SatNOGS is an open source project to place satellite ground stations around the world and connect them through a network. It is part of the Libre Space Foundation and a Hackaday prize winner. In amongst a very clever system that no doubt is a gazillion lines of code are two very rewarding projects. Both of them involve a bit of radio and a bit of raspberry pi. The only choice is whether you can jump in and build a full rotator or a static ground station floats your boat. Oddly enough I have both. It is much better to hear from the leaders of the project so here’s a bit of video that explains it well.

The story starts a year or so ago. The local radio club used to like a project that meant a bit of collaboration between members. I volunteered to build the mechanical parts of the rotator and the electronics were to be completed be someone else. It didn’t quite work out but there are some good learning points.

Testing the mechanical set up without antennas

Non Rotator build

The rotator may be the big daddy and requires a bit of investment it isn’t the only ground station. A lesser performance can sound a bit dull but a static station can still be a great addition to the shack. The reception principal is the same as the rotator ground station. If I can over simply the whole system an Rtl-sdr is connected to a Raspberry pi and an omni directional antenna tops it off. There is currently no preferred design, it really is a roll your own system but I will explore this later.

Rotator build

A bit more complex but none the less a similar Idea. Instead of static antenna’s a device will be used to point higher gain antenna’s. The hard bit is that you need to either buy the pointing bit (expensive… Very expensive. Upwards of £600 for a couple of motors) or make one yourself. Fortunately the SatNOGS team provides all the details.

3D design and printing

The project makes use of some modern tools and techniques. Firstly the mechanical design has been achieved using 3D modeling. The designs have been pulled together using a modern parametric that is free. Now that’s good. It’s called FreeCAD. One of the things you need to get your head around is that you design each piece and then create an assembly. You only design the part once. If you change it you only change it once. This does not happen with 2D CAD.

The design is a mix of commercial parts like T Slot and motors as well 3D printed parts. These are accessed through the SatNOGS GitLab repository. In there you will find a series of STL files and a BOM (Bill Of Materials). This is your shopping list.

Print yourself v’s sending off

There is a pretty big question here, if you’ve not already got a 3D printer then you could buy one or you could rent one. This may not sound as daft as it looks. If you own your own printer then you’ll undoubtedly be familiar with STL files. So this is not really aimed at you. I didn’t own a printer at the time (I do have a very simple one now) so I could beg, borrow or steal one or a more realistic approach is to rent one, or rather rent time and materials on another. I was lucky that a local company has an Ultimaker 2 that was reasonable to rent. I sent them the stl files from the BOM and a couple of days later I got the parts. Services like 3D Hubs could do the same, although I’ve never used them they look like they offer a marketplace for owners of equipment to print parts by you uploading parts and they price the parts, you like the price you pay and they are then delivered.


SatNOGS use a Open Hardware Assembly Instructions system to produce very good assembly instructions. I do not intend in regurgitating these. There are a few things I noticed that may be worth sharing:

  1. It will take you time to assemble a rotator. more than you might think.
  2. You will almost certainly do things more than once. Something will not fit or rub or need adjustment.
  3. Some of the parts are expensive – I spent a long time finding the T Slot at a reasonable price ready cut. Well worth it in the long run
  4. You will almost certainly need to hand finish some of the parts. The worm gears spring to mind here

The system requires a micro-controller to be connected to the Raspberry Pi. These PCB’s will need to be bought and populated. Generally you need one of these but end up buying 3 or more from OSH Park, SEEED Studio or Dirty PCB’s ( I notice that the latter two also do 3D printing) for example, unpopulated. They will also need populating

Raspberry Pi – Really simple this bit. Download the image and follow the instructions on the wiki

Here’s a nice video of the whole build sequence that was supplied by the development team.

Using the network

There are some good articles on the wiki regarding scheduling observations and moving from development to production environments. I can’t stress enough that this is a really complex project under the hood. There is plenty for the budding hacker to mess about with but at the heart is a great opportunity to blend ham radio, coding and making into one project.

Open source – An invitation to tinker.

There is no reason why you need to be happy with a feature you don’t like. Its all open source so you can modify parts very easily through 3D modeling software and the controller firmware is also available for tinkering and if you are really keen then there is no reason why you shouldn’t roll your own client. Way beyond me.