Equatorial Mount
Introduction to Equatorial Mounts
Due to the Earth’s rotation, distant objects in the sky moves around by one cycle per day. Thus if we try to take long-exposure pictures of stars, they move away and create “star trails” on the pohtos. An equatorial mount is a mount for cameras or telescopes that compensates for Earth’s rotation by having one rotational axis parallel to the Earth’s axis of rotation and spinning with the same angular speed as the earth but opposite direction so that stars appear stationary.
Source: HKU Department of Physics
Hardware Design
In my design, the camera to be used is mounted on a shaft which would be aligned to the Earth’s rotation axis. The shaft is driven to rotate at the same angular speed as the Earth via a worm drive.
When the camera is in Bulb mode, as long as the shutter pin and the ground pin in the camera’s built-in shutter release port remain shorted, the shutter remains open. To control this, I designed a MOSFET switching circuit.
The servo, MOSFET and an LED are all controlled by an Arduino Pro Mini microcontroller. I added a BluetoothLE module so that I can control the whole device wirelessly using my phone.
As for its casing, I created 3D models in Autodesk Inventor and got it manufactured by a CNC milling contractor.
Software
Now that I have several functions for my equatorial mount, I came up with the idea of defining a variable for each of the functions and assigning values to the variables which indicate the state of each function, for example, tracking/not tracking, exposing/not exposing… I didn’t actually realise that this was the way state machines work until afterwords when I read about them.
To control when to change states and execute these functions, I developed an Android application so that my phone could connect with the Arduino via Bluetooth and control all the functions of the mount. The application was developed in the MIT APP Inventor.
Testing and Evaluation
After finishing building my mount, I experimentally tested its accuracy using a laser. I attached a laser to the spinning shaft and set different angles for the motor to rotate. Then, by measuring the linear displacement of the light spot on the wall, I could calculate the real angular displacement and compare with my set value. Thus I could obtain my mount’s percentage error for different tracking times.
From the graph I can see that the percentage error is quite big for very short tracking times and much smaller for longer tracking times. This, I suspect, is because the servo has a fixed absolute error which gets diluted for large angular displacements.
When I tested it in a real nightscape photography session, I was amazed to see that it could support a 240-sec exposure, without clearly visible star trails even when zoomed in. Normally, with that combination of camera and lens without an equatorial mount, I could only shoot a maximum of 30 sec before everything gets spoiled by star trails.
Here is a video demonstrating my Android app controlling my equatorial mount: