When the Sun Briefly Turns Off the Power
February 28, 2015 10:47 AM   Subscribe

Europe's Solar Power Industry Braces For Solar Eclipse On the morning of March 20, 2015, a solar eclipse will pass over all of Europe, visible from Turkey to Greenland. A decade ago, that probably wouldn't have mattered to anyone except people who love astronomy... But now, 3 percent of Europe's electricity grid comes from solar power...

This more in depth article indicates: Hundreds of millions of Europeans will experience a solar eclipse ranging from 40% to 97% obscuration on March 20th. Iceland will have a total blackout.
posted by Michele in California (53 comments total) 17 users marked this as a favorite
 
Cool story.

I can't help thinking that they should just take a vacation day. Power the hospitals and other crisis centers and tell everybody else to expect a brownout and spend the day goofing off.
posted by anotherpanacea at 11:04 AM on February 28, 2015 [7 favorites]


Other than happening a little more rapidly, how is this different from sunset?
posted by jacquilynne at 11:05 AM on February 28, 2015 [27 favorites]


Clearly, the answer is more sacrifices to Sol Invictus to help him conquer the Devouring Dark. This was never a problem for the Romans, because they had the Old Time Religion!
posted by GenjiandProust at 11:11 AM on February 28, 2015 [26 favorites]


Time to throw some folks into a cenote.
posted by Stonestock Relentless at 11:12 AM on February 28, 2015 [9 favorites]


Wow, an interruption that happens for about 15 minutes, over a relatively small area? For a power supply that is down half of the day on average anyway? Which only makes up 3% of the grid? This has to be a manufactured problem.
posted by Mitrovarr at 11:12 AM on February 28, 2015 [52 favorites]


Luckily, we've known about the solar eclipse for a while.

And the award for Driest Sentence in a News Story goes to...
posted by Etrigan at 11:13 AM on February 28, 2015 [8 favorites]


Well, it is a little different than a sunset, as far as I can tell. Firstly, it affects a very large area, almost instantly, rather than a gradual onset thing that slowly moves across the continent. And the timing isn't great - morning for most time zones, it seems, when people turn on the kettle for the morning tea / coffee.

Overall, though? From my layman's perspective, it really does look a trivial problem.
posted by YAMWAK at 11:15 AM on February 28, 2015 [3 favorites]


Yeah, it's interesting that they have to do so much preparing when everything returns to normal in a few hours. I'm wondering what kind of incidents they're expecting.
posted by erratic meatsack at 11:17 AM on February 28, 2015 [1 favorite]


This has to be a manufactured problem.

I would think so, the main problem with solar and wind in terms of their network management impacts is predictability. Since this was highly predictable it seems likely to have been planned for in advance.
posted by biffa at 11:20 AM on February 28, 2015 [1 favorite]


We're all laughing now. But what happens if the sun doesn't come back? What if the moon blocks the sun forever? Then those coal plants won't look so unwelcome.
posted by Nelson at 11:20 AM on February 28, 2015 [34 favorites]


The PDF is interesting reading.

Dealing with sunrise/sunset is much easier because we're very predictable animals in aggregate -- anecdotally power companies can see the spikes from our alarm clocks going off in the morning. They're very good at matching supply to demand, and don't like surprises.

There's also a big imbalance in solar between the different countries to worry about -- Germany for instance is 50% of that 3%.
posted by RobotVoodooPower at 11:21 AM on February 28, 2015 [3 favorites]


But what happens if the sun doesn't come back?

The same thing that happens when your rooftop PV panels are buried under two feet of snow. No output. Six weeks now, and the temperature has barely gone above freezing twice. I had to actually pay an electric bill this month. That ain't right.
posted by Kirth Gerson at 11:31 AM on February 28, 2015 [5 favorites]


Kirth Gerson: "But what happens if the sun doesn't come back?

The same thing that happens when your rooftop PV panels are buried under two feet of snow. No output. Six weeks now, and the temperature has barely gone above freezing twice. I had to actually pay an electric bill this month. That ain't right.
"

I suppose warmers for the panels powered by the panels themselves somewhat defeats the purpose, and the water has to go somewhere...
posted by pwnguin at 11:34 AM on February 28, 2015


Given how darkish and cloudy it is on a March morning here in Iceland, I doubt most of us will feel the difference.

