Share My Creation ISS and LEO satellites tracker

antenna_finished.jpg

Hi all,

Here is my latest creation. As usual it is pretty much useless, but I had a lot of fun designing it and seeing it in real life.

As you may know I have been working in satellites activities and spent most of my pro life into Low Earth Orbite satellites. So here is a moke up of a ground station antenna, able to track any LEO sat you want.

The antenna is 3D printed from a design I made starting with a 6 meter dish antenna picture.
image005.jpg
The dish is now 20cm wide, has a real paraboloid shape and is moved by two stepper motors fully hidden into the pedestal of the antenna
the 3D models are available here
Two hall effect sensors and two magnets allow to setup the antenna horizontal and pointing to north at boot time. They are also hidden into the structure.

CAD_antenna.jpg


An ESP32 is also hidden in the basement and is the heart of the system. It computes the position of the satellites every 1ms and computes the azimuth and elevation angles from your antenna to the satellite.
electronics.jpg
A B4X App allows to visualize the track and to send parameters to the antenna : which satellites to follow, the Two Lines Elements and the time.

The orbit propagator is embeded in the ESP32 and uses the excellent SGP4 library ported by hopperpop
the firmware was directly inspired by Alex Chang project

The App is quite simple with a single activity which mainly uses GoogleMap and GooglemapsExtra libraries (thanks @Erel and @warwound )

screenshot.jpg


Finally two video of a real ISS tracking in my home :)
One at 16x speed and one much longer but with explainations !


 
Last edited:

freedom2000

Well-Known Member
Licensed User
Longtime User
if you are interested in satellites orbit determination, please sit confortably, take a beer (or two) and read this excellent paper.

SGP4 orbit determination

As said in the introduction: "The Simplified General Perturbations (SGP) model series began development in the 1960s (Lane 1965), and became operational in the early 1970s (Lane and Cranford, 1969). A variety of publications presented the mathematical theory, and the paper by Vallado et al. (2006) tracks the history of these publications. Unfortunately, there has never been a release of any kind of differential correction code to implement the SGP4 method in a systematic approach to create Two-line Element (TLE) data. With the increased number of observing sites, and the availability of low-cost high quality optical observations, it is desirable to have such codes. The primary uses would be to obtain a more accurate TLE from independent data, and to have the ability to examine covariance data to support mission operations (such as conjunction operations) "
 
Last edited:

mangojack

Well-Known Member
Licensed User
Longtime User
if you are interested in satellites orbit determination, please sit confortably, take a beer (or two) and read this excellent paper.
SGP4 orbit determination

The supplied link points to he B4A Google maps tutorial .. ?

Not that I am likely to read the Paper ... Your followup statement was enough for me ... I had to have 3 beers just to slow the brain swell.o_O
 

kimstudio

Active Member
Licensed User
Longtime User
This is a real fancy amazing toy! I just came back from Shanghai astronomical museum this afternoon and want to look up to the sky... Some nights I saw a quite bright star slowly (not very slow) crossing the sky, is it one of these LEO satellites?
 

freedom2000

Well-Known Member
Licensed User
Longtime User
This is a real fancy amazing toy! I just came back from Shanghai astronomical museum this afternoon and want to look up to the sky... Some nights I saw a quite bright star slowly (not very slow) crossing the sky, is it one of these LEO satellites?
yes for sure :)

you can see most of them either descending from north to south or ascending from south to north.
They cross the sky in less than 15 minutes.

If you see more than 60 in a line it is for sure the Elon Musk's starlink constellation !
 
Last edited:

mangojack

Well-Known Member
Licensed User
Longtime User
I deleted my post thinking I was diminishing attention from your excellent work / project ..


@kimstudio , and others interested .... another one just recently found , mainly for planets and the star system , an interactive night sky display based on your location.
I have been watching Saturn & Jupiter the last few nights.

 
Last edited:

hatzisn

Well-Known Member
Licensed User
Longtime User
Hi @freedom2000, can you please tell me, the two hall effect sensors are one for the x axis and the other for the z axis?

You've put some homework to me and I checked in Excel the calculated results for 2 hall effect sensors (magnetic fields) in the horizontal plane perpendicular to each other. I figured out that you do not need it designed like this (but I may be wrong). Can you please tell me if the second sensor is for the z axis? Further more, why do you need the z axis sensor if it is so?
 
Last edited:

freedom2000

Well-Known Member
Licensed User
Longtime User
Good questions !

Ok let's open the bowels of this antenna 👿

The azimuth axis is mounted on a roller ball bearing to insure a smooth rotation of the crappy motor I use. One interest of this design is that I have been able to push the 4 wires of the Elevation motor into the axis of the ball bearing. No tangle of wires inside the head nor inside the ball bearing !

rotary table.jpg

On the right of this picture you can see the hole for the Azimuth magnet
Into the foot of the antenna is enough (just enough) room to push a tiny hall sensor

Az hall.jpg

hall sensor Az.jpg

So now, when the magnet is close to the hall sensor I get signal and can calibrate the north.
At boot time the antenna will rotate and stop its motion when pointing always in the same direction "the north".

Same thing for the elevation. A hall sensor is hidden into the "bump" you can see on the foot. And into one of the conterweight is hidden the magnet.
hall sensor El.jpg
El hall sensor.jpg


  • Note that the counterweight is half the weight of the antenna dish and made with lead molten into a metal shape of the plastic counterweight. This is absolutely needed to keep the antenna balanced, as the dish weights 120g... So 60g of lead on each side, thus a total of 240g. Just small enough for the motor to be able to move all this !

  • Note also that the elevation calibration is performed when the antenna is calibrated to north (if you want the magnet to fly in front of the hall sensor) !

Of course you still have to align one side of the base to the north for all this to work. But you don't have to be precise as nobody will check that your antenna is really pointing to north :cool:

All what we want is to point the antenna always to the same direction at boot time even after a loss of power !
After this initial calibration the software takes care to count the steps provided to the steppers and to go back to "zero" when the pass is finished. This is done in a way to untangle the wires which are sitting around the azimuth motor shaft (the very long screw on the picture) ...
 
Last edited:

hatzisn

Well-Known Member
Licensed User
Longtime User
Thanks a lot for the explanation. I know what a hall effect sensor is due to a on-line tutorial but I have never used one. When you said two hall sensors I imagined that you use them with the two magnets because there is a minimum threshold of magnetic field for the sensor to function. What I imagined was that you place the two hall effect sensors along with their magnets in the horizontal plane perpendicular to each other in order to measure the constituents of the earth's magnetic field. I created thus a calculation of this and found that this is not needed as you can use only one and find the maximum voltage of the sensor when the magnetic induction of the magnet is parallel to the horizontal constituent of earth's one and they are pointing in the same direction. This way you could measure immediately the voltage of the sensor at boot and then move it slightly to one side. If the voltage decreases then you start moving to the other side until you get the maximum voltage (if it increases you keep on moving to the same side until the maximum). When you find it you know that this is the magnetic south (assuming the proper direction of the magnet) and the magnetic north is exactly on the opposite side (you still have to correct due to the difference in magnetic and geographical north pole). With this design you do not have to align the base to the north south geographical axis as the sensor-magnet-microcontroller system does it automatically for you.
 
Last edited:

freedom2000

Well-Known Member
Licensed User
Longtime User
To find real north I would have used a magnetometer (HMC58831 for instance).

But not needed in my case !
 

hatzisn

Well-Known Member
Licensed User
Longtime User
Thanks for the suggestion. I didn't know about it.
 

freedom2000

Well-Known Member
Licensed User
Longtime User
I remember when I was a child and dreamed of knowing where the satellites are. Excellent work! Congratulations!
thanks !

I do believe that this moke up as a great potential to explain to children "how it works". And possibly not only children (most of our young ingeneers have never seen a real antenna "tracking their babies").

I have just implemented a "time acceleration" option --> a lot of fun to see the pass with x10 speed (both antenna and ground track of course)
 
Top