GOES-17 Satellite Imagery

I had seen some cool pictures online of images obtained using a SDR from GOES-17, a geostationary weather satellite over the west coast of the US. There is also a complementary GOES-16 over the east coast. There are also a few other similar satellites from different countries, in different locations, and different communication protocols. Due to the quarantine for COVID-19, now seemed like a good time to stay at home and point a dish at the sky with hopes of receiving images.

The first thing I had to do was to buy a few parts to go with my SDR

Premiertek 2.4GHz WiFi grid antenna ($75)

Nooelec GOES+ LNA and filter ($35)

A few SMA and type-N connectors to piece it all together (~$10)

Next up, I had to align the dish. I used dishpointer to get the azimuth, elevation, and skew from my current location. I then used a compass and bubble level app on my phone to set the dish up pointing in the general direction of where dishpointer told me. I knew I would refine this later as I found signal

My bike repair stand makes for a GREAT adjustable dish stand…

Aligned dish

Aligned dish

The antenna I bought was a 2.4GHz antenna intended for extending WiFi coverage. Since the signal from GOES-17 is 1.6941 GHz, there is a common modification that people have been doing to tune the antenna closer to this frequency. The modification is either to flip the reflector around, or install a small spacer between the reflector and the feed. I opted for a ~2.5cm spacer. This is a bit of a hack since the dipoles in the feed are still tuned for 2.4GHz, but it is enough to work.

Spacer installed to tune antenna closer to 1.7GHz (from 2.4GHz)

Spacer installed to tune antenna closer to 1.7GHz (from 2.4GHz)

Then I connected up my low noise amplifier/filter combo from Nooelec, and connected it to my SDR. The LNA would be powered by the SDR which has a built in bias-t circuit to supply DC voltage.

After a bit of toying around (a lot a bit to be honest… different spacer lengths, reflector layouts, antenna orientations, etc..)… I found the signal and it was coming in pretty clearly!

I then switched to Ubuntu and built/setup goestools. Goestools is a set of tools used to receive and decode the signal from the GOES satellite. Once I configured goestools to enable the bias-t on my sdr to provide power to the LNA, I fired it up and started adjusted the antenna slightly to try and minimize the viterbi error and minimize packet drops.

Once this was done, I then started up goesproc and it started processing data and saving images!

Here is the first thing that came in!

First image!

Since this satellite is geostationary…I was done! I left it running for ~24 hours and stitched together the images using ImageMagick to make some nice gifs.

GOES-17

GOES-17

GOES satellites have a few imagers and some can be requested by government agencies to point at specific things for certain reasons, these 2 gifs/sets of images are 2 additional images that get produced.

GOES-17 M1, changed where it was pointing part of the way through

GOES-17 M1, changed where it was pointing part of the way through

GOES-17 M2

GOES-17 M2

You can read about what it is pointed at and why here at the NOAA message list.

The GOES satellites also have a relay system where they relay smaller images from a few other similar satellites.

NOAA GOES-16

NOAA GOES-16

Japanese Himawari-8

Japanese Himawari-8

I also received some National Weather Service data from the satellite…





Reading more about this specific satellite (GOES-17), it has definitely had some problems. Most notable was GOES-17 ABI cooling failure which lead to them running reduced hours at different parts of the year to prevent the images from being over-saturated due to overheating.