TBS Crossfire

[badger1]

I'm toying with getting a TBS Crossfire Micro Tx module & Nano Rx - does anyone have any experience with them?

I'm not looking to go super-long distances, but I seem to get RSSI warning from my 2.4GHz Taranis rather a lot & it would just be nice to have a bit more control security.  I did look at FrSky R9 but it doesn't seem to have a happy history & I'm generally a bit wary of FrSky's direction at the moment what with ACCST 2.0, etc.

[atomiclama]

I'd stay away from R9.

If you are lucky it works if not you are stuffed, many people have good experience with it others not, I'm one of the latter.
Initial walk tests looked promising but as soon as I started flying it RSSI was way down and  the link started dropping out. The R9MM antenna connector is so flimsy, once it's taken apart it won't go back.

[Sweet Pickle]

Yeah agreed. I dabbled with R9 twice, what do they say about fool me once? Failsafe after failsafe and nothing like the signal strength I was expecting. I really didn't want to go down the TBS route in much the same way I didn't want to go down the DJI digital route but you know what? You just can't argue with the quality of either. You might not like it or what they stand for but... it works and it works well. Take my money you ba$tards.

[ched]

Yeah agreed. I dabbled with R9 twice, what do they say about fool me once? Failsafe after failsafe and nothing like the signal strength I was expecting. I really didn't want to go down the TBS route in much the same way I didn't want to go down the DJI digital route but you know what? You just can't argue with the quality of either. You might not like it or what they stand for but... it works and it works well. Take my money you ba$tards.

I assume you have looked at the dji controller? As it would integrate with the dji camera and goggles very well and seems to have good range?

[Sweet Pickle]

I did look at it, range isn't really a consideration. I have a Mavic thats only flown 3 times so know where I'm best suited. It's all about 5" and below for me. I also prefer the portability of a small transmitter. My current is the X-Lite with slimmed down TBS Micro but I'm considering a Tango 2. If it wasn't for the fact I have loads of micros that have Frsky I'd have switched as soon as it hit market. Finding it hard to justify now.... knowing me I will be swayed though, I always am. I'm a fool for spending for the sake of it.

[badger1]

thanks chaps

there may be a post from me in the "what did the postman bring you" thread tomorrow. 

[badger1]

I'm just pondering whether it's worth trying to run my shiny new Crossfire Micro Tx/Nano Rx on an F4 board with limited UARTs ...

I see that it's possible to output SBUS either inverted or non-inverted from the Nano, so I could potentially use the SBUS pad on my F4 FC for this.  I haven't seen anything about being able to run SmartPort though, so I assume this isn't supported?

I also see that it's possible to run Smart Audio directly off the Nano, which would free up a pad on the FC.  anyone doing this?  is this Smart Audio generically - ie including Tramp - or specifically TBS Smart Audio?

[ched]

I thought crossfire was a tx/rx type rx. i.e. non inverted but needs both pins on a uart. So you get rx and telemetry signals. I think ch1 is tx ch2 is rx but check that.

[badger1]

I thought crossfire was a tx/rx type rx. i.e. non inverted but needs both pins on a uart. So you get rx and telemetry signals. I think ch1 is tx ch2 is rx but check that.

for Crossfire the protocol - CRSF - you are correct.

for Crossfire the system, the Rx can output either CRSF or SBUS (or PPM/PWM, Smart Audio, etc.) on whatever pins you tell it.  I guess it doesn't do much other than, well, receiving () & passes-thru whatever it receives to the FC (or servo, VTx, etc.).

Crossfire Manual

[ched]

Looks like smartaudio is just for direct connection to tbs Unify vtx control.
So if you want telemetry you need ch1 & ch2 to tx and rx uart pad.

[badger1]

thanks ched, looks that way re. the Telemetry.

I suspect you could run any VTx that supports (specifically) Smart Audio off it though, but not Tramp.

[badger1]

so, maiden flights with Crossfire today & the results are  ... mixed.

from a control point of view it seems like an excellent system.  even before the end of the first flight I was flying behind  trees that I wouldn't have dared go behind on FrSky for fear of fail-safe'ing.

however, it doesn't seem like it takes much of a drop in LQ to cause it to drop from 150Hz to 50Hz mode & it takes a while once LQ is restored for it to go back to 150Hz.

now, I don't really care about 150Hz vs. 50Hz mode, that's not why I bought Crossfire, but when it drops to 50Hz mode, the Telemetry update rate drops off a cliff - to once every five seconds - 0.2 Hz!

here is some graphing of Taranis logging:



the red line is link quality (LQ), which is somewhere up near 100% most of the time, but spikes down to 82% (presumably behind tress) a couple of times.

the green line is 150/50Hz mode (RFMD) - when it's at 2 it's in 150Hz mode & when it's at 1, it's in 50Hz mode.  even a very slight drop in LQ causes it to drop into 50Hz mode, & then it takes a long time to come back to 150Hz.

the pink line is GPS speed (as an example of a telemetry reading that the quad is sending back).  when RFMD is at 2, the pink line is a real-time monitor - at elast as fast as the 0.5s rate I log in the Taranis.  however when RFMD drops to 1, the pink line becomes a succession of 5s-long horizontal lines.

the problem is that the low Telemetry refresh rate in 50Hz mode makes for lousy GPS traces:



the nice smooth lines are @150Hz & the jagged lines are @50Hz. 

also, it sends telemetry data back in metric units (m/km/h) vs. imperial in FrSky (ft/mph) which has the interesting side-effect of making my traces appear to be underground. 

[ched]

Silly question but do all crossfire systems have switchable power levels?
Were you on low power?
I thought that the power could be set to be adaptive upto something like 1W?
Seems the micro version is 10mW or 100mW which were you on?

[badger1]

hmm, fair point, thanks ched.  I just looked & I was on 25mW with dynamic power disabled, so I'll have a tinker with that for next time.  I might not have flown behind those trees if I'd realized!

I still think the telemetry refresh rate of once every 5s is incredibly slow though, if the control rate is once every 1/50th of a second (even at the reduced rate).

the Micro Tx will go up to 250mW.  the bigger ones, with the XT30 socket so you can plug in external power, go to the higher transmit powers.

[ched]

hmm, fair point, thanks ched.  I just looked & I was on 25mW with dynamic power disabled, so I'll have a tinker with that for next time.  I might not have flown behind those trees if I'd realized!

I still think the telemetry refresh rate of once every 5s is incredibly slow though, if the control rate is once every 1/50th of a second (even at the reduced rate).

the Micro Tx will go up to 250mW.  the bigger ones, with the XT30 socket so you can plug in external power, go to the higher transmit powers.

5Sec does seem slow maybe it's a safety thing as I think we would always want control with it being acceptable to have mega slow telemetry. If the telemetry say contained gps 5 sec is probably enough to find a downd quad and again 5 sec for batt volt is fine in an iffy situation.
It will be interesting to see your results with a bit more power   

[Six6siX]

Can I ask how big are those fields just wondering what sort of range you were getting for the telemetry to drop it's refresh rate. 

[badger1]

Can I ask how big are those fields just wondering what sort of range you were getting for the telemetry to drop it's refresh rate. 

from take-off to the furthest point is about 500m.  the model was probably quite end-on to the Tx antenna there too.

to be fair, after a bit more looking, telemetry for signal strength/quality/etc. - ie from the Rx itself - doesn't appear to back off in refresh rate, only the data the Rx is learning from the FC.

I will have a play with the power levels next time I'm out, maybe I was expecting too much from it @25mW.

[badger1]

5Sec does seem slow maybe it's a safety thing as I think we would always want control with it being acceptable to have mega slow telemetry.

I presume that control (Tx->Rx) & telemetry (Rx->Tx) are using different frequencies, so I wouldn't really expect the data rate of one to affect the data rate of the other.  if they were time-sharing the same frequency, then yes, it would absolutely make sense to back off the telemetry a lot to give control priority.

If the telemetry say contained gps 5 sec is probably enough to find a downd quad and again 5 sec for batt volt is fine in an iffy situation.

5s of flight could easily be 100m+

It will be interesting to see your results with a bit more power   

I may have to go full Clarkson - "MORE POWER"   

[ched]

I presume that control (Tx->Rx) & telemetry (Rx->Tx) are using different frequencies, so I wouldn't really expect the data rate of one to affect the data rate of the other.  if they were time-sharing the same frequency, then yes, it would absolutely make sense to back off the telemetry a lot to give control priority.

5s of flight could easily be 100m+

I may have to go full Clarkson - "MORE POWER"   

I'm not sure if they share same freq. I had assumed they do but I don't know that for certain. It certainly complicates the freq hopping if you are using 2 freq each time.
I might try and find that out just out of interest.

More power always good as long as you have control 😁😁😁😁

[atomiclama]

I think you can fix the TX into 150Hz mode if you want but not sure you can do anything about the telem.

Just bump it up to full power and fixed 150Hz and see how you go. 

If you want nice smooth GPS traces then you can get the GPS data from the FC black box logs.

As to the freq used, they use Frequency hopping. So there is a range of frequencies and telem is slotted in between the RC update from the TX. Telem is low priority so gets ditched quickly.