I'll be sure to be BRACED all the same.
posted by Aya Hirano on the Astral Plane at 11:38 AM on February 28, 2015 [5 favorites]


An electrical event you have this much warning for? It's nothing.

What PV system designers fear most is the winter sunrise. Cold silicon puts out very high voltages, and high voltages let the magic smoke out of your inverters. I was at a solar project up on the Ontario-Minnesota border last week: 25 MW (AC), all mono-Si modules, with a dawn temperature of -35°C and blinding reflections from the snow. The inverters were screaming
posted by scruss at 11:40 AM on February 28, 2015 [20 favorites]


So, I searched the main article, and there was no mention of night time. How is this a story? Solar power, like many renewables, are a bonus in the energy network, not the main feature.
posted by filthy light thief at 11:40 AM on February 28, 2015


This hysteria brought to you by Koch Industries.
posted by benzenedream at 11:43 AM on February 28, 2015 [11 favorites]


This battery ad brought to you by Elon Musk.
posted by runehog at 11:45 AM on February 28, 2015 [2 favorites]


Don't clouds obscure the sun every now and then too? Is that equivalent, or is the effect of cloud cover much more localized?
posted by cyberscythe at 11:49 AM on February 28, 2015


Oh no! Whatever shall they do about something that happens LITERALLY EVERY NIGHT
posted by Sys Rq at 12:07 PM on February 28, 2015 [1 favorite]


I would think so, the main problem with solar and wind in terms of their network management impacts is predictability. Since this was highly predictable it seems likely to have been planned for in advance.

The thing about solar and wind power and their predictability is that in large aggregate amounts the output tends to follow the law of large numbers especially when the individual turbines and panels are further apart from each other. The wind sites are also specially selected for consistency along with wind speeds and possible output (hence offshore turbines where wind is both consistent and faster).

The total contribution to the grid is not trivial. A loss of 3% of your power almost instantaneously? A single power plant going offline was the root cause of the 2003 Northeastern power blackout. You route through interconnects which may not often be used as heavily and haven't had their clearance checked lately and *bam* a 345kV wire sags into a tree and you have yourself a shitstorm.

Which is why everyone is on edge and ready for the worst.

Also, that Murphy guy is a fucking dick.
posted by Talez at 12:11 PM on February 28, 2015 [8 favorites]


YAMWAK: "Firstly, it affects a very large area, almost instantly, rather than a gradual onset thing that slowly moves across the continent."

It affects a large area over all but only a small fraction at anyone time. An Eclipse passes any particular area in a few minutes (say 5 minutes). If solar is 3% (5/120*.03) an overall reduction of 1/8 of 1 percent of generating capability during the duration of the event.

A lot of big numbers but probably less noticeable than a cloudy period to the grid.

Also for any particular plant there is a period of partial totality; so the panels will relatively slowly be darkened and lightened; it won't be like flipping an insolation switch.

While individual power plants might need to take special steps I can't imagine they will amount to much more than the precautions already in place for sunrise or fast moving rain.

I'm sure it'll be filling social media though as a talking point on the unreliability of renewable energy. I should just unplug from Tumblr and facebook right now.

On a different note I now wonder about the effect of solar eclipses on wind.
posted by Mitheral at 12:16 PM on February 28, 2015 [5 favorites]


Cloud cover is localized, and clouds move slowly, so region-wide PV generation varies very little from minute to minute. TSOs already model both wind and cloud coverage to predict renewable generation on any given day, since they have to either generate or purchase electricity to meet demand, and not all of these sources are available at a moment's notice. Unscheduled energy surging *into* the grid in a very short time is not something the TSOs have a lot of experience with, so they're being extra-careful.

The media is making a big deal of it -- for the power companies it's a mildly exciting event, and prior planning prevents piss poor performance.

BTW, ENTSO-E has a groovy web site which shows you all the sources of generation in a given day, and the load forecast for tomorrow, among other things.
posted by RobotVoodooPower at 12:20 PM on February 28, 2015 [1 favorite]


From this other article:
The spread of huge solar arrays across the EU means that more than 10 per cent of the continent’s electricity now comes from solar panels.

