@daniellee This log shows a normal flight happening in altitude hold mode. RC inputs are coming in and motors are spinning. Was there some issue with the flight this log was captured from?

@Jetson-Nano Can you rebuild to add UAVCAN? The code is here: https://github.com/modalai/px4-firmware/tree/v1.14.0-2.0.133-dev

@Eric-Katzfey thanks. before I did sdk reflash I found out the px4 params were somehow corrupted. I have tried to reproduce these values without success.
what made it work is taking a similar drawn read the params usign px4-param show -a and writing the to the corrupted one.

@Mustafa_Cayci You can get the Starling 2 Max with a ToF sensor for depth mapping indoors (The sensors don't really work outdoors). https://www.modalai.com/products/starling-2-max?variant=49129974923568

@gerard I also just remembered we had a stint of bad MCBL-00010's a few months ago...
We have since cleared that up, but there is no telling how long ago you may have ordered that cable...
Please see this thread for guidance and if you do think your MCBL-00010 is at fault, we will replace your M0048, just reference both threads in your RMA (this one and this one -->) https://forum.modalai.com/topic/4632/flight-core-2-firmware-update-through-qgc?_=1781284416780

Hi @Jetson-Nano,

Can you please clarify the following - in your original post, you mentioned that the third party ESC also had issues, however in the post above you said no issues. I am quoting your notes from above:

"
Scenario 3 – Third-Party ESC Evaluation
To further isolate the issue, we replaced the VOXL FPV ESC with a NeutronRC X-CLASS 4IN1 ESC V2 (3–8S, 70A) configured using the DShot protocol.

During testing, we observed the following:

ESC temperature increased rapidly, reaching approximately 80°C to 90°C within two minutes of flight.
One motor would subsequently stop during flight.
The aircraft would lose stability and crash.
"

I just wanted to clarify what actually happened and then we can investigate further. Than, you for providing the diagram, i think it looks OK.

I have never seen this behavior where the ESC reports incorrect voltage and these short spikes in temperature. This may point to a hardware issue on the ESC board, but i do not actually know where it might be or what would cause it. Have you visually inspected the FPV ESC that has this issue to see if any components appear damaged or anything looks unusual?

Alex

@astro_phil ,

I just tested your configuration on voxl2 mini

ov7251 connected to M0135 JL imx214 connected to M0135 JU M0135 interposer connected to Voxl2 mini J7 cameras are connected to hardware camera slots 2 and 3 respectively

This particular board is using an old SDK but it should not matter, kernel variant 2.0.0

hw platform: M0104 mach.var: 2.0.0 -------------------------------------------------------------------------------- voxl-suite: 1.3.3 voxl-camera-server -l DEBUG: Attempting to open the hal module DEBUG: SUCCESS: Camera module opened on attempt 0 DEBUG: ----------- Number of cameras: 2 DEBUG: Cam idx: 0, Cam slot: 2, Slave Address: 0x00E2, Sensor Id: 0x7750 DEBUG: Cam idx: 1, Cam slot: 3, Slave Address: 0x0020, Sensor Id: 0x0214 DEBUG: Note: This list comes from the HAL module and may not be indicative DEBUG: of configurations that have full pipelines DEBUG: Number of cameras: 2 dmesg: [ 209.617719] CAM_INFO: CAM-ICP: cam_icp_mgr_hw_open: 3879 FW download done successfully [ 209.648730] qcom,camera ac50000.qcom,cci:qcom,cam-sensor3: Linked as a consumer to regulator.59 [ 209.649003] qcom,camera ac50000.qcom,cci:qcom,cam-sensor3: Linked as a consumer to regulator.55 [ 209.649103] qcom,camera ac50000.qcom,cci:qcom,cam-sensor3: Linked as a consumer to regulator.60 [ 209.649137] qcom,camera ac50000.qcom,cci:qcom,cam-sensor3: Linked as a consumer to regulator.79 [ 209.666156] CAM_INFO: CAM-SENSOR: cam_sensor_driver_cmd: 918 Probe success,slot:3,slave_addr:0x20,sensor_id:0x214 [ 209.666263] qcom,camera ac50000.qcom,cci:qcom,cam-sensor3: Dropping the link to regulator.79 [ 209.670389] qcom,camera ac50000.qcom,cci:qcom,cam-sensor3: Dropping the link to regulator.60 [ 209.670476] qcom,camera ac50000.qcom,cci:qcom,cam-sensor3: Dropping the link to regulator.55 [ 209.670541] qcom,camera ac50000.qcom,cci:qcom,cam-sensor3: Dropping the link to regulator.59 [ 209.688989] qcom,camera ac50000.qcom,cci:qcom,cam-sensor2: Linked as a consumer to regulator.59 [ 209.691170] qcom,camera ac50000.qcom,cci:qcom,cam-sensor2: Linked as a consumer to regulator.55 [ 209.693356] qcom,camera ac50000.qcom,cci:qcom,cam-sensor2: Linked as a consumer to regulator.60 [ 209.693454] qcom,camera ac50000.qcom,cci:qcom,cam-sensor2: Linked as a consumer to regulator.79 [ 209.717968] CAM_INFO: CAM-SENSOR: cam_sensor_set_alt_id: 731 [Applying Alternative Address] : Sensor Index : 2538021044 [ 209.717979] CAM_INFO: CAM-SENSOR: cam_sensor_set_alt_id: 732 [Applying Alternative Address] : Slave Address : 0xE2 [ 209.717988] CAM_INFO: CAM-SENSOR: cam_sensor_set_alt_id: 733 [Applying Alternative Address] : Slave sensor id: 0x7750 [ 209.717997] CAM_INFO: CAM-SENSOR: cam_sensor_set_alt_id: 734 [Applying Alternative Address] : cci_client sid : 0x71 [ 209.718308] CAM_INFO: CAM-SENSOR: cam_sensor_set_alt_id: 770 [Applying Alternative Address] : Probe : Alternate address updated [ 209.718543] CAM_INFO: CAM-SENSOR: cam_sensor_driver_cmd: 918 Probe success,slot:2,slave_addr:0xe2,sensor_id:0x7750

