Replies

  • Paul, great ongoing work.  Are you still using the restricted non-commercial frsky library?  I would just carry on in your own repo, you're the de facto maintainer now :)

  • Another code commit today to my repo - link in previous post.

    Thanks to tester feedback I have ironed out some flight mode bugs, but also took this opportunity to replace both plane217.lua and coptr217.lua with a single new telemetry script called telempc.lua which combines both AP types. Also offset.lua has been updated to include a new option to select between copter = 1 or plane = 2.

    Please feel free to give it a go. I will update the Clooney82 wiki once I merge this into that repo.

    Thanks, Paul

  • Another code commit today to my repo: https://github.com/athertop/MavLink_FrSkySPort

    I have removed the reliance on the flight mode/global variables table in OpenTx, and added some other code fixes and minor enhancements. Only been tested on the simulator so far and on my Taranis Plus on the bench. If you guys plan to test this, please make sure you get the latest version from the above repo. If all is OK, then I will merge into the Clooney82 repo. Thanks, Paul

    athertop/MavLink_FrSkySPort
    APM MavLink to FrSky S.Port converter. Contribute to athertop/MavLink_FrSkySPort development by creating an account on GitHub.
  • I'll do my best to test it out next weekend .... 


  • Cheers Garry - my spelling is atrocious TBH!  (I'm amazed I spelled "atrocious" :-)


    Garry Kraemer said:

    Paul,

    I found a spelling mistake in the first sentence on the wiki:
    https://github.com/Clooney82/MavLink_FrSkySPort/wiki
    'Ardupolot' should be 'Ardupilot'.

    Keep up the good work!
    Thanks for making the improvements!

    Garry

  • Paul,

    I found a spelling mistake in the first sentence on the wiki:
    https://github.com/Clooney82/MavLink_FrSkySPort/wiki
    'Ardupolot' should be 'Ardupilot'.

    Keep up the good work!
    Thanks for making the improvements!

    Garry

    Clooney82/MavLink_FrSkySPort
    This MavLink_FrSkySPort repository is discontinued! The development is moved to athertop/MavLink_FrSkySPort. Please do not use this repo, and follow…


  • Paul Atherton said:

    I am currently looking for anyone interested in testing a new version of the Mavlink->FrSkySPort software.

    In a nutshell, I have made some improvements to the telemetry screen and added some new features.

    The new pre-release is available from my fork here: https://github.com/athertop/MavLink_FrSkySPort

    Using this will require a few things to happen:

    • The Teensy sketch will need to be re-flashed from my repo above - to do this you will first need to install the updated FrSkySportTelemetry library provided in my libraries folder, as there are some changes in this library

    • Note that before writing the sketch, there are a number of new options which can be enabled in MavLink_FrSkySPort.ino These options are discussed in the updated Wiki here https://github.com/Clooney82/MavLink_FrSkySPort/wiki 

    As for the new features, I have:

    • refreshed the telemetry screen layout a bit - mostly to allow the inclusion of Airspeed and units for both speed and altitude (plus climb rate) values displayed. The screen now looks like this:
      main_screen1.jpg
    • Switchable units for both speed and Alt (/alt climb rate) can be configured in the offset.lua mixer script

      In offset.lua, the following values can now be configured:
      • SpeedUnits - 1=m/s, 2=kph, 3=mph.
      • AltUnits - 1=m (m/s for climb rate), 2=feet (or feet/s for climb rate)
    • Additionally, there is now a second telemetry screen called txtmsg.lua which should be added as Screen2 through the telemetry sensors page. This provides mavlink status messages and looks like this*:

      msg_screen.jpg?width=660*Note - occasionally data corruption does sometimes corrupt the text of messages displayed here - this is still a bit experimental.

    Other recent features recently added (in release v1.7) - already in the main Clooney82 repo include:

    • New features allowing the teensy to be used in a ground station for Ultimate LRS - either as part of a repeater/relay station, or located in the ULRS Tx module itself, directly connected to the Taranis.

      This is possible as ULRS downlinks the Telemetry port data to the ULRS Tx (at 19200baud), so we can connect the Teensy to the serial port of the Tx on the ground side to pick up a Mavlink connection, and output the teensy SPort sensor data directly into the Taranis using the SPort Polling feature now included in the project (by uncommenting #define POLLING_ENABLED in MavLink_FrSkySPort.ino).

