RA encoder.
Dec encoder.
Espressif ESP32
Orwell Astronomical Society (Ipswich)
A Push-To System For The Tomline Refractor
The setting circles on the Tomline Refractor are original to the instrument, dating from 1874. Over the intervening decades, the ravages of time and corrosion have rendered them somewhat difficult to read, and many observers have struggled to use them to align the instrument to locate faint objects. This has encouraged instead the use of "star hopping", to first set the instrument on a bright, easily located object close to the desired coordinates, and then, by relating the stars visible in the eyepiece to detailed star maps, to "hop" successively from star to star until it is possible to "zero in" on the desired location. However, star hopping is itself a skill not easily mastered.
In the 2010s, the impressive capabilities of relatively low-cost electronics and control systems encouraged interest in the potential to equip the instrument with digital setting circles which could be easily read and used to align the telescope to a desired location without the observer requiring particular training or experience.
The 2015 International Astronomy Fair was held at Stoneleigh Park, near Coventry, on the weekend of 02-03 October. OASI members John Wainwright, Mike O’Mahony and then-Chairman David Murton attended, and were much impressed with the equipment on display. While there, David spoke with representatives of AWR Technology, who told him about push-to systems that the firm had fitted to various historic telescopes, including at Greenwich and Herstmonceux. Some weeks later, at an OASI committee meeting on 17 November 2015, David asked Martin Cook to investigate the possibility of engaging AWR Technology to fit a push-to system to the Tomline Refractor.
Martin contacted AWR Technology and, after several conversations, was offered a system consisting of encoders fitted to the RA and dec axes of the Tomline Refractor, a sidereal clock which had to be permanently powered on and a controller with a display showing the RA and dec to which the telescope was pointing. To find an object, the observer would first have to look up the coordinates, then move the telescope, while watching the display until it matched the desired position. This would not be easy when the observer was on the other side of the dome from the display! The price quoted was £2500 plus the cost of site visits plus the cost of manufacturing brackets for encoders plus the cost of fitting the brackets and encoders to the telescope. It proved hard to pin down how the project was going to proceed and, at a committee meeting on 31 May 2016, the idea was dropped.
Early in 2017, OASI member Adam Honeybell, a software engineer, approached Martin with a proposal to construct a push-to system, at very modest cost. In common with the approach of AWR Technologies, the system would utilise shaft encoders on the RA and dec axes of the Tomline Refractor. However, instead of displaying RA and dec on a display in the dome, it would use an Arduino Uno (microcontroller) to convert the output of the shaft encoders to a form which an astronomy sky-mapping app could use to display graphically the direction in which the telescope was pointing. There are ever-increasing numbers of such apps; they are relatively cheap (some are free) and run on phones and tablets, so the observer could hold in his/her hand a device indicating graphically where the telescope was pointing.
Martin, with access to engineering equipment, set about designing and manufacturing brackets to fix RA and dec shaft encoders to the telescope. The bracket for the RA encoder was relatively easy to manufacture and positioned the encoder to run on the inside of the RA wheel, out of harm’s way. Alas, the dec encoder was a different story! The only location where it could be fitted out of harm’s way was on a large gear wheel where the telescope tube meets the dec shaft. Working out the details of a suitable bracket required standing precariously on a ladder while trying to envision the interaction of fixed and moving components in 3D. Martin eventually hit on a solution to plan the engineering in detail by taking photos of the area with a ruler in the frame, then printing the photos full size to provide a large, scale view of the relevant components. The photos could be examined to provide a view of the associated metalwork from all angles; using this approach, a mount was devised for the dec encoder. By early March 2017, both encoders were fitted.
In the meantime, Adam was working on the software and associated hardware of the project. He aimed for both software and hardware to be compatible with modern, commercially available telescopes as well as the Tomline Refractor. The software would need to keep track of the pulses from the encoders, work out the RA and dec to which the telescope was pointing and pass this to the sky-mapping app. The hardware consisted of a plastic box housing an Arduino Uno with a Bluetooth shield attached. Connections between components were made using individual wires; unfortunately, there were many wires, because of which the hardware bore a passing resemblance to spaghetti!
With all necessary parts fabricated and assembled, testing could begin. To calibrate the hardware and software, it was necessary to establish how many pulses the encoders produced per rotation of each axis. This required turning the telescope through 360° in RA and in dec several times then working out the average figure, a feat not easy to accomplish with a 3-metre-long Victorian instrument. As the months of testing, redevelopment and retesting continued, it became evident that the Bluetooth connection was not stable, and it was replaced with a wi-fi connection. Later, Adam tracked down an intermittent problem to an extension cable used with one of the encoders. When this was rectified, he nevertheless remained unhappy with the accuracy of the system and a slow drift of the RA coordinate.
Work on the project slowed considerably when the company which employed Adam closed, and he subsequently took up employment as a contractor in Sheffield. Then, in March 2020, the Covid-19 lockdown made access to Orwell Park Observatory impossible, and the project stalled.
Over the next year, Martin followed posts on the Cloudy Nights astronomy forum by Speo (forum username). He had developed a system like Adam’s and was hoping for commercial sales. In January 2021, in a change of heart, Speo posted full details of the project on Cloudy Nights, free for use by self-builders. (Web link below.)
Speo’s system was based on the Espressif ESP32 microcontroller rather than the Arduino; this has the major advantage of an integrated wi-fi capability, enabling a much simpler design and eliminating much of the potential for loose connections which had proved so troublesome. The ESP32 tracks the number of pulses registered by the shaft encoders. The system connects to SkySafari, an app running on a tablet/phone. SkySafari repeatedly polls the ESP32 for an update on the pulse count received from the encoders since the last poll. SkySafari performs the complex calculations to convert the pulse counts into celestial coordinates and plots the position pointed to by the telescope (RA/dec) as crosshairs superimposed on a map of the sky.
Martin ordered the necessary parts to build a prototype on a piece of PCB stripboard and tested it with a couple of encoders. All worked well so he then ordered five custom-made PCBs from China (Speo had produced a set of drawings to facilitate PCB manufacture), with express delivery, for less than £20!
Prior to testing, it was necessary to set in SkySafari the number of pulses per revolution of the telescope around each axis. This arrangement is acceptable when the app is used with only one telescope but, if used with two or more different instruments, the observer would need to re-enter the values when connecting to each telescope. Martin contacted Speo via Cloudy Nights suggesting an update to the software to store the pulse counts in the system and upload them to SkySafari on start-up. The very next day, Speo replied advising that he had updated the code as requested. An outstanding level of support!
Martin assembled the components on one of the PCBs, mounted the PCB in a plastic box, loaded the latest software and installed the system on the Tomline Refractor. Testing proceeded smoothly. Total costs were less than £100.
OASI now has at its disposal a very rare instrument, a Victorian push-to telescope linked to a modern astronomy app, enabling celestial objects to be found quickly and easily.
RA encoder.
Dec encoder.
Espressif ESP32
Housings for the ESP32 and RA drive.
Rigel in the crosshairs of the system.
Adam working on the push-to.
James Appleton