APOD: 45 Days in the Sun (2015 Feb 21)

[APOD Robot]

45 Days in the Sun

Explanation: From January 11 to February 25 2013, a pinhole camera sat in a field near Budapest, Hungary, planet Earth to create this intriguing solargraph. And for 45 days, an old Antonov An-2 biplane stood still while the Sun rose and set. The camera's continuous exposure began about 20 days after the northern hemispere's winter solstice, so each day the Sun's trail arcs steadily higher through the sky. These days in the Sun were recorded on a piece of black and white photosensitive paper tucked in to the simple plastic film container. The long exposure produced a visible color image on the paper that was then digitally scanned. Of course, cloudy days left gaps in the solargraph's Sun trails.

<< Previous APOD

This Day in APOD

Next APOD >>

[/b]

[Antony Rawlinson]

Interesting that the position of the sunset barely moves, but sunrise moves markedly east throughout the period. Must be an effect of perihelion, but I don't understand orbital mechanics well enough to see the reason straight away. Is there a short explanation?

[Rules For]

Interesting that the position of the sunset barely moves, but sunrise moves markedly east throughout the period. Must be an effect of perihelion, but I don't understand orbital mechanics well enough to see the reason straight away. Is there a short explanation?

The migration of sunrise and sunset should be symmetric. I think the explanation for this picture might be that the photographic paper was curved inside inside the film canister (making it distorted towards the edges), and that the pinhole opening was not facing due south (making the distortions asymmetric).

[Simen1]

What does it take to make a digital solargraphy?

I don't mean scanning the film or photopaper, but with a digital sensor. It's not as simple as setting the camera to bulb. The scene may be much to contrasty for digital sensors. The sensor will get saturated and the battery will drain. An very dark ND filter will enable longer shutter times without saturation, but at the expense of thermal noise. Single exposures for hours will drown in thermal noise. Using a AC adapter in stead of a battery will solve the battery problem. Combining tens of thousands of short exposure images will be far to heavy for any normal computer. Tens of thousands of shutter actuations may wear out the shutter mechanism too.

Any suggestions on how to do this digital without scanning? Suggested equipment?

Why? While I like the look and charm of the canister film method, but when something seems hard to do and out of the ordinary i want to try that to get something different. In this case much better image sharpness, less distorted and better colors.

[geckzilla]

What does it take to make a digital solargraphy?

Here's my idea: Set the camera up in a safe environment with attached to its own computer and power supply converter instead of battery (kind of obvious). The camera needs a custom interval timer and to somehow dump the photos straight onto the computer's hard drive. Someone must know how to do this and it probably only works with certain cameras. At noon, a slightly longer exposure is needed to capture the daytime scenery. This noon shot will have to be scripted to automatically adjust between bright, sunny days and dim, cloudy days. At the end of each day, the script then stacks the images into a single one. If the camera is actively dumping photos onto the computer's hard drive it can add each one to the stack while the camera is busy. It's not that cumbersome for the computer in this way and that way you can also just delete the photos if you don't want to store them although keeping a cache of them might be useful for diagnosing unforeseen issues. Anyway, save the daily images and even if you go for a whole year, 365 images to adjust and stack is a lot easier than a few hundred thousand. No idea how to deal with the camera's mechanical wear and tear. Have a spare ready or maybe a video camera would work better?

[Chris Peterson]

What does it take to make a digital solargraphy?

Here's my idea: Set the camera up in a safe environment with attached to its own computer and power supply converter instead of battery (kind of obvious). The camera needs a custom interval timer and to somehow dump the photos straight onto the computer's hard drive. Someone must know how to do this and it probably only works with certain cameras. At noon, a slightly longer exposure is needed to capture the daytime scenery. This noon shot will have to be scripted to automatically adjust between bright, sunny days and dim, cloudy days. At the end of each day, the script then stacks the images into a single one. If the camera is actively dumping photos onto the computer's hard drive it can add each one to the stack while the camera is busy. It's not that cumbersome for the computer in this way and that way you can also just delete the photos if you don't want to store them although keeping a cache of them might be useful for diagnosing unforeseen issues. Anyway, save the daily images and even if you go for a whole year, 365 images to adjust and stack is a lot easier than a few hundred thousand. No idea how to deal with the camera's mechanical wear and tear. Have a spare ready or maybe a video camera would work better?

Don't use a photographic camera. Use some sort of webcam. No moving parts. I'd probably hook it up to something like a Raspberry Pi to avoid tying up a computer for months.

[Rusty Brown in Cda]

Interesting that the position of the sunset barely moves, but sunrise moves markedly east throughout the period. Must be an effect of perihelion, but I don't understand orbital mechanics well enough to see the reason straight away. Is there a short explanation?

The migration of sunrise and sunset should be symmetric. I think the explanation for this picture might be that the photographic paper was curved inside inside the film canister (making it distorted towards the edges), and that the pinhole opening was not facing due south (making the distortions asymmetric).

I disagree. I charted on a spreadsheet graph the times of sunrise and sunset for all of 2015 and found the same asymmetry plainly visible. I, too, suspect that it is caused by perihelion in early January, perhaps because the earth travels faster in orbit up to that point but slower thereafter as we move gradually farther away from the sun - i.e. uphill.

[Nitpicker]

Interesting that the position of the sunset barely moves, but sunrise moves markedly east throughout the period. Must be an effect of perihelion, but I don't understand orbital mechanics well enough to see the reason straight away. Is there a short explanation?

The migration of sunrise and sunset should be symmetric. I think the explanation for this picture might be that the photographic paper was curved inside inside the film canister (making it distorted towards the edges), and that the pinhole opening was not facing due south (making the distortions asymmetric).

I disagree. I charted on a spreadsheet graph the times of sunrise and sunset for all of 2015 and found the same asymmetry plainly visible. I, too, suspect that it is caused by perihelion in early January, perhaps because the earth travels faster in orbit up to that point but slower thereafter as we move gradually farther away from the sun - i.e. uphill.

For that kind of distortion to be real, the declination of the Sun would have to vary drastically within each day. This doesn't happen. The variation in declination is slow, gradual and periodic over a year, not a day. The distortion we see is mainly due to the pinhole not facing due South.