5 ms·

How to Make a Sundial

97 points by app4soft 4y ago

teeray 4y ago
Very cool. Love the simplicity of paper, but it would be cool to get an STL output I could 3D print for something more weather-proof.
- app4soft 4y ago
  Here is sundial generator for laser cut, which produces SVG.[0,1] It might be easy to make a 3D model from generated SVG in Blender and export it to STL. [0] https://blog.singleton.io/posts/2018-07-05-glowforge-sundial/ https://blog.singleton.io/posts/2018-07-05-glowforge-sundial... [1] https://github.com/dps/sundial https://github.com/dps/sundial
defrost 4y ago
Hmm .. for anyone that wants to level up here . . . The Digital Sundial is a thing: https://www.youtube.com/watch?v=wrsje5It_UU https://www.youtube.com/watch?v=wrsje5It_UU Architectural tricks with shadows as time passes in a fascinating area all of its own and possible catnip for anyone with an interest in raytracing and projective geometry.
- teeray 4y ago
  I’ve often idly wondered if the reverse has been done: taking video capture of the shadows of objects over time and using the sun’s known position to generate models of the building. I was thinking that might be one clues some of the auto-generated 3D buildings use in, for example, Apple Maps.
  - defrost 4y ago
    My guess would be "yes" - at the very least within the spooky photogrammetry community where many images from far away planes | sats can be compiled to produce "best guesses" on ground structures and routine movement. I stress "guess" as its essentially impossible to uniquely construct higher dimensional objects from sliced signals (or shadows) squashed to a plane (or lower dimension). Another application area for this is remote exploration geophysics that "scans" a airplane (or drone or helicopter or car etc) across a landscape recording magnetics, gravity, radiomentrics, EM return strength, etc in order to build a best guess model of deeper objects of interest .. copper deposits, iron, geological structure, etc. This is the arcane art of building 3D models from "shadows" .. but not the shadows you were thinking of teeray :)
    - adolph 4y ago
      Muon Tomography is kind of like interpolating the interior structure of things using shadows, right? Cosmic ray muons have been used for decades to radiograph objects such as pyramids and geological structures. The technique of muon transmission imaging was first used in the 1950s by Eric George to measure the depth of the overburden of a tunnel in Australia. In a famous experiment in the 1960s, Luis Alvarez used muon transmission imaging to search for hidden chambers in the Pyramid of Chephren in Giza, although none were found at the time; a later effort discovered a previously unknown void in the Great Pyramid. https://en.wikipedia.org/wiki/Muon_tomography https://en.wikipedia.org/wiki/Muon_tomography
      - defrost 4y ago
        With a few additional twists .. rather than "simple" direct photon paths originating from an effective point source with some diffraction into Umbra, penumbra and antumbra you have a diffuse source (the entire visible sky) with varying strength from angle | amount of atmosphere and a spectrum of energies from primary high energy rays and scattered impact products of lesser energy. But, in essence, similar principle; a fixed position detector within a pyramid records the HE spectrum across a period of time and moved (or several detectors operate in unison) and the spectra are used used to reconstruct probable variations in density .. In airbourne radiometry the HE counts are recorded and calibration flights (lines at multiple altitudes) are performed to estimate the cosmic spectra coming in from above .. in order to subtract that from observed spectra to reach an esimate of spectra from the ground ( from U, K, Th .. radon, etc ).
ngcc_hk 4y ago
I wonder if you only know time and a sundial. Can you know where you are? Tl;dr Btw the pivot moment the first generation of Catholic astronomers in china (after all the chinese one killed and the Italian in jail) was restored their position as they are right by using sundial and their prediction (of equinox?). Always wonder about this time and sundial.
- tejtm 4y ago
  ... what is a sextant? https://en.wikipedia.org/wiki/Sextant https://en.wikipedia.org/wiki/Sextant
  - adrian_b 4y ago
    A sextant requires you to wait the local noon (the maximum rise of the Sun over the horizon), when measuring the angular height of the Sun allows you to compute the latitude and the time when the noon happened allows you to compute the longitude. Instead of a sextant, one could use a vertical rod fixed on a horizontal and plane surface (in a good sundial either the gnomon is inclined, or the base is inclined, or both, so it would be less appropriate for this task), and wait for the moment of the shortest shadow, when the length of the shadow gives the Sun's angular height and the time gives the longitude (and the direction of the shadow gives the North-South direction). With a more sophisticated instrument able to measure the 2 angles of the Sun's direction (height and azimuth), together with knowing the time (including the date, not only the time of the day) and together with knowing the local North-South direction, you could compute the latitude and longitude at any time. The local North-South direction could be given by an inertial compass. A cheap magnetic compass could be used, but then for high precision in computing the position, a detailed geomagnetic map of the Earth would be needed and a complex computation would be required to find a place on Earth where the angles of the Sun at the known time and the indication of the magnetic compass would match the observed values.
  - mellavora 4y ago
    Doesn't a sextant also require a rather thick book listing the positions of many celestial bodies at certain times/locations?
gorgoiler 4y ago
How can one amplify the sun’s motion across the sky, to make a sundial with faster motion? I would love a sort of sun vernier caliper.
- albert_e 4y ago
  using a prism, maybe?
  - carapace 4y ago
    I think you'd have to use a curved mirror or prism?
- adrian_b 4y ago
  One could use a rotating mirror, whose angle of rotation would be determined by a circuit with photo-receptors sensing the direction of the Sun, which could give any desired proportionality between the angle of the Sun and the angle of its projected image. One could also cheat and only give the illusion of a faster motion of the Sun, with a rotating mirror whose angle would be determined by a clock mechanism, i.e. which would need a knowledge of the time independent of the observed position of the Sun.
ninefathom 4y ago
This was absolutely fascinating, and a wonderful application. It seems to have a bit of trouble with daylight savings time- I think perhaps I might add a manual UTC offset selection- but on the whole, very well done.
arethuza 4y ago
There was a "Sundial" that I'm very pleased was never built. "Sundial" was the codename for a 10 gigaton H-bomb design produced by Edward Teller and his team - the gigaton primary being called "Gnomon". There was no military value in these monsters - they would have been enormous (probably thousands of tons) and would need to be detonated at extreme altitude. http://blog.nuclearsecrecy.com/2012/09/12/in-search-of-a-bigger-boom/ http://blog.nuclearsecrecy.com/2012/09/12/in-search-of-a-big...
- dtgriscom 4y ago
  > ... would need to be detonated at extreme altitude. Perhaps it would be more accurate to say "would need to never be detonated."
moritzwarhier 4y ago
What a great idea for a web app. The no-nonsense description sparks interest without any tricks.
atahanacar 4y ago

