Eclipse Production Dip

[steveholtam]

I saw quite a dip here in Sacramento. Anyone have anything cool to share?
ScreenHunter_599 Aug. 21 14.04.jpg

[tyab]

No to much drop here, surprisingly overcast and a lot of haze from the fire in Yosemite mitigated it. We watched it with a quick home made pinhole projector.

https://pvoutput.org/intraday.jsp?id=49477&sid=45049&dt=20170821

[J.P.M.]

Best guess here was the eclipse reduced the day's production by ~ 5.1%. The output curve looked similar to what you show.

[bcroe]

We are north of the complete eclipse; clouds were so heavy you wouldn't realize there was one if
you didn't know it. Clouds cleared enough in later minutes to actually see the partial eclipse. The
array was hardly affected; ran about 50% for the day at 63 KWH. Bruce Roe

[Mike90250]

Here's my production chart. You can see when the eclipse happened at 10:15, and when my irrigation pump kicked in for 3 hours (at 11:30am)
(sun hit panels at 8am, and left at 6m, when I used the microwave to heat dinner)
8-21-2017_eclipse.png

[EnergyKurt]

Got me pretty good in Kentucky

[peakbagger]

And yet there were not riots in the streets and rolling blackouts as the eclipse worked its way across the country

[SunEagle]

And yet there were not riots in the streets and rolling blackouts as the eclipse worked its way across the country

Due to the small % of solar power generation (~1% of all fuels used in 2016) throughout the US I would not have expected any power loss since the existing base generation capacity would be more than enough to support the brief outage from the sun.

Now if that % of solar increases significantly then I would think covering the power generation from other sources would be more of a problem. That % point for solar will be different across the US based on what type of power generation is available.

[BFW577]

Cool article on how California was effected.

https://www.washingtonpost.com/news/powerpost/paloma/the-energy-202/2017/08/22/the-energy-202-eclipse-tests-growing-power-of-renewable-energy/599b316a30fb0435b8208f5f/?utm_term=.2070e3613467

Screenshot_2017-08-22-10-57-41.png

[SunEagle]

Cool article on how California was effected.

https://www.washingtonpost.com/news/...=.2070e3613467

I am glad that the CA POCO's were ready and communicating to make sure their grid did not go down during the eclipse. But as I stated in another thread the amount of solar that CA rely on is maybe 10% of the total in-state power generation and it still imports close to 29% of the power it consumes which makes that 10% closer to only 6%.

Now if CA goes to say 40% of their power coming from solar I would say that the POCO's could be scrambling to keep the lights on if there was a wide spread outage (due to an eclipse or something else) without their spare base power being more than 3 times what it is today. That is going to be costly to keep around that many Peakers to fill in where RE can't do the job but maybe battery cost will be much lower and allow the state to ride through a 5 minute outage.

[sensij]

Now if CA goes to say 40% of their power coming from solar I would say that the POCO's could be scrambling to keep the lights on if there was a wide spread outage (due to an eclipse or something else) without their spare base power being more than 3 times what it is today. That is going to be costly to keep around that many Peakers to fill in where RE can't do the job but maybe battery cost will be much lower and allow the state to ride through a 5 minute outage.

That is a big IF. Yesterday, something around 20% of power dropped out over the eclipse, so we were halfway there. What if the same 31000 MW of demand existed, but 40% of it dropped out? That would mean another 6200 MW or so would have been needed to make up the difference. Non-variable sources contributed just under 27000 MW at the peak of this eclipse, but contributed 33000 MW later in the day, so that 6000 MW or so of capacity was clearly there, if it can be ramped in and out at the required rate. Peak load throughout the year has consistently been around 45000 MW, so there is evidence that at least another 12000 MW or so in capacity could have been available if needed. [all this is ignoring the impact of rooftop solar, not captured in CAISO's numbers, but not big enough to really change the thinking]

If you are talking about a world in which 40% of CA's *energy* comes from solar, that probably means mid-day demand is being met by something approaching 100% solar. 40% is really tough... using today's day ahead data, we can see that something around 732000 MWh of energy generation was planned, but 35% of that occurs between dusk and dawn, and can not possibly be directly fed by solar... and that is in late august. Winter months would be even tougher, I don't think you get to 40% without significant storage, and that changes the situation quite a bit.

This isn't to say the grid is robust against all possible scenarios, but handling the eclipse in CA doesn't look like it was much of a challenge on paper. The human factor in execution probably brought some excitement to the small number of people who are aware of it.