I also tested the same configuration in J6:

voxl-camera-server -l DEBUG: Attempting to open the hal module DEBUG: SUCCESS: Camera module opened on attempt 0 DEBUG: ----------- Number of cameras: 2 DEBUG: Cam idx: 0, Cam slot: 0, Slave Address: 0x00E2, Sensor Id: 0x7750 DEBUG: Cam idx: 1, Cam slot: 1, Slave Address: 0x0020, Sensor Id: 0x0214 DEBUG: Note: This list comes from the HAL module and may not be indicative DEBUG: of configurations that have full pipelines DEBUG: Number of cameras: 2 dmesg [ 216.996797] CAM_INFO: CAM-ICP: cam_icp_mgr_hw_open: 3879 FW download done successfully [ 217.029303] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor1: Linked as a consumer to regulator.58 [ 217.029395] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor1: Linked as a consumer to regulator.55 [ 217.029484] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor1: Linked as a consumer to regulator.60 [ 217.029516] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor1: Linked as a consumer to regulator.79 [ 217.046069] CAM_INFO: CAM-SENSOR: cam_sensor_driver_cmd: 918 Probe success,slot:1,slave_addr:0x20,sensor_id:0x214 [ 217.046191] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor1: Dropping the link to regulator.79 [ 217.050324] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor1: Dropping the link to regulator.60 [ 217.050407] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor1: Dropping the link to regulator.55 [ 217.050470] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor1: Dropping the link to regulator.58 [ 217.069409] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor0: Linked as a consumer to regulator.58 [ 217.071541] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor0: Linked as a consumer to regulator.55 [ 217.073710] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor0: Linked as a consumer to regulator.60 [ 217.073801] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor0: Linked as a consumer to regulator.79 [ 217.097221] CAM_INFO: CAM-SENSOR: cam_sensor_set_alt_id: 731 [Applying Alternative Address] : Sensor Index : 390152372 [ 217.097229] CAM_INFO: CAM-SENSOR: cam_sensor_set_alt_id: 732 [Applying Alternative Address] : Slave Address : 0xE2 [ 217.097234] CAM_INFO: CAM-SENSOR: cam_sensor_set_alt_id: 733 [Applying Alternative Address] : Slave sensor id: 0x7750 [ 217.097239] CAM_INFO: CAM-SENSOR: cam_sensor_set_alt_id: 734 [Applying Alternative Address] : cci_client sid : 0x71 [ 217.097532] CAM_INFO: CAM-SENSOR: cam_sensor_set_alt_id: 770 [Applying Alternative Address] : Probe : Alternate address updated [ 217.097747] CAM_INFO: CAM-SENSOR: cam_sensor_driver_cmd: 918 Probe success,slot:0,slave_addr:0xe2,sensor_id:0x7750

