The tradeoff between good luck, bad luck, good planning, and bad planning
This entire trip was balanced by trying to use good planning to minimize the negative effects of bad planning, while trying to set us up to take advantage of any good luck, while not feeling disappointed by the possibility of bad luck.
Our trip to see the 2017 solar eclipse was worth it, but the memory of that eclipse was also tied to the memory of the subsequent 12-hour-drive from Glendo, Wyoming back to Denver, Colorado. We wanted to avoid any chance of that happening again, which meant:
1. Finding a hotel in totality that was driving distance from home. In our case, that meant somewhere in Cleveland-ish, despite the historically mediocre weather in April.
2. Finding a hotel within walking distance to food and groceries.
3. Finding a location to shoot the eclipse that was within walking distance of the hotel.
The location
In a stroke of bad luck, our original hotel canceled our reservation two months before the eclipse, claiming their planned HVAC upgrade was taking longer than expected. In a stroke of bad planning, I didn't have any backup hotel reservations in case our reservation fell through.
In a stroke of good luck, I was able to find another hotel in a Cleveland suburb that, according to Google Maps, was a block away from a sledding hill. This meant a potentially interesting place to photograph the eclipse, far better than my original plan of “hotel parking lot”, while still avoiding the need to drive anywhere.
Preparation beforehand
The Sunday before the eclipse, we went to the hill at around the same time as the eclipse the next day. It was gorgeous, clear conditions. Perfect for understanding where the sun (and moon!) would be, and the arc they’d trace across the sky the next day. I set up everything I had planned:
• A Nikon Z7 with a 2x teleconverter on the Nikon Z 180-600mm lens, sitting on top of the Sky-Watcher Star Adventurer 2i. Using my phone’s compass, I could point it north(ish), and knowing the latitude (41.36161°), I could angle it up towards where the North Star would be if it were nighttime. I wanted to verify that I could get it at the correct angle, and the star tracker would keep the sun mostly in frame. At 1200mm, it would be a pain to need to hand-track it for three hours.
• A Nikon D800 with a Sigma 35mm f1.4 Art lens. I used Stellarium before the trip to ballpark where the sun should be at the start and the end of the eclipse on a 35mm focal length. I wanted to verify that if I started shooting the sun in the top left corner, based on the timing and movement, it would stay within the frame by the end of the eclipse.
• Both lenses were connected to a Vello Shutterboss II external trigger. It allowed me to set up a timer to fire automatically every five minutes, and also meant I could take a photo without introducing camera shake from pressing the shutter button.
Heading out the day before was invaluable!
• I verified that the star tracker would be decent enough to keep the sun in frame, as long as I checked the framing every ~15-20 minutes.
• The D800 with the wide-angle lens was well framed, and verified that a shot every five minutes would create an aesthetically pleasing sequence as the moon eclipsed the sun.
• I also learned that it would be a good idea to set up close to the wooden fence on the hill. I had no idea how many people would come to this hill for the eclipse, so I wanted to be in a place that was unlikely to have people walking by.
• Wind would sweep over the hill, which normally wouldn’t be an issue, except that it would blow the solar filter off the D800. I would need to find a bit of tape to keep it on.
• We ran into a local, Joe, who was also out at the hill to prepare to shoot the eclipse the next day. He had picked that location because of the good view, and the added benefit of the surrounding nature, birds, and insects that could react during totality. Hearing someone nearby validated that it seemed like a reasonable place to shoot!
The day of
The night before and early morning of, a heavy rainstorm rolled through (bad luck). Thankfully, it left by sunrise, leaving mostly clear skies (good luck). I spent the hotel breakfast trying to listen to Cleveland morning news anchors predict the weather. Clouds were still projected, but they were looking like high-altitude cirrus clouds which shouldn’t impact the eclipse itself.
When we arrived to the hill around 11am, it was a gorgeous day, almost perfectly clear.
The first nibble out of the sun. Shot at 2:02:34pm. You can see the cloud cover passing over the sun, but it still shone through. A couple of small sun spots also dotted the surface.
My telephoto setup, albeit with my Z7 missing off the back. Shot at 2:39:33pm.
This cloud coverage was pretty consistent throughout the eclipse, and something worth remembering in a couple of decades. As far as weather apps are concerned, this would be partially or mostly cloudy, but when it comes to being able to photograph an eclipse, it didn’t make a significant issue.
