My take on an autonomous satellite image collector grabbing from NOAA 15/18/19.
(Leave Behind - Collection Device)

Similar projects worth following
This idea came from seeing "rugged" pi designs and basically any style of field computer that I found. I originally wanted to just make a cool computer using a Pi that I could take with me on hiking/camping trips and not have to worry about it. Although I do still want to make a computer like that, I also wanted to get more familiar with SDR (software-defined radio) and Linux. After doing some research about SDR projects I learned that NOAA broadcasts images from their weather satellites. Now if that isn't awesome then I'm not sure what is!

So the plan is to stuff a bunch of electronics in a pelican case, build a QFH (quadrifilar helix) antenna and plop it on a roof somewhere and start collecting. Easy as that...right.

In addition to what was said above, this project is my first leap into RF related projects, something that I actively want to pursue and learn about. I also am super interested in the outdoors and tactical space so of course I wanted to add that type of spin to the project. All the attached files are pictures of my progress so if you want to follow along, that's probably the best way. I will post major updates in logs. 

Find the build photos here!

Feel free to let me know what I'm doing wrong, and hopefully how I can fix it. 

Thanks for reading!

View all 9 components

  • 4.0

    Nate Schroeder03/31/2020 at 19:50 0 comments

    Reflashed the RPI4 and used this info to try and get more consistent results as well as have a location to store the images.  

    I added the website to the project under "Images Server". This is just a basic website that tracks the next satellite it will be capturing along with the recent images and upcoming passes. I might flush out this website and post more of my RF projects on it, or create an entirely separate one. Using AWS S3 which was very easy to setup and getting running and it was free!

    Also 3D printed a couple of parts to flesh out the build. This included a mound for the RTL-SDR module along with an adapted for the antenna so I could use a tripod as a stand for it. Check the description for all the pictures of the build.

    Once I (hopefully) get consistent results with this revision, I am going to try and find a nice spot for it, nice and elevated somewhere. If anyone has a roof I can borrow, let me know. I might also add a SAWBIRD filter and LNA depending on what the images look like.


  • 3.2

    Nate Schroeder01/08/2020 at 03:55 0 comments

    Another pretty decent capture (I think). Still struggling with those black lines and very inconsistent results. Not sure if it's antenna placement, or what. However, I think my antenna works, only saying that because I got these pictures, so I am thinking the problem might stem from location or software. As the good scientist that I am, I changed two variables at once in hope that both work and my issues get resolved. I will also try and go outside, assuming I don't get blown away. (gale warning in effect). I will say though, this is addicting. It's just crazy that I am able to grab data from a XXX? million dollar satellite with some wire and PVC pipe! 

    Once these bugs are worked through and I know I can reliably get data from the sats, I will proceed with final assembly and then deployment.


  • 3.1

    Nate Schroeder01/07/2020 at 03:48 0 comments

    Eyes on! Re-flashing the Pi with a fresh Raspian image and following the tutorial seemed to work. This is the first pass that I was able to grab. Not the best, and not really sure what the lines are scattered about, but it’s a picture! I am really excited to further improve this. I am going to let it keep collecting for maybe another day before I strip the Pelican case down and paint the aluminum plate. I will also glue up the antenna as I know it works, you better not have just jinxed yourself. More to come!

  • 3.0

    Nate Schroeder01/06/2020 at 22:47 0 comments

    Well progress has been made on the antenna structure. I believe everything is wired right with the antenna. I followed this instructable but ended up modifying my antenna once I understood how it functioned. The antenna instructable has the coax as part of the loop. This revision I had problems with and could only hear the satellite after I tuned the antenna by rotating it and watching the waterfall on SDR#. The design I used actually just uses two loops of wire and are connected in a caddy corner fashion at the top of the antenna and one loop is connected to the coax shield and the other loop the signal. I found this wiring diagram the most helpful and easy to understand, although this antenna really isn't that complicated after you think about it for awhile, or a long while, a really really long while.

