@tom yes, will try and do so tomorrow.

actually it might not be feasible for me to snap actual photos I just made an image showing the basic concept

Hello, I'm Chris from Berlin and I just wanted to kind of just introduce myself and say hi and wave over and all. Because I'm likely to ask questions here in the near future but I'm not at the point yet where I have all the information I need to ask the right specific questions you know?
I'm mostly whirling about MAVLINK errors atm but am personally more on the PX4 flying-side, like I know my way around flight controllers and drones and stuff. =P

@modaltb said in GPS/Mag - Connection woes and Firmware versions:

OK OK, I see what you are talking about.

Yes J1009 is a UART... but not configured for TELEM, this is where we map MAVLINK instances to UARTs:

https://github.com/PX4/PX4-Autopilot/blob/master/boards/modalai/fc-v1/default.cmake#L12

There is/was a max MAVLINK instance count (at least in 1.10/1.11) that prevented enabling another instance.

Are you setup to compile PX4? It might be you just need to add another instance after the line called out above.

For the UART instances, this is what we have by default, and you could mess around by changing these to fit your need.

		GPS1:/dev/ttyS0 # UART1  / J1012
		TEL1:/dev/ttyS6 # UART7  / J1010
		TEL2:/dev/ttyS4 # UART5  / (internally routed)
		TEL3:/dev/ttyS1 # USART2 / J1002

The form factor is off of a "racer" style, where we don't have all the interfaces like PX4IO etc, so you might find some limitations if comparing to other flight controllers with more IO.

Hey man, just wanted to let you know that against all odds this actually worked.
I put in the line describing TELEM4 and built the firmware myself and - tadaa!
Thanks again.

Hey everyone,
just wanted to let you know that my fix got merged into master so it should work for everyone now...

I'd have loved to change the title of this to something a little more descriptive but it's too late now I guess.
anyway, cheers everyone, happy flying.

One more thing to add I can think of is
that on another drone the same setup but without the LTE module is working normally.
Please tell me if it would make a big difference to have actual photos and I'll see what I can do...

@tom yes, will try and do so tomorrow.

actually it might not be feasible for me to snap actual photos I just made an image showing the basic concept

Hi guys and gals
Sorry for coming up with such a probably weird and quirky issue again and also I really tried my hardest with the title here, so please have some lenience with me here...

I startup the drone with an LTE module (v2) attached and a camera connected to the VOXL-Flight's USB port.
Scanning wifi-IP's this VOXL doesn't show up anymore and I can't connect to it through a GS nor SSH.
Now the thing is, as soon as I pull the camera from the VOXL's USB the wifi on the VOXL starts up and it will connect to a GS and I can SSH in normally.

I tried plugging the camera back in and the wifi (wlan) on the VOXL actually ceases to work again.

Are there possibly any resource conflicts going on here?
Should I try connecting the camera to the LTE module instead? (As I understand this doubles as a USB-hub?)

thanks for any insights.
cheers

@Artur yes, try connecting to Mission Planner. Did you change any MAV related settings?
cheers

Hey once more.
After trying to go back to a basic and default setup today I narrowed my issue down to the QGroundControl version being used.
With the stable release the parameters actually do save (but I'm missing lots of other improvements in regards to PX4 now) so I guess this specific issue is solved
but
please let me ask, while you're here, what's the preferred sequence of steps to go back to an absolute default PX4 configuration for a VOXL-Flight (i.e. combination of flashing PX4-version / resetting parameters to vehicle defaults /resetting to firmware defaults /when to reboot etc)?
cheers and have a good week.

@RichieRich Hello there.
Awesome, thanks so much for taking the time as these are exactly the hints I'm looking for.
Will gather logs and parameters to share here, today, and try the combination "165" as well.

I'm using the combined "VOXL-Flight" units (one-PCB Flight-Core and VOXL).
Not using any VIO data, only GPS, mag, baro basically.

Will report back with more data. Thanks again and have a nice one everyone.

Hi,
I'm just checking in because this also concerns magnetic declination handling it seems... ‍️

@blakekrpec As I understand the ECL/EKF library integration so far, you can run several instances that each will fuse different sensor combinations (with individual blending) to accommodate for estimation state errors and the like....

@Chad-Sweet how many ekf_imu instances would you suggest running on a VOXL-FLIGHT in general, considering a decent workload already on the VOXL -part?
cheers

Hey,
I played around with them today a bit and found yes, your values did reduce the frequency with which the failure occurred so I'll continue tweaking there, thanks a lot.
Actually your link was double-helpful since I was looking for those specific parameters that regard the takeoff for quite some time now as well (without success...) ;D

I understand this is actually more closely related to PX4 but I very much appreciate the help anyway.
I sincerely don't even remember anymore why yesterday I thought this must somehow boil down to the fact that it's running on the VOXL-Flight hehe.