The two weeks before April 8th, it was a near-daily ritual checking the weather, watching the projected cloud coverage go up and down in likelihood. Ultimately, it was entirely out of control, and worked out.
My other setup, my 14-year-old Nikon D800 with a 35mm lens, and the solar filter from 2017 held on by a bit of tape. This was lovely in its simplicity. I set it up around 12:30pm, started the trigger to fire every five minutes at 1:45pm, and then ignored it for three hours.
As the eclipse progressed, one of the sun spots was eaten by the moon. Shot at 2:45:43pm. An even deeper crescent ate the other sun spot around 25 minutes later, at 3:04:23pm.
At this point, I took my camera off to shoot a wide angle using the 28mm f1.8. It was probably fine to shoot directly towards the sun for a moment.
By 3:09:10pm, it felt noticeably cooler, and marginally darker, but still more like a significantly overcast day. If you stepped outside, you wouldn’t think “Boy, the sun sure is acting weird today.”
This was how the sun looked through the solar filter at (effectively) the same time as above, at 3:10:25pm. Just a sliver at this point. Note that we’re down to just a 1/15 sec exposure. The aperture doesn’t help, but there’s definitely less light at this point.
From here on out, after over an hour, you could feel time shift gears suddenly.
What was minutes ago a slow, indiscernible, shift watching a celestial body make the solar system’s slowest overtake in the passing lane was now watching fractions of a sliver disappear.
The left image was shot at 3:12:53pm, and then, less than a minute later, at 3:13:24pm, the last bit of sun until..
It was safe to remove the solar filter! Totality, at 3:13:29pm!
A stitched panorama from the top of the hill. I was shocked with how dark it truly was. Going back to preparation, I practiced the swap between lenses back and forth, but didn’t get the muscle memory in to swap the ISO and aperture settings to account for just how dark it got. This was 45-minutes-after-sunset darkness, not just twilight-kinda-darker darkness. In a couple of decades, I’ll need to remember to practice swapping settings during totality.
Shot at 3:15:04pm.
The advantage of the 2x teleconverter, with the high resolution sensor, means the flexibility for a lot of cropping. I wanted to come in close to this solar prominence just because my mind continues to boggle at the juxtaposition of the dark moon against the solar prominences that are otherwise blown out and impossible to view.
Fun fact: the solar prominence at the bottom of the frame is ~5 Earths tall.
Shot at 3:17:05pm.
On my list of shots I wanted to get, but didn’t in 2017, was a wide(er) angle shot. For all of totality in 2017, I shot at 1200mm. It meant getting a ton of detail close up, but I didn’t get any shots to emphasize the corona. Here, towards the end of the eclipse, I was able to catch some of it.
This is my favorite shot of the lot.
Shot at 3:17:11pm.
Shot at 3:17:16pm, five seconds after the shot above. Another lesson learned was that it’s really, really hard to get the eclipse in focus. All of the machine learning-backed auto-focusing in the world doesn’t work well on something that happens for three minutes every 18 months.
Serendipitously, my antique 500mm f8 mirror lens meant lots of practice shooting manual focus, so it was (pretty) quick and comfortable to change focal lengths and quickly refocus using the 180-600mm.
Note that the shutter speed is starting to creep back from a 1/4 sec up to 1/250 sec now.
The left image was shot at 3:17:17pm, a second later, and the diamond ring effect was starting to show. The right was shot at 3:17:18pm, another second later. This is a great capture to show what it’s like to be about a second later than I probably should have put the solar filter back on the end of the lens.
And finally, shot at 3:17:22pm, the solar filter safely back on the lens, and the barest sliver of the sun now peaking back out from the moon.
A minute later at 3:18:14pm, another greater sliver. And at this point on the right is when things started to feel back to normal again, at 3:24:53pm. By this point, the majority of people have left on the hill, leaving a handful of folks who were capturing the final transition out back to normalcy.
And when it was all said and done, I was also able to get the series of shots across the sky. There's no editing here, just the camera taking a photograph every five minutes, and then Photoshop to stack all of the layers on top of each other.
You can see how moving cloud cover caused the sun to look brighter or darker depending on what was over it at the time.
Some of the distortion on the top-left is from being at the outer corner of the lens.