97 points by app4soft 4y ago

Very cool. Love the simplicity of paper, but it would be cool to get an STL output I could 3D print for something more weather-proof.

app4soft 4y ago
Here is sundial generator for laser cut, which produces SVG.[0,1] It might be easy to make a 3D model from generated SVG in Blender and export it to STL. [0] https://blog.singleton.io/posts/2018-07-05-glowforge-sundial/ https://blog.singleton.io/posts/2018-07-05-glowforge-sundial... [1] https://github.com/dps/sundial https://github.com/dps/sundial

defrost 4y ago

Hmm .. for anyone that wants to level up here . . . The Digital Sundial is a thing: https://www.youtube.com/watch?v=wrsje5It_UU https://www.youtube.com/watch?v=wrsje5It_UU Architectural tricks with shadows as time passes in a fascinating area all of its own and possible catnip for anyone with an interest in raytracing and projective geometry.

teeray 4y ago
I’ve often idly wondered if the reverse has been done: taking video capture of the shadows of objects over time and using the sun’s known position to generate models of the building. I was thinking that might be one clues some of the auto-generated 3D buildings use in, for example, Apple Maps.
- defrost 4y ago
  My guess would be "yes" - at the very least within the spooky photogrammetry community where many images from far away planes | sats can be compiled to produce "best guesses" on ground structures and routine movement. I stress "guess" as its essentially impossible to uniquely construct higher dimensional objects from sliced signals (or shadows) squashed to a plane (or lower dimension). Another application area for this is remote exploration geophysics that "scans" a airplane (or drone or helicopter or car etc) across a landscape recording magnetics, gravity, radiomentrics, EM return strength, etc in order to build a best guess model of deeper objects of interest .. copper deposits, iron, geological structure, etc. This is the arcane art of building 3D models from "shadows" .. but not the shadows you were thinking of teeray :)
  - adolph 4y ago
    Muon Tomography is kind of like interpolating the interior structure of things using shadows, right? Cosmic ray muons have been used for decades to radiograph objects such as pyramids and geological structures. The technique of muon transmission imaging was first used in the 1950s by Eric George to measure the depth of the overburden of a tunnel in Australia. In a famous experiment in the 1960s, Luis Alvarez used muon transmission imaging to search for hidden chambers in the Pyramid of Chephren in Giza, although none were found at the time; a later effort discovered a previously unknown void in the Great Pyramid. https://en.wikipedia.org/wiki/Muon_tomography https://en.wikipedia.org/wiki/Muon_tomography
    - defrost 4y ago
      With a few additional twists .. rather than "simple" direct photon paths originating from an effective point source with some diffraction into Umbra, penumbra and antumbra you have a diffuse source (the entire visible sky) with varying strength from angle | amount of atmosphere and a spectrum of energies from primary high energy rays and scattered impact products of lesser energy. But, in essence, similar principle; a fixed position detector within a pyramid records the HE spectrum across a period of time and moved (or several detectors operate in unison) and the spectra are used used to reconstruct probable variations in density .. In airbourne radiometry the HE counts are recorded and calibration flights (lines at multiple altitudes) are performed to estimate the cosmic spectra coming in from above .. in order to subtract that from observed spectra to reach an esimate of spectra from the ground ( from U, K, Th .. radon, etc ).