[SunEagle]

That is a big IF. Yesterday, something around 20% of power dropped out over the eclipse, so we were halfway there. What if the same 31000 MW of demand existed, but 40% of it dropped out? That would mean another 6200 MW or so would have been needed to make up the difference. Non-variable sources contributed just under 27000 MW at the peak of this eclipse, but contributed 33000 MW later in the day, so that 6000 MW or so of capacity was clearly there, if it can be ramped in and out at the required rate. Peak load throughout the year has consistently been around 45000 MW, so there is evidence that at least another 12000 MW or so in capacity would have been available if needed. [all this is ignoring the impact of rooftop solar, not captured in CAISO's numbers, but not big enough to really change the thinking]

If you are talking about a world in which 40% of CA's *energy* comes from solar, that probably means mid-day demand is being met by something approaching 100% solar. 40% is really tough... using today's day ahead data, we can see that something around 73200 MWh of energy generation was planned, but 35% of that occurs between dusk and dawn, and can not possibly be directly fed by solar... and that is in late august. Winter months would be even tougher, I don't think you get to 40% without significant storage, and that changes the situation quite a bit.

This isn't to say the grid is robust against all possible scenarios, but handling the eclipse in CA doesn't look like it was much of a challenge on paper. The human factor in execution probably brought some excitement to the small number of people who are aware of it.

I agree with what you said. Handling this eclipse was not much of a challenge. But IMO getting to 40% solar generation might be an uphill climb without a solid base of energy storage or fast starting gas fired Peakers. Using wind power can fill in the times that solar is not available but it isn't reliable. Another area of concern is when a high percentage of those reserve power sources are down for maintenance.

Back in the winter of 1989 Florida had a big freeze at Christmas. Due to a high number of generating plants that were down for maintenance and the cold temperatures up in GA my area was hit with rolling black outs due to the lack of power available. Since winter is a low consumption period the POCO's took down their equipment for repairs. No one had foreseen the problem of not getting the power needed from out of state.

As I said sometimes the Merde hits the fan at the most inappropriate time. Having multiple resources to generate power is a good thing.

[J.P.M.]

That is a big IF. Yesterday, something around 20% of power dropped out over the eclipse, so we were halfway there. What if the same 31000 MW of demand existed, but 40% of it dropped out? That would mean another 6200 MW or so would have been needed to make up the difference. Non-variable sources contributed just under 27000 MW at the peak of this eclipse, but contributed 33000 MW later in the day, so that 6000 MW or so of capacity was clearly there, if it can be ramped in and out at the required rate. Peak load throughout the year has consistently been around 45000 MW, so there is evidence that at least another 12000 MW or so in capacity could have been available if needed. [all this is ignoring the impact of rooftop solar, not captured in CAISO's numbers, but not big enough to really change the thinking]

If you are talking about a world in which 40% of CA's *energy* comes from solar, that probably means mid-day demand is being met by something approaching 100% solar. 40% is really tough... using today's day ahead data, we can see that something around 732000 MWh of energy generation was planned, but 35% of that occurs between dusk and dawn, and can not possibly be directly fed by solar... and that is in late august. Winter months would be even tougher, I don't think you get to 40% without significant storage, and that changes the situation quite a bit.

This isn't to say the grid is robust against all possible scenarios, but handling the eclipse in CA doesn't look like it was much of a challenge on paper. The human factor in execution probably brought some excitement to the small number of people who are aware of it.

Time to think out of the box. If/As less reliable power sources (solar/wind/etc.) carry more of the load, more backup will be necessary to maintain or improve current reliability/unintrerruptability in the system, either from online quick response generation or more storage or both, or other combinations of load management techniques. While that backup generation/storage will add to the cost of R.E. generation, it'll happen if the combo of R.E. and storage/backup generation is deemed as cost effective as other methods (say, nukes without required storage). At this time backup generation has the lion's share of the dealing w/ R.E. induced system generation variability. Storage will be the next tool, and maybe the next big money making opportunity. When ?? I've got some spare investibles looking for a home just now.

The grid will never be robust in all scenarios, but it will evolve as sources of generation evolve and never be the same as it is today, the same as it is different today than it was as Edison envisioned it.

The recent eclipse was a relatively easy event to plan for because it was known well in advance for both time and place, and a pretty good guess on max. impact was possible. As R.E. supplies a bigger chunk of the system load, things like clouds and wind will probably not so easy to plan for as an eclipse. I could see storage and quick start gen. both handling the variability problems with gen. handling most of the need at the outset, but storage taking a larger portion of the need as battery tech. matures.

[azdave]

Had 63% totality here in Phoenix at 10:33 local time yesterday. Wenzlick - Google Chrome 8222017 54451 PM.jpg

[kmm]

image_9805.jpg My chart from the San Diego area.