SatNOGS is a home built Azimuth and Elevation rotator that comprises of a number of 3D printed parts as well as commercial off the shelf items as well as some simple PCB’s and sensor modules. It is built around a pair of stepper motors that are connected to a worm and spur gear that turns a shaft. The speed and magnitude of the rotation is controlled by GPredict on a Raspberry Pi generating a standard command set and an Arduino motor controller which translates that into movement. Tracking can be controlled through a client / server arrangement on the SatNOGS website with the resultant signals captured through an RTL SDR dongle back to the user.
The project won its originators the 2015 Hackaday Prize and a slowly growing following. The project is beginning to provide a number of ground stations across the world. There have been a few demonstrations at Dayton and more recently at the NARSA rally.
Workington Club is a small club with a core membership of less than 20. We are involved, as are many clubs, in contesting, DXpeditions, training and special event stations. There is little that is remarkable about what we do. The club like to do a build each year, we’ve tackled simple antenna projects, kit builds and on occasions been involved softer projects like the RSGB Inspire project. We share knowledge with our closest club, the Furness Amateur Radio Society (FARS).
The SatNOGS project was chosen as a bit of collaborative project as well as a bit of a leap into the unknown for us. None of us had any real experience of 3D printing, between us we have a working knowledge of microprocessors and a passing awareness of the Raspberry Pi.
The team consisted of a core of 5 that looked after a specific part of the project
- Mechanical – Alex Hill, G7KSE (Also the project lead)
- PCB’s – Derek Atkinson, G1LZL
- Raspberry Pi – Daniel Smith, 2E0DNX
- Arduino – John Bracegirlde, 2E0JPI
- Antennas – Andrew Holland, G4VFL
We also received considerable support from the Cockermouth FabLab and the team there as well as the countless members of the SatNOGS forums as well as the SatNOGS team themselves.
The image blow shows the assembly drawing of the SatNOGS system in its current form (Version 3). An external framework of extruded 20×20 T Slot and hidden edge connectors retain bearing housings and gear bracketry. 3D printed parts were also used for sensor’s as well as the retaining rings. This is a very simple design that is similar to a lot of commercial rotators and the use of Nema 17 stepper motors gives a surprising amount of torque.
Antennas can either be to the design’s noted by the SatNOGS group or in our case a fairly common 3 element yagi for 2m and 7 element yagi for 70cm’s that allows this to be fitted inside the shack roof and away from the Cumbrian elements.
Both antennas feed an RTL-SDR dongle with the I&Q data being served through to the web interface.
The bill of materials provided proved to be a little inaccurate and we printed a few parts that weren’t necessary that got caught between versions 2 and 3 as with it a number of fasteners. This didn’t prove to be too costly a mistake but added to the bill. Most of the parts were sourced through common internet suppliers or eBay although the T-Slot proved to be expensive and no amount of search could find a cheaper alternative. Some of the T Slot fasteners were too expensive to purchase in the EU and an order through AliExpress was fulfilled, but very slowly. The stepper motors and their boards were easily procured as they are commonly referred to as RepRap parts (The RepRap was an early homebrew 3D printer)
Assembling the parts was a straightforward but time consuming process. There are many very tight areas that cannot be accessed easily with conventional tools and so modified tools were made. It also doesn’t help when your 4mm Allen key is borrowed by a family member and then dropped down a drain. The less said about that the better. Space in the envelope for belts and the adjustment of them is time consuming and the mechanical engineer in me had a few grumbles about some of the design decisions that were made. All said and done the assembly was completed without too much fuss.
Current 3D printing methods, even on >£2000 printers, can provide some ‘interesting’ results. Whilst the spur gear printed perfectly the worm proved to be more of a challenge and so 6 were printed and the best 2 used. Even these 2 needed some hand finishing and bedding in with the use of hand drill and silicone grease.
In addition to the mechanical parts 2 PCB’s were needed, one for power and one to hold the Arduino and is motor control boards as well as provide a link to and from the 2 optical sensors. These were manufactured using the toner transfer method and drilled to suit the components. There was only a small amount of through hole soldering to do in order to complete the boards.