ngcc_hk 4y ago

I wonder if you only know time and a sundial. Can you know where you are? Tl;dr Btw the pivot moment the first generation of Catholic astronomers in china (after all the chinese one killed and the Italian in jail) was restored their position as they are right by using sundial and their prediction (of equinox?). Always wonder about this time and sundial.

tejtm 4y ago
... what is a sextant? https://en.wikipedia.org/wiki/Sextant https://en.wikipedia.org/wiki/Sextant
- adrian_b 4y ago
  A sextant requires you to wait the local noon (the maximum rise of the Sun over the horizon), when measuring the angular height of the Sun allows you to compute the latitude and the time when the noon happened allows you to compute the longitude. Instead of a sextant, one could use a vertical rod fixed on a horizontal and plane surface (in a good sundial either the gnomon is inclined, or the base is inclined, or both, so it would be less appropriate for this task), and wait for the moment of the shortest shadow, when the length of the shadow gives the Sun's angular height and the time gives the longitude (and the direction of the shadow gives the North-South direction). With a more sophisticated instrument able to measure the 2 angles of the Sun's direction (height and azimuth), together with knowing the time (including the date, not only the time of the day) and together with knowing the local North-South direction, you could compute the latitude and longitude at any time. The local North-South direction could be given by an inertial compass. A cheap magnetic compass could be used, but then for high precision in computing the position, a detailed geomagnetic map of the Earth would be needed and a complex computation would be required to find a place on Earth where the angles of the Sun at the known time and the indication of the magnetic compass would match the observed values.
- mellavora 4y ago
  Doesn't a sextant also require a rather thick book listing the positions of many celestial bodies at certain times/locations?

gorgoiler 4y ago

How can one amplify the sun’s motion across the sky, to make a sundial with faster motion? I would love a sort of sun vernier caliper.

albert_e 4y ago
using a prism, maybe?
- carapace 4y ago
  I think you'd have to use a curved mirror or prism?
adrian_b 4y ago
One could use a rotating mirror, whose angle of rotation would be determined by a circuit with photo-receptors sensing the direction of the Sun, which could give any desired proportionality between the angle of the Sun and the angle of its projected image. One could also cheat and only give the illusion of a faster motion of the Sun, with a rotating mirror whose angle would be determined by a clock mechanism, i.e. which would need a knowledge of the time independent of the observed position of the Sun.

ninefathom 4y ago

This was absolutely fascinating, and a wonderful application. It seems to have a bit of trouble with daylight savings time- I think perhaps I might add a manual UTC offset selection- but on the whole, very well done.

arethuza 4y ago

There was a "Sundial" that I'm very pleased was never built. "Sundial" was the codename for a 10 gigaton H-bomb design produced by Edward Teller and his team - the gigaton primary being called "Gnomon". There was no military value in these monsters - they would have been enormous (probably thousands of tons) and would need to be detonated at extreme altitude. http://blog.nuclearsecrecy.com/2012/09/12/in-search-of-a-bigger-boom/ http://blog.nuclearsecrecy.com/2012/09/12/in-search-of-a-big...

dtgriscom 4y ago
> ... would need to be detonated at extreme altitude. Perhaps it would be more accurate to say "would need to never be detonated."

moritzwarhier 4y ago

What a great idea for a web app. The no-nonsense description sparks interest without any tricks.

atahanacar 4y ago