Pictures and movies
Here you find pictures, movies and test results of some of the previously built prototypes. These prototypes are all equipped with a physical universal joint construction. You can recognize that by the aluminium fork that points to the pole, in our vicinity that is approximately 52°.
In all occasions, the mounts here are driven by one single motor. The precisely right motions of the horseshoe axis and the fork axis are passively retrieved by the universal joint parts from the rotating azimuthal ring. At the base you find the single motor that drives the azimuthal plane by an inverted time belt between two rollers. The two rollers are there to deal with slack in the time belt. There are many ways that you can think of to drive the azimuthal plane.
|Movie (176MB) Circumpolar slewed rotation||Movie (480MB) Circumpolar slewed rotation||Movie (77MB) Slewing from east to west|
A test session of sidereal tracking over a long time
We chose a star (Alcyone, Eta Tauri of the Pleiades) in the east in order to see to what extend the software and the mechanics are able to track an object at sidereal speed. The next picture shows the stacked and cropped result of the 71 integrations, 60 seconds each that we took. The individual pictures are calibrated for dark frames, no calibration has been done for flat fields. (Hence the crop..)
Before you reach a verdict about the potential it is worth to know that the tested mount was entirely built by hand. CNC machined parts will have a higher degree of precision. Most parts were built from the type of plywood that you find in concrete formworks. I 'm pretty confident that there are plenty of opportunities to do further improvement.
The stars in the individual pictures and as a result in the final stack aren't very round. A problem.. that turned to be a blessing in disguise. The main cause for the stars not being round clearly was some kind of speed error. I suspected that the main cause was the fact that I didn't do something like an alt-az alignment. I made simulations, found the way to have mounts like these alt-az aligned and developed an algorithm to compensate for speed deviations by counting pixels. Later I found what was the main cause: It was the combination of a wrongly entered hour angle and having an insufficient alt-az alignment of the mount.
We did not use any form of autoguiding or encoders, the speed of the steppermotor was calculated by an Arduino zero, based on the math that you can find on these pages. This test also showed other deviations such as slipstick effects caused by friction on bearings and elasticity of the construction.
In the models that I propose here I intend to drive all axes each with their own motor. By doing so we can do without the physical universal joint. As a result we will see a decrease of negative factors like elasticity, friction and slipstck effects that can cause trouble.
The test report of the stacked pictures of Eta Tauri
The following picture shows the stacked result of all the unregistered pictures that were used in the previous picture. The top-down component of motion is purely caused by the low altitude of the star above the horizon, it is caused by atmospheric refraction. The refraction causes the stars to apear higher on the sky. While the the star gets higher above the horizon, the airmass decreases which results in lesser refraction.
The east west component in every star trail is the result of speed errors of the mount. We learned from that! Now we have an algortithm to detect alignment trouble, so it is possible to get the mount running at the right speed.
Some pictures of two basic designs
This first example shows a Newton telescope. It is based on a 25 cm design, the OTA is balanced at the combined declination/fork axis when it is equiped with one of the popular Canon DSLR cameras. It is likely that the cradle combined with the OTA will be top heavy. That can be solved by bars for counter weights connecting the horseshoes at the lower corners, or by lowering the declination/fork axis in the cradle. That demands for larger diameter horseshoes.
The animation shows siderial tracking of a mount at a latitude of 52° north for an object at 40° declination. Viewed from the south east, hour angles of 18h, 21h, 0h, 3h and 6h.
The second design shows an example with a SCT. The dimensions are based on a balanced OTA consisting of a Celestron C14 and a SBIG ST10-XME camera.
The animation shows circumpolar sidereal tracking of a mount at a latitude of 52° north and an object with a declination of 52°. Viewed from the north west, hour angles 0h, 3h, 6h, 9h, 12h, 15h, 18h and 21h.