Yeah, flipped that as well. sure it’s something I’m overlooking just can’t find it.
Which DEV. build you have used? The one linked above?
It’s not in the master yet.
For my current test’s I am using the latest posted above, arduplane_matekf405wing_v0.8.zip
I start to run out of ideas. And I thought I was one of the guys who tried everything possible wrong first…
But the “classical” (barely 2 month old) CRSF telemetry seems to work properly?
Nope, and that’s the thing… I am not getting any… not even the original CRSF telem I was getting before switching to this build. O well, I keep poking at it.
Found the problem… had RCIn enable on two ports… guess it didn’t like that.
1 Like
Thanks for this information - was very helpfull for me. I will compile the Durandal by myself - learning by doing 
.
Maybe we could ask Eric for some ideas regarding the FrSky sensors. He has a FrSky2Mavlink converter. Something possible with a Teensy?
During the test in my home I have seen a issue wiht the CRSF telemetrie. From time to time (within 2-5min) I got the alarm “Telemetrie lost” and the red borders on the widget. It occours only a part of a second and not generate a error on the flight controler. But it ouccours many times not only once.
What is your use case? You have frsky sensors on board the aircraft and would like to sent them down the crossfire link? GAS suite, jet etc?
Many reported telemetry link jumping between rf mode 0, 1 and 2 when TX and RX are very close on the bench, I had the same issue myself, I move the RX a couple meters away and it stops.
OK, thanks, my suspect is that older TBS firmware versions do not update the link statistics frame that I parse in ardupilot to derive rf_mode, I tested version 4.0.3 and 4.0.5 and can’t reproduce it.
this can became a real issue since I need rf_mode to toggle between high rate telemetry and low rate telemetry.
Which firmware are you using?
We could have a combination of timing and the link statistics potentially
Like your old code did, yes I though about that but what would be the threshhold? unfortunately rates vary a lot and a 50Hz polling rate coul dbe mode 1 on a full speed uart or mode 2 in a slower one 
Hi Alex,
tested with TBS fw 2.42. I will perform the test with a newer fw too.
[edit: typo]
1 Like
My use case will be the on board sensors from the aircraft - especially the gas suite (Turbine), the rpm sensors and the Frsky Lipo sensors. Most importend the gas suite for the turbine start and use.
I will take on the “jump behavior” at the test flight.
ok, so you have many frsky sensors chained on the bus…not an easy task to “wrap” them on crossfire 
most importend would be only the gas suite. The others are like “nice to have”.
@yaapu updated TBS fullsize TX to fw 4.05 and get:
APM: ArduPlane V4.1.0dev (0b3bc7b4)
APM: ChibiOS: 331fe75d
APM: omnibusf4pro 00300029 3137470A 30363730
APM: RCOut: PWM:1-8
Received 1091 parameters (ftp)
Flight battery 100 percent
Mode MANUAL
APM: CRSF: rf mode 2, rate 156
APM: CRSF: rf mode 2, rate 155
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 153
APM: CRSF: rf mode 2, rate 156
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 155
APM: CRSF: rf mode 2, rate 156
APM: CRSF: rf mode 2, rate 157
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 156
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 156
APM: CRSF: rf mode 2, rate 157
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 156
APM: CRSF: rf mode change 2->1
APM: CRSF: rf mode change 1->2
APM: CRSF: rf mode 2, rate 155
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 154
on the xLite display I see RFMD0
and optionally (depending of the build) the debug messages:
Let it run a little bit longer I get the rf mode switching messages too. Maybe I need more distance. Here is the mavproxy output:
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 159
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 155
APM: CRSF: rf mode 2, rate 156
APM: CRSF: rf mode 2, rate 159
APM: CRSF: rf mode change 2->1
APM: CRSF: rf mode change 1->2
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode change 2->1
APM: CRSF: rf mode change 1->2
APM: CRSF: rf mode change 2->1
APM: CRSF: rf mode change 1->2
APM: CRSF: rf mode change 2->1
APM: CRSF: rf mode change 1->2
APM: CRSF: rf mode change 2->1
APM: CRSF: rf mode change 1->2
APM: CRSF: rf mode 2, rate 142
APM: CRSF: rf mode 2, rate 155
APM: CRSF: rf mode 2, rate 156
APM: CRSF: rf mode 2, rate 157
APM: CRSF: rf mode 2, rate 158
APM: CRSF: rf mode 2, rate 158
shown on the xLite display too:
@jelkins Glad you found it.
Ok, so the newer firmware solves the ardupilot rf mode detection issue!
Now please rediscover your sensors, check if the RFMD sensor is there and check it’s value via a telemetry screen, I do not manipulate it so should be there irregardless of my patches
A rediscover of the sensors made the RFMD indicator alive!! 
I found an easy way to create “virtual” distance. Tests with a SMA-terminator in place of the TX aerial made it possible to achieve low RSSI or LQ settings in a repeatable way. Maybe helpful for others too.
with aerial:
with SMA dummy load:
1 Like
Thanks a lot Sasha,I’ve been assured by TBS that same version firmware on Full TX and Micro TX behaves the same, this allowes me to test only those versions I can flash on my Micro TX V2, i.e. >= 3.72
1 Like
Hi Alex,
It’s a pleasure for me to be able help you. You are the one who makes the magic happens. As satisfied user of your FrskyTelemetryScript that’s the least I can do. And I learn a lot . . .
I have a TBS Nano TX coming. More things to test. 
According to Trappy it should be the same hardware as the MIcro TX V2. The new TX may help to reduce the amount of uncertainties.
Good to know you are able switch back to a stable firmware like FW3.72. I am not the only one keep running on older FW’s.
Why I switch back to CRSF FW2.42? It worked very well for me in the past. If it’s not broken don’t fix it was the idea behind that.
1 Like