + others), but nonetheless your cost effectiveness will decline once your market penetration becomes large. Still, these are some superb numbers that bode very well for the future of solar.
It's best not to base it on "hours per day", and instead just look at capacity factors. Capacity factors on commercial scale solar plants range from under 15% to over 30%, depending on the tracking tech (none, single axis, dual axis) and plant design (as well as the most critical aspect, of course - location).
A nice thing about solar is that it tends to align pretty well with the demand curve, so up to a point adding actually makes grid operators' jobs easier, not harder. It also runs contrary to wind, which tends to blow stronger at night, and periods of low sun tend to most often be high wind and vice versa.
Tesla X + big boat = huge aero drag problem, not weight problem. And even concerning weight, freight truck tires are much more efficient than normal car tires.
2kWh/mi is what the electric cargo crate haulers at the Port of Los Angeles get. Maybe reduce the efficiency some for higher average speeds, but it's ballpark.
I think you need to cut down on your opiate dosage