    I also got all the software and scripts running on the Pi, however I think I have an issue. My recordings are really long. Like 1 hr+. Which doesn't make any sense to me because each pass is around 15 mins. The predict software is spot on with my coordinates because when I look at the schedule of satellite passes they match with NY20 online tracking. The result is that I get garbage for results. Here is a link to a pass and the results (the audio and images). Around the 30 min mark I am getting (I think) pretty good reception, and it lasts for what I am assuming is the full length of the pass. However, when the recordings 70 mins, I think it is getting lost in the noise. I plan on doing a fresh install on the Pi and redoing the scripts because maybe a dependency for GQRX is messing with rtl_sdr? If anyone has any suggestions about this, it would be greatly appreciated. I modified the rtl_fm command that runs to correct for the RTL's PPM offset so I think that is correct. Here is the tutorial I used for the Pi automation. It uses a Noolec SDR so that's why the rtl_fm command is slightly different, I think. This is starting to become a little bit of a rabbit hole. If I do a fresh install and still can't get clear results I am not sure what should be next. Maybe a nice paper weight? Nah, I'll keep messing around and see what happens.

    Looks cool nonetheless. I could only imagine what people driving by looking in the window think I'm doing. If I told them I was talking to satellites vs. summoning aliens I'm not sure which would be more believable.

    (The electrical tape is temporary. I want to make sure it works before gluing it up in case I need to take it apart.) 

  • 2.0

    Nate Schroeder01/02/2020 at 03:04 0 comments

    I would say the collection "box" is about 95% of the way there. Besides paint and some last minute wiring tiding upping (lol), and painting the plate, it's about done. I also have been looking into RF shielding because upon further research I believe that the switching power supply is quite noisy so I'm thinking of wrapping that in aluminum foil. If that's a stupid idea please let me know. Upon preliminary testing using GQRX I am able to get FM broadcasts and UHF/VHF transmissions using one of my Beofeng radio antenna's in a roundabout way using some unnecessarily long cable and adapters. Like I said I've never used and SDR or SDR software so I am not sure what is normal. Radio stations are extremely clear, and I picked up the local NOAA broadcast but it is a little noisy. I tried the SDR on my windows machine and got better results so I am thinking that the power supply is adding some noise? I am going to pickup some PVC pipe for the QFH antenna and will be using this link to build it. Also going to be adding a coax loop which I've heard reduces noise and will be shielding the exposed connections. With any luck I will have some satellite images by the weekend. I will keep everyone, or anyone who is actually reading this, updated! I am very excited with how well it is coming along, it's crazy how accurate CAD is. The drawing that I printed from my CAD sketch for the top plate to cut to size and drill holes was extremely accurate, and I did not measure one thing for this project. All models were from Gradcad.

    (BTW: Those exposed wires by the PSU are not live, the switch on the side of the box attached to the outlet is turned off, safety first :D)


  • 1.1

    Nate Schroeder12/27/2019 at 02:28 0 comments

    Cut my aluminum mounting plate out. I will be painting after I do a final mock up. Ordered all my parts for the project so waiting on that. Starting to come together.

  • 1.0

    Nate Schroeder12/23/2019 at 04:14 0 comments


    Just created this project so this update will probably be the shortest one. Worked on modeling a lot. Trying to plan everything out in the case before actually cutting into it. Acquired some scrap sheet metal, measuring 3/32" thick so whatever gauge that is. Probably overkill for just mounting electronics but what I have. Starting to print out drawings of each component in order to get a physical idea of where things will line up. 3D printed the panel mounts for the case which can be found here. Also starting to think about/work the problem in regards to heating and cooling along with weather management. I obviously want the case to be weather resistant as it will be sitting on a roof, but I don't want to restrict air flow to the Pi and cause overheating. I believe it will get quite hot in there because I have a power supply, Pi, and RTL-SDR module running. I am wanting to add some sort of vent but need to find/make the right kind to keep as much weatherproofness, it's fun making up words, as possible. And it needs to look cool...always needs to look cool. If anyone has an suggestions, please let me know. Also I'm thinking of using those waterproof pass-throughs for the antenna feed line input so if anyone has any experience with that I'd love to hear about it. 

    I have a couple other projects going on so this might not get updated as often as I want to but that is life. 


View all 7 project logs

Enjoy this project?



andrewmurphy83 wrote 08/23/2022 at 22:05 point

rtl-sdr has the fm filter and your other problem could be the power supply that might be throwing hash out and interfearing

  Are you sure? yes | no

andrewmurphy83 wrote 08/23/2022 at 22:04 point

The black lines is from static or interfernce of some come i would recomend put a fm filter u can get and get a noaa sawbird filtar from amazon and or newlec i can get you links if you need it the fm signal block also can get there or from another sdr sights as well

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates