I am building a rover using for controller a Pixhawk 6C and the GPS M10 unit. The connection for Control and Mission Planner is accomplished through optical fiber (The transceivers have 2 UART ports).
The issue I am facing is the following: when I power-up the whole system (PixHawk and Optical Fiber Transceiver) the GPS is unable to get a fix (the Sat count often stays at zero) and even after waiting 10 to 15 minutes it stays at 0.
If I connect the PixHawk through USB to Mission Planner the GPS get a 3D lock very quickly ( sees around 14 to 18 satellites), If I power the PixHawk through batteries without powering the Optical fiber transceiver same as previously.
But, as soon as I power up the transceiver, the sat count drops drastically (and even sometimes reaches 0).
If I let the pixhawk+ GPS on for a little while (3 to 5 minutes) and then fire-up the transceiver. There is still a drop in the sat count but not as big (losing about 2/4 satellites in the count) and even after a while the count starts to climb back to its original count…
What could cause this behavior? I suspect RF noise from the transceiver? And what would be the be the best course of action to mitigate this?
Thanks in advance for your time
Have a nice day
Alain
Some more information about the setting: the GPS Unit is on its mast at the front of the Rover, the transceiver is on the back under an aluminium plate ( as you can see on the attached picture), there is also a little 12V fan sitting on top of the box for cooling.
All the system is powered by a PM06 Power Module
RF noise from various equipment (including, but not limited to, USB3 devices and video encoders in VTXs and/or DVR recorders) can interefere with GPS reception, with symptoms exactly as you describe.
It is quite unlikely that a one-size-fits-all universal solution exists, especially given that the size of your machine does not allow to easily separate the offending circuits. Solutions can involve shielding your transceivers, better power source isolation, better wire management in general, using a higher GPS mast, adding a ground plane below the GPS antenna, etc. Occasionally, playing with orientation of the offending circuit also helps.
For instance, in the past I had similar problems with a recorder module of a Caddx Tarsier (a dual camera assembly where one camera was recording in 4k) for which I had to make a shield, which in one case was made of tin can, and in another one was 3D-printed and covered with aluminium foil. Both looked awful but worked well.
So one solution which is you can actually monitor the rf noise affecting the GNSS sats through U center application, if your GNSS is ublox chip, by doing trail and errors like OFF the trans receiver and ON the other subsystems to identify the noise. Even USB 3.0 will also cause rf noise a lot. And other solution may be place a copper tape below the GNSS unit.
Let me suggest a workaround: Why are you using optical fiber (whose transceiver seems to be the source of the interference) instead of a simple RC receiver? With mLRS, for example, you’ll have plenty of range and very good bidirectional Mavlink telemetry.
First thanks a lot for your time and your very helpful answers and sorry for the late reply, got a little bit busy lately…
Anyway, I follow your advice @MaxBuzz and put some shielding (copper tape) around the box containing the transceiver and for good measure I added some foil below the GNSS unit. It worked perfectly and now I have a solid count of satellites (around 18) and no drops when switching on the transceiver.
I also proceed the following way when starting the system: switching on the pixhawk first, wait a couple of minutes for the GNSS to establish a solid fix then I switch on the transceiver….
It works like a charm, got now a disco rover with the foils and all the LEDs blinking all over the place.
Anyway, thanks again for your time and your inputs