With power supply so reliant on the sun’s rays, the solar eclipse on 20 March might lead to blackouts.

Electricity system operators have warned that the eclipse poses ‘an unprecedented test for Europe’s electricity system’.
I was sort-a-ish studying stuff like this as part of my environmental studies degree ( that I never finished cuz life got in the way) and I still read a tad about it here and there. One issue you see with, for example, wind power is that when there is a storm, you get a sudden large increase in energy generation and it overloads the system. The system is really not prepared to deal with so much current coming down the lines so suddenly. They also have no place to readily store it.

I have less understanding of the technical aspects of solar, but, yeah, sudden, widescale disruption of supply is potentially an actual problem, not just someone making crap up.

And I think what these articles are really saying is that this is the first TOTAL eclipse in a long time -- the last one was 1999 -- and since the last one, we have a lot more solar and we really don't know what will happen, so people are concerned, and reasonably so. There are two to three solar eclipses per year. They are always paired with a lunar eclipse (separated by two weeks). But most eclipses are partial eclipses. Total eclipses happen years apart, thus the system currently in place -- built up to the degree it currently is -- has never experienced such an event and we don't know how the system will react to it. Since they are so rare, whatever we need to learn needs to be learned as quickly as possible, from as few incidents as possible, so that as the system grows, future eclipses are less likely to cause problems for the power grid.
posted by Michele in California at 12:22 PM on February 28, 2015 [1 favorite]


I was sort-a-ish studying stuff like this as part of my environmental studies degree ( that I never finished cuz life got in the way) and I still read a tad about it here and there. One issue you see with, for example, wind power is that when there is a storm, you get a sudden large increase in energy generation and it overloads the system. The system is really not prepared to deal with so much current coming down the lines so suddenly. They also have no place to readily store it.

Wind turbines in storm areas will either be feathered to match the wind speed (if it's mild) or just have the brakes put on completely and shut down. Natural gas peaking power plants can usually pick up the slack until the storm passes.

Solar doesn't really work like that though.
posted by Talez at 12:24 PM on February 28, 2015 [1 favorite]


Wind turbines in storm areas will either be feathered to match the wind speed (if it's mild) or just have the brakes put on completely and shut down.

Oh, sure. I just recall reading about some of the challenges in dealing with the fact that wind power is not a steady supply of power. So, for example, a lot of the power goes to waste precisely because when there is a big surge, sometimes you just cannot capture it. I don't recall all the technical details. I just recall reading about some of the challenges that happen and some of the things they are trying to pioneer to make the wind power system work better.

And I know even less about solar. But I don't think engineers and what not are simply being histrionic here to say "Yeah, we have some concerns. The current system, as it is now, has never seen an event of this sort and we aren't sure what that means for it."
posted by Michele in California at 12:31 PM on February 28, 2015


Michele: Sure, those happen, I think it was a tree falling in Switzerland that took down the entire Italian grid in 2003 after there were already some constraints in place, but those are unpredictable events. How is this not similar to a planned shutdown of multiple nuclear plants say, in the case of a generic technical fault coming to light? Most places in Europe peak in the early evening not in the morning so there should be plenty of capacity there to cope with a 3% hit.
posted by biffa at 12:40 PM on February 28, 2015


Because, you know, power company execs had no prior knowledge of the possibility of any kind of solar event at some guesstmated date and couldn't possible consider putting any thought toward considering, you know, their responsibility to their customers or the community at large. But, by Sol, there could be an impact on the bottom line. And it could happen. It probably will. Oh, great sadness! Who will pay for that crisis?
posted by BlueHorse at 12:51 PM on February 28, 2015


3% of a dollar is 3 cents. 3% of a billion dollars is something more than that. We are talking a widespread thing here. They are predicting brownouts. That sounds like a reasonable prediction to me.

