IEEE Power & Energy Magazine - November/December 2016 - 76

The path of totality will be 70 mi wide; any person, or
any PV cell, within that band will not have any sunlight during the period of totality. But totality will only last for a short
period, from about 1 min to a maximum of 2 min 40 s.
While the total absence of sunlight will only last for a couple
of minutes, the partial phases of the eclipse-from the time
when the disk of the moon first touches the rim of the sun
until it disappears-will last several hours. Also, areas lying
outside the path of totality will still see a partial eclipse and
experience enough obscuration to affect solar generation;
the duration and amount of reduction will depend on their
distance from the path of totality.
Every area of the continental United States, and adjacent Canada as well, will see some effects. Only one major
city, Nashville, Tennessee, will find itself right in the path
of totality. Parts of the Kansas City and St. Louis metro
areas will border it. Essentially all of the United States will
see more than a 50% obscuration of the Sun, with the only
exception being the northern tip of Maine, where 48% will
be blocked. In fact, the only major metropolitan areas with
less than 70% reduction are Los Angeles (62%), Phoenix
(63%), and Boston (63%). Here are the figures for a few other
major American cities: Atlanta 97%, Denver 92%, Salt Lake
City 91%, Chicago 87%, Dallas 76%, and New York 72%.
The United States has about 27,000 MW of solar nameplate capacity. Almost all of it is PVs. Approximately onethird of this solar capacity is distributed energy resources
(DERs), a.k.a. behind-the-meter solar. It isn't connected
to the bulk power system directly, only through local distribution systems. This is not a lot of capacity compared
to the approximately 844,000 MW of the combined Eastern, Western, and ERCOT Interconnections, it's about 3%.
Further, solar has a much lower capacity factor (about 13%

Watching an Eclipse
If you plan on viewing an eclipse, make sure you remember not to look directly at the sun. That's a sure way
to cause permanent blindness. Seriously-do not take
this warning lightly. And don't fall for some of the old
stories about watching it through sunglasses or a strip
of unexposed film. Use a #14 welder's glass or a specially made solar filter. You can order a set of five pairs
of cardboard "glasses" with solar filter material for $20
or lower. A photographic filter that's mounted between
glass or resin will cost you considerably more. You can
also purchase solar filter material by the sheet. Check
online. To photograph the May 2012 annular eclipse in
New Mexico, I used a cut-down sheet of filter material
slotted into a Cokin filter adaptor, which in turn screwed
into the front of the camera lens; though not perfectly
flat and plane parallel, this setup worked pretty well.

ieee power & energy magazine

overall) than more conventional forms of power generation.
So it's likely that actual solar output at any given time would
be lower than its 3% share of installed or nameplate capacity.
By the way, wind generation can be affected, too. Wind
speeds in the path of totality tend to decrease by 0.7 m/s, or
1.6 mi/h. This is so slight, though, that it would hardly be
noticed, nothing more than noise within the general variability of wind.
Local weather, especially cloud cover or lack thereof,
will play a critical role. Overcast skies in the 2015 European
eclipse lessened power generation prior to the event; hence
the reduction was a lot less dramatic than it would have been
if skies had been clear. Many things can get in the way of
sunlight besides lunar eclipses-smoke from forest fires,
industrial pollutants, dust, haze, and UFOs, to name a few.
Customer demand or load is impacted, too. Contrary
to what one might expect, load tends to decrease during a
solar eclipse. The temperature in the area of totality typically declines by 3 °C (5 °F); and it gets dark! But that area
is very small compared to the size of the power system.
A larger impact comes from the tendency of people, even
in the partially eclipsed areas, to turn off their stoves or
computers or whatever else they're doing and go outside
to view the eclipse firsthand. Studies have shown that the
resultant drop in demand is enough to partially offset the
net decrease in solar generation, but there is no general
consensus on the amount.
So what's the bottom line? I have a lot of confidence in
our system operators, specifically, those who control the
various regional transmission organizations, independent
system operators and other control areas, nowadays called
balancing authorities. Consequently, I believe that no untoward events will occur. However, prudence is advised, as is
cooperation. More on that shortly.

Dealing with the 21 August 2017 Eclipse
The main advantage system operators have this time is that
we know exactly what's going to happen and when it's going
to happen. This is completely unlike just about any other
emergencies we face-generators tripping suddenly, lightning strikes, multiple transmission outages, breaker failures, solar storms, and cyber or physical attacks. For the 21
August 2017 solar eclipse, all system operators will know far
in advance when sunlight will start to decrease at every solar
facility in their respective control areas, how much solar radiation will decline, at what rate it will do so, when it will start
to increase again, and so on. This is a rare luxury for operators! In a way, a solar eclipse is comparable to a storm cloud
passing over each facility, only we know about it hours/days/
weeks/months in advance. As Jordan Wirfs-Brock has said,
"A solar eclipse is a cloudy day that we can predict." WirfsBrock has also noted, "A solar eclipse-which lasts at most a
few hours-is not going to cause an energy apocalypse." We
would add, provided we're prepared. For more information
about being prepared, see "Watching an Eclipse."
november/december 2016

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - November/December 2016