      In a repeater/relay configuration, the Teensy output would connect into a FrSky X4R Rx SPort as normal, where the SPort data would get to the Taranis over the regular XJT radio connection (no SPort polling required for this).

      As no SPort connectivity exists on the airside in this configuration, it is not possible to use an FLVSS  LiPo sensor in the model to monitor individual cell voltages, so in this case its preferable to uncomment the line #define USE_FLVSS_FAKE_SENSOR_DATA in MavLink_FrSkySPort.ino before writing the sketch to the teensy.

    • The second feature aimed at ULRS users is the ability to configure the teensy code to provide RSSI via mavlink (by uncommenting #define USE_MAV_RSSI in MavLink_FrSkySPort.ino) - this is the only way to get RSSI when using Ultimate LRS (at the moment) (requires the ULRS Rx to have its RSSI output connected to the SB port on the Pixhawk with suitable LP filter, and RSSI to be configured in AP). When this is enabled it provides RSSI data on the A4 sensor (replacing A4 original purpose of displaying Pitch angle value). It also sets A3 to 0 (instead of Roll angle). Using this configuration the RSSI value is automatically picked up by the telemetry screen which will replace rssi: on the top bar (which is for FrSky XJT RSSI) with rx-rssi: as you see in the image above to reflect Rx supplied RSSI over Mavlink.

    Yes, I jumped the gun and updated the wiki, but the software has performed well so far in tests so will be released soon - just hoping a few others might be willing to do a quick test before I merge the code changes to the Clooney82 repo)

    Cheers, Paul

  • I am currently looking for anyone interested in testing a new version of the Mavlink->FrSkySPort software.

    In a nutshell, I have made some improvements to the telemetry screen and added some new features.

    The new pre-release is available from my fork here: https://github.com/athertop/MavLink_FrSkySPort

    Using this will require a few things to happen:

    • The Teensy sketch will need to be re-flashed from my repo above - to do this you will first need to install the updated FrSkySportTelemetry library provided in my libraries folder, as there are some changes in this library

    • Note that before writing the sketch, there are a number of new options which can be enabled in MavLink_FrSkySPort.ino These options are discussed in the updated Wiki here https://github.com/Clooney82/MavLink_FrSkySPort/wiki 

    As for the new features, I have:

    • refreshed the telemetry screen layout a bit - mostly to allow the inclusion of Airspeed and units for both speed and altitude (plus climb rate) values displayed. The screen now looks like this:
      main_screen1.jpg
    • Switchable units for both speed and Alt (/alt climb rate) can be configured in the offset.lua mixer script

      In offset.lua, the following values can now be configured:
      • SpeedUnits - 1=m/s, 2=kph, 3=mph.
      • AltUnits - 1=m (m/s for climb rate), 2=feet (or feet/s for climb rate)
    • Additionally, there is now a second telemetry screen called txtmsg.lua which should be added as Screen2 through the telemetry sensors page. This provides mavlink status messages and looks like this*:

      msg_screen.jpg?width=660*Note - occasionally data corruption does sometimes corrupt the text of messages displayed here - this is still a bit experimental.

    Other recent features recently added (in release v1.7) - already in the main Clooney82 repo include:

    • New features allowing the teensy to be used in a ground station for Ultimate LRS - either as part of a repeater/relay station, or located in the ULRS Tx module itself, directly connected to the Taranis.

      This is possible as ULRS downlinks the Telemetry port data to the ULRS Tx (at 19200baud), so we can connect the Teensy to the serial port of the Tx on the ground side to pick up a Mavlink connection, and output the teensy SPort sensor data directly into the Taranis using the SPort Polling feature now included in the project (by uncommenting #define POLLING_ENABLED in MavLink_FrSkySPort.ino).

      In a repeater/relay configuration, the Teensy output would connect into a FrSky X4R Rx SPort as normal, where the SPort data would get to the Taranis over the regular XJT radio connection (no SPort polling required for this).

      As no SPort connectivity exists on the airside in this configuration, it is not possible to use an FLVSS  LiPo sensor in the model to monitor individual cell voltages, so in this case its preferable to uncomment the line #define USE_FLVSS_FAKE_SENSOR_DATA in MavLink_FrSkySPort.ino before writing the sketch to the teensy.

    • The second feature aimed at ULRS users is the ability to configure the teensy code to provide RSSI via mavlink (by uncommenting #define USE_MAV_RSSI in MavLink_FrSkySPort.ino) - this is the only way to get RSSI when using Ultimate LRS (at the moment) (requires the ULRS Rx to have its RSSI output connected to the SB port on the Pixhawk with suitable LP filter, and RSSI to be configured in AP). When this is enabled it provides RSSI data on the A4 sensor (replacing A4 original purpose of displaying Pitch angle value). It also sets A3 to 0 (instead of Roll angle). Using this configuration the RSSI value is automatically picked up by the telemetry screen which will replace rssi: on the top bar (which is for FrSky XJT RSSI) with rx-rssi: as you see in the image above to reflect Rx supplied RSSI over Mavlink.

    Yes, I jumped the gun and updated the wiki, but the software has performed well so far in tests so will be released soon - just hoping a few others might be willing to do a quick test before I merge the code changes to the Clooney82 repo)

    Cheers, Paul

    athertop/MavLink_FrSkySPort
    APM MavLink to FrSky S.Port converter. Contribute to athertop/MavLink_FrSkySPort development by creating an account on GitHub.
  • OK ... I was messing about and got the Cmin working, it was something i am SURE i tried earlier and hadn't worked, maybe it was due to the old Telem script and not the new one 

    I think your wiki should say

    Method - Cell 
    Cells Sensor - Cels
    Source 1 - Lowest

    This is a screenshot of my settings - https://goo.gl/photos/zmk2kzYzuwaSSt9V6

    Ewan Fletcher said:

    Paul,

    Thanks alot for the reply,

    I've taken a quick screenshot of my TX v Mission planner.

    Everything was updated today to the latest version of the repo, the fake sensor was uncommented and the max cells was set to 4

    Looking at the link above i thought A2 was the HDOP reading x 10?

    I'm running 2.1.9 and 3.3.3, i was using the previous Telem1.lua but have swapped to copter2.1.7 and its made no difference

    I've stuck a couple of screnshoots up to show what i have on A2 and how i have A2, and the options i get for Cmin - https://goo.gl/photos/1yiqewBpzGFMqHnW9

    Paul Atherton said:

    Ewan,

    OK, I have checked some things. The telemetry value T1 contains the HDOP info. The Wiki describes correctly how to configure this here: https://github.com/Clooney82/MavLink_FrSkySPort/wiki/1.2.-FrSky-Tar...

    I also checked my working configuration for the calculated 'Cmin' sensor, and this matches the Cmin details exactly as per the same wiki page.

    Just for info, I am running OpenTx 2.1.8 and the latest copter217.lua script available from the Clooney82 repo on github. At the copter end, I am running Ardupilot 3.3.3 on Pixhawk FC. I have also tested this successfully on the 3.4-rc5 firmware, and also on latest Plane f/w.

  • Paul,

    Thanks alot for the reply,

    I've taken a quick screenshot of my TX v Mission planner.

    Everything was updated today to the latest version of the repo, the fake sensor was uncommented and the max cells was set to 4

    Looking at the link above i thought A2 was the HDOP reading x 10?

    I'm running 2.1.9 and 3.3.3, i was using the previous Telem1.lua but have swapped to copter2.1.7 and its made no difference

    I've stuck a couple of screnshoots up to show what i have on A2 and how i have A2, and the options i get for Cmin - https://goo.gl/photos/1yiqewBpzGFMqHnW9

    Paul Atherton said:

    Ewan,

    OK, I have checked some things. The telemetry value T1 contains the HDOP info. The Wiki describes correctly how to configure this here: https://github.com/Clooney82/MavLink_FrSkySPort/wiki/1.2.-FrSky-Tar...

    I also checked my working configuration for the calculated 'Cmin' sensor, and this matches the Cmin details exactly as per the same wiki page.

    Just for info, I am running OpenTx 2.1.8 and the latest copter217.lua script available from the Clooney82 repo on github. At the copter end, I am running Ardupilot 3.3.3 on Pixhawk FC. I have also tested this successfully on the 3.4-rc5 firmware, and also on latest Plane f/w.

This reply was deleted.

Activity