I have already said twice that a) this is the first total eclipse in years (since 1999) & we don't really know what it will mean and b) I don't think the engineers expressing concern is some kind of histrionic overreaction. I don't know what more to say than that. If that doesn't make sense or you don't agree, I don't see how me saying it ten more times adds any value to the conversation.
posted by Michele in California at 12:52 PM on February 28, 2015 [3 favorites]


Given how darkish and cloudy it is on a March morning here in Iceland, I doubt most of us will feel the difference

Same here in Germany :) That being said, thanks for posting OP, I had no idea this was coming up! Will be sure to keep an eye outside that morning.
posted by photo guy at 1:04 PM on February 28, 2015 [1 favorite]


So, they'll deal with it like they've dealt with every bad winter storm ever (minus the storm part)?
posted by Brocktoon at 1:13 PM on February 28, 2015


Obviously solar power is just completely impractical. I really don't know what we were all thinking of.
posted by Segundus at 1:14 PM on February 28, 2015


Unscheduled energy surging *into* the grid in a very short time is not something the TSOs have a lot of experience with, so they're being extra-careful.

This bears repeating : 3% of your power going out is only half of the problem - if it's going to come back in 15 minutes, that is not a lot of time to balance things out on a continent-wide-grid level.
posted by Dr Dracator at 1:42 PM on February 28, 2015 [2 favorites]


The European Network of Transmission System Opterators for Electricity seem to mostly be worried about the rate of change in power output, which will be rather more rapid than the normal day/night transition. But they go on to say

"Based on evidence from the 1999 eclipse, we expect a depression in demand around the time of maximum obscuration from human causes - people stopping work and going to look at the phenomenon...We refer to this as the human - demand effect. This should depress demand before maximum eclipse – negative rate of change, and increase demand after – positive rate of change. These rates of change work in the opposite direction to the PV changes. From 1999 evidence we expect the demand effect to dominate the PV effect."

In other words, the drop in power consumption brought on by people going outside to look at the eclipse is expected to outweigh the drop in production brought on by the eclipse itself.
posted by mr vino at 1:42 PM on February 28, 2015 [4 favorites]


> I have already said twice that a) this is the first total eclipse in years (since 1999)

This will be a total eclipse for the Faroe Islands, Svalbard, the North Atlantic (part of Iceland) and the North Pole. For continental Europe it's a partial eclipse.

I wonder how large the impact is when you compare it for instance to a football World Cup final match.
posted by bjrn at 1:56 PM on February 28, 2015 [1 favorite]


Unscheduled energy surging *into* the grid in a very short time is not something the TSOs have a lot of experience with, so they're being extra-careful.

Naah. Virtually all climates - and certainly most places in Europe - share a reasonably common situation where the sun is completely obstructed for a brief period of time mid-day throughout most or much of the local grid causing everyone's solar systems shut down, and where the sun suddenly comes back out and all those systems spring back into action simultaneously. This is a situation that happens frequently enough that I'm sure that the TSOs have already experienced it and built in protections against it.

It's called a thunderstorm.

TSOs already have demand-based generators to deal with this - natural gas or even petroleum-based - that go from idle to full generation in less than 15 minutes and cease generating just as quickly based on fluctuation in demand. This problem, at this scale or greater, has existed for quite some time.
posted by eschatfische at 2:00 PM on February 28, 2015


What if it is a rainy day?
posted by notreally at 2:18 PM on February 28, 2015


-This thing may have a problem.

-No it won't. You just hate thing.

-Actually, I love this thing which is why I work on thing for a living. And why I'm prepare to deal with the problem, so don't worry.

-So you're getting paid! You made up the problem for money!

-I'm not actually going to get paid more for solving the problem with the thing.

-See! There is no problem with this thing. I am an amateur on the Internet but I have engineer's disease. So clearly you are making up the problem with thing.
posted by anotherpanacea at 2:20 PM on February 28, 2015 [23 favorites]


Yes, BUT, when engineers express "concern," they are obviously being hyperbolic and histrionic. Yeesh.

{/}

(Also, I <3 your reply.)
posted by Michele in California at 2:25 PM on February 28, 2015


We're all laughing now. But what happens if the sun doesn't come back? What if the moon blocks the sun forever? Then those coal plants won't look so unwelcome.

