Multi-day time lapse of a solar installation - Basic math

Over the next couple of weeks, I'm going to post a series of essays illustrating the process of capturing a long term time lapse. This project presented some issues that required a little thought in setting up my system. I thought this would be a good opportunity to share some of the challenges that were encountered and solutions to overcome them. There are a few unknowns regarding the project(as always) and I'm going to share my learning process to the solve those issues too.

A while back, I was contacted by a large solar installer and asked to shoot a time lapse of an solar array installation. The installation would be on the San Diego office of the American Lung Association. After a couple of bumps and derailments, the solar installation is finally commencing next week.

As I mentioned earlier, there were several issues that caused me to scratch my head. Here are some of the considerations to be addressed:

• The installation will take between 3-7 days
• The final time lapse video needs to be 1 minute
• Need to power a camera and intervalometer for multiple days of shooting
• The installation is on a flat roof with no locations that a camera can be mounted to in order to look down on the roof to capture the process
• Camera needs to be protected from potential weather and elements
• Camera(s) can't interfere with the workers doing the install

The basic math
There are a couple ways to determine how long a time lapse needs to be shot and how many shots per minute need to be shot. If the priority is creating a certain amount of movement in clouds or the sun, one will choose the amount of shots per minute to get the desired effect. Visit Timescapes to get a better understanding of all the elements you should consider.

If the final time lapse needs be a specific amount of time, then a little math needs to calculated to determine how many shots need to be shot. For my scenario here, I have a target time of 60 seconds for my final time lapse video. My final output video will be 24 frames per second, since the time lapse footage may accompany interview or b-roll footage for other videos. Now multiply 24(fps)x60(seconds of final video) and you get 1440 shots to fill a 60 second time lapse.

Now comes the trickier part. In a perfect world, the solar installer would give me a specific amount of time that the installation will take to complete. In reality, he told me this morning that it will take between 3-7 days depending on complications and other factors. My challenge is distributing those 1440 shots in a 3-7 day timeline.

For sake of simplicity, let's assume that the project will take five days. Each work day will be 7am to 5pm or 10 hours. 10 hours a day multiplied by 5 days equals 50 hours of total capture time. We then divide 1440(number of shots) by 50 hours and we get the total of shots per hour of 28.8 or 29. If I shoot one shot every 2 minutes, it will get me very close to making a 60 second time lapse.

As I mentioned before, the actual project time is fairly unknown right now. How I am going to deal with that issue will be addressed in a another post. Stay tuned!