6 ms·

Solar eclipse poses energy challenge

19 points by GotAnyMegadeth 12y ago

sxcurry 12y ago
On the surface, this seems really silly. What happens on a cloudy day? Don't the utilities need to deal with power output drops all the time?
- JoeAltmaier 12y ago
  Or, dare I say it, night?
  - wodenokoto 12y ago
    People use less power at night. This is why electricity is often discounted at night time.
    - JoeAltmaier 12y ago
      This will change with upcoming grid storage. Then we can generate all night and store it, spread the burden out and extend the life of our generation capacity. Oh, except for solar, It won't be helping.
- tjradcliffe 12y ago
  They do, but rarely across the whole grid at once. This event is like it becoming extremely cloudy everywhere for about five minutes, followed by a return to normal conditions. That's a fairly large transient, so it isn't really "solar power is variable" that's the story, but "solar eclipses produce an unusually large transient in solar output". The same is true of wind power, which in some areas (the province of Alberta, in Canada, for example) wind resources are concentrated in a small geographic area, so when the wind stops blowing all the wind power drops off the grid. Alberta has limited their total wind capacity to 3% of average demand to prevent this from creating instabilities. Distributed storage is going to be a significant part of our renewable future, which will mitigate these problems, but for the moment the growth of solar and wind capacity has outstripped storage capacity, so we can expect there will be a few unfortunate events along the way (which will stimulate NIMBYs to denounce the entire technological sector, quite likely.)
  - mikeash 12y ago
    It's not that fast. In London, for example, the eclipse starts at 8:25AM, and doesn't reach its maximum until over an hour later, at 9:31AM. It then takes another hour and change before it finishes at 10:41AM. It is unusual that this event takes place nearly simultaneously over a whole country-sized area, but it's not a five-minute thing.
- ape4 12y ago
  Also, its in the UK. Land of rain.
- ChuckMcM 12y ago
  Actually the panels on my roof generate about 55% of the power they normally would on a cloudy day. They get some efficiency gain by being cooler, and they lose some light insolation.
femto113 12y ago
Um, isn't there an "eclipse" for roughly 12 hours every day?
- mikeash 12y ago
  Yes, and that one corresponds nicely with reduced demand for electricity. This one doesn't.
  - femto113 12y ago
    According to the EIA [1] Peak demand (in New England, which I imagine is similar to the UK) appears to be around 7PM, which is well past the time significant solar production is occurring. [1] http://www.eia.gov/todayinenergy/detail.cfm?id=830 http://www.eia.gov/todayinenergy/detail.cfm?id=830
    - mikeash 12y ago
      The timing certainly could be worse, but this eclipse is right at the "morning ramp" period, where demand is substantially higher than during that big eclipse we get every night.

On the surface, this seems really silly. What happens on a cloudy day? Don't the utilities need to deal with power output drops all the time?

JoeAltmaier 12y ago
Or, dare I say it, night?
- wodenokoto 12y ago
  People use less power at night. This is why electricity is often discounted at night time.
  - JoeAltmaier 12y ago
    This will change with upcoming grid storage. Then we can generate all night and store it, spread the burden out and extend the life of our generation capacity. Oh, except for solar, It won't be helping.
tjradcliffe 12y ago
They do, but rarely across the whole grid at once. This event is like it becoming extremely cloudy everywhere for about five minutes, followed by a return to normal conditions. That's a fairly large transient, so it isn't really "solar power is variable" that's the story, but "solar eclipses produce an unusually large transient in solar output". The same is true of wind power, which in some areas (the province of Alberta, in Canada, for example) wind resources are concentrated in a small geographic area, so when the wind stops blowing all the wind power drops off the grid. Alberta has limited their total wind capacity to 3% of average demand to prevent this from creating instabilities. Distributed storage is going to be a significant part of our renewable future, which will mitigate these problems, but for the moment the growth of solar and wind capacity has outstripped storage capacity, so we can expect there will be a few unfortunate events along the way (which will stimulate NIMBYs to denounce the entire technological sector, quite likely.)
- mikeash 12y ago
  It's not that fast. In London, for example, the eclipse starts at 8:25AM, and doesn't reach its maximum until over an hour later, at 9:31AM. It then takes another hour and change before it finishes at 10:41AM. It is unusual that this event takes place nearly simultaneously over a whole country-sized area, but it's not a five-minute thing.
ape4 12y ago
Also, its in the UK. Land of rain.
ChuckMcM 12y ago
Actually the panels on my roof generate about 55% of the power they normally would on a cloudy day. They get some efficiency gain by being cooler, and they lose some light insolation.

femto113 12y ago

Um, isn't there an "eclipse" for roughly 12 hours every day?

mikeash 12y ago
Yes, and that one corresponds nicely with reduced demand for electricity. This one doesn't.
- femto113 12y ago
  According to the EIA [1] Peak demand (in New England, which I imagine is similar to the UK) appears to be around 7PM, which is well past the time significant solar production is occurring. [1] http://www.eia.gov/todayinenergy/detail.cfm?id=830 http://www.eia.gov/todayinenergy/detail.cfm?id=830
  - mikeash 12y ago
    The timing certainly could be worse, but this eclipse is right at the "morning ramp" period, where demand is substantially higher than during that big eclipse we get every night.