By the way, i think the SDKs are now not including the imx214 driver for slot 1. If you need it, you can use the following drivers https://storage.googleapis.com/modalai_public/temp/imx214_test_bins/imx214_eis_drivers_20250815.zip for all camera slots. The only functional change of these drivers is that the maximum resolution of the sensor has been changed from 4208 x 3120 to 4196 x 3120 due to some optimizations in gpu-based debayering that require certain image width (long story, but you can still use this driver, just may need to adjust your camera server config so that you are not requesting impossible resolution).

When you do your tests, please make sure you have the correct sensormodule id's for each camera in /usr/lib/camera . The slot number where the camera is physically plugged in has to match the index at the end of the sensormodule file name (before .bin).

Alex

@voxltester As I tried to explain before, don't wait for voxl-mavlink-server to connect to your application. You need to have it initiate the connection to voxl-mavlink-server.

@Eric-Katzfey We are experiencing connection issues between VOXL2 and QGC.

QGroundControl is showing disconnected after inputting the comms link with 192.168.7.2:14550 in UDP. mavlink_to_gcs and mavlink_from_gcs are both reporting data. voxl-px4 and voxl-mavlink-server are running.

voxl-my-ip
192.168.7.2 (usb0)

@PatkirkMartin , I am looking into the command routing via PX4 for the on/off function.. stay tuned..

@LukeNow This worked, thanks.

@JB1 Our voxl-remote-id software with WiFi dongle is intended only for reference and is not meant to be a commercially ready system for our customers who need a remote id solution for their VOXL2 based drone products. Integration, testing, validation, and FAA registration with desired hardware is up to the end user. At a minimum the manufacturer id would need to be changed from ModalAI's id to your companies issued manufacturer id.

@NikosMav Also in QGC no orientation is shown, just the position.

@PatkirkMartin No, as I said a UART mapped to the applications processor would also work. PX4 has a split architecture on VOXL2 with all flight critical operations happening on the DSP and support functions such as Mavlink, logging, and data manager are on the applications processor. But there is flexibility to move functions between them as needed. So, to match the ELRS receiver with the Mavlink module you could run it on the applications processor instead of the DSP. However, you would need an expansion board that exposes such a connector.

@ApoorvThapliyal : the problem (Error message: "set_position_target_local_ned: coordinate frame 0 unsupported") appeared in QGC after updating VOXL SDK from 1.5.0 to 1.6.3.
The error blocked me from being able to use QGC to calibrate or arm the drone. Stopping and restarting voxl-vision-hub solved the problem.
For now I'm using voxl-vision-hub as it comes on SDK 1.6.3 (no changes yet) and have not made changes to the .conf file (yet).

@vmbasvi That appears to be a Microhard modem, not a WiFi dongle.

@Ben-Linne : Thanks for the excellent instructions. Worked perfectly first shot. betaFPV is back to normal and binds to the radio controller ! Best regards, Hector

@jon When running that way make sure you wait a few seconds before restarting. If you ctrl-c out of voxl-px4 and immediately restart voxl-px4 you will see that error condition.

@Eric-Katzfey UAVCAN is working in the latest versions, when enabled. And yes I meant heading error, thats my mistake.
The collision prevention is the major issue I am having.
The heading error is in the latest version, and I also came across it in the issues section and it was shown solved, but I did not test it out still with VIO, I was testing with only Optical flow and it was not working.

@Alex-Kushleyev

Thank you. We will do, but for future reference:

@Alex-Kushleyev said in Burnt board components:

what type of vehicle was the ESC a part of?

Singer Vision FPV 3.5"

@Alex-Kushleyev said in Burnt board components:

what was the behavior of the motor that was not functioning properly?

Not operating a full power with intermittent operation

@Alex-Kushleyev said in Burnt board components:

did the whole ESC stop working or only one channel, which was originally struggling?

Only one channel

@Alex-Kushleyev said in Burnt board components:

did you notice the burned component when the ESC arrived or did it burn after arrival?

We didn't noticed or inspected it.

@Alex-Kushleyev said in Burnt board components:

was the black paint already on the ESC when you received it?

We didn't notice or checked