Easy, we just have to appoint a sacrificial king to be torn apart so the sun can rise again and spring can return to the land. It's a little known fact that it gets sunny right after winter in Greece because we regularly sacrifice prime ministers.
posted by ersatz at 3:21 PM on February 28, 2015 [7 favorites]


Greece does owe a lot of debt to Northern Europe.. perhaps we have a form of payment that would satisfy the Bundestag?
posted by Nelson at 3:28 PM on February 28, 2015


We are all Fire Nation now, is that what you're saying?
posted by SPrintF at 3:32 PM on February 28, 2015 [3 favorites]


Unless somebody moved Germany to the equatorial tropics, this eclipse is going to have less of an effect than progressing a week deeper into winter does. Places in the higher latitudes see progressively more dark time as they approach the winter solstice, and that really is widespread. This is a blip.
posted by Kirth Gerson at 3:34 PM on February 28, 2015


I bet the Europeans can't handle a Total Eclipse of the Heart, either
posted by thelonius at 3:35 PM on February 28, 2015 [11 favorites]


In fairness, every now and then they do fall apart.
posted by Etrigan at 3:59 PM on February 28, 2015 [6 favorites]


Here is the ENTSO-E technical report. Since Germany is the booster for PV and has by far the most PV capacity it is Germany that will be at the centre of any problems that might arise. Italy after that. The report mentions the downward gradient will be as much as 400MW/minute and the post eclipse uptick might be as much as 700MW/min, which is a lot when it goes on for an hour, broken down, the central problem is Germany which might get a drop off of 273MW/min and an upward gradient of 361MW/min. Germany has a lot of experience at managing intermittency as a result of being the big name in wind and PV so is used to internal and cross-border management of balancing, most especially drawing on Scandinavian hydro which can ramp up much faster than the gas that eschatfische mentions, but those are big numbers over an extended period. So I guess they'll be aiming to have as much baseload on and keep their fairly copious hydro, pumped storage hydro and gas in reserve for flexible response, wheeling in from France and the north as necessary to try and keep up, I guess much will depend on what is available if others are trying to do something similar. Not sure of Italy's flexibility, will be interesting to see how they both cope. Will be interesting to see the write up afterwards.

What if it is a rainy day?

It should be noted that the ENTSO-E calculation is based on starting from and returning to clear conditions, poorer weather will reduce the rate of change and make the situation easier to deal with.

It should be borne in mind that this is a transmission company trade association report. I'm not familiar enough with the politics of EU wide TSOs to offer the current political perspective but if the solution is reinforce the transmission networks to deal with more distributed generation then it can't be assumed to be without bias.
posted by biffa at 4:55 PM on February 28, 2015 [4 favorites]


This is an absurd bit of FUD for all the reasons noted above.
posted by humanfont at 5:38 PM on February 28, 2015


What's behind ENTSO's alarm mongering?

“In several reports, policy and position papers, ENTSO-E has pointed out that in order to guarantee security of supply, a series of policy and regulatory changes are needed to take into account the evolution of Europe’s energy mix.

Ah.
posted by Omnomnom at 1:00 AM on March 1, 2015


"I bet the Europeans can't handle a Total Eclipse of the Heart, either"

But a total eclipse of the heart is only 7m02 long. This is like two eclipses of the heart, and most of us barely have the one heart in the first place...
posted by sodium lights the horizon at 1:22 AM on March 1, 2015 [5 favorites]


In fairness everyone is calling for that, the old regulatory frameworks aren't a great fit for all the new issues raised by having lots of RE.
posted by biffa at 1:30 AM on March 1, 2015 [2 favorites]


Electricity production in Germany in week 12 2015. Here's a screenshot. Shows the dip in power production during the eclipse. This more detailed graph is also interesting. (Found via Reddit).
posted by Nelson at 8:26 AM on March 21, 2015 [1 favorite]


Shows the dip in power production during the eclipse.

Yep. And a more-than-compensatory spike in the hours prior.
posted by Sys Rq at 8:31 AM on March 21, 2015 [1 favorite]


German power net survives solar eclipse, unsurprisingly.
posted by scruss at 1:31 PM on March 21, 2015


« Older I've traveled over dried earth and floods   |   Right Place Art; Wrong Place Vandalism Newer »


This thread has been archived and is closed to new comments