New FrSkyX protocol

thanks for your quick advice.

I have copied the whole file system that comes with the deviation-devo7e-v4.0.1-4ed4de6 pack (including the protocol folder), except the model folder and the hardware.ini / tx.ini

btw, it is not only the frskyx, but also skyartec that uses the same CC2500 module that gives the same error

I see that you have good result in Ver. 1f0. I also want to check it out, where can I get this version?

Hello,
if you get the Error with all Files ready the there is a Hardware or hardware.ini Error ( =13 and =14 -> the Solder Point at the TMS TCK Pad ?)
The Ver. 4ed does work on Channel number set to 8 and lower. Danger here at higher 8 and Power Cycle on RX !
The Ver. d76 does work on Cannel number set to 12 and lower without Problems on the Bench Test with my 7e. Only a litte Jitter some Times.
Ver. 1f0 was not working right, same Bug 4ed,073

Here is d76, but careful it is not 100% testet .

i think the jitter has nothing to do with the protocol itself.
Its caused by the cc2500 messing up TXs 3.3V voltage.
Therfore Analog inpus and TX voltage readings are unstable

@ Alexandro
Do you have an Oscilloscope on hand?

Guys I feel like we are moving in circles here,
I think It needs an LC filter tuned on that specific frequency(In can be calculated see timing on scope).
The same LC filter type used for filtering esc motors noise on FPV.
blog.oscarliang.net/power-filter-video-transmitter-fpv-camera/
www.dronetrest.com/t/what-in-an-lc-filte...d-do-i-need-one/1381

@Wene001

Yes but it is to bad to get a good Display.
I ordert a LC Filter to get the Voltage jumping lower, may be next week i can answer at the other Post.
The little Jitter is only on the Stick Channel and there it is max on Channel 1 and falling lower to Channel 4. On 3 and 4 it is most Time ok but on 1 it is moving the Digital Servos.
On D8 Protocol it is lower .
I posted long Text to explain all the Things but some times the Jitter is a bit Magic and it goes low and nearly disappear.
On Stick movement it change the Speed from Point to point, it looks like same as watching an airscrew through a Video cam and changing the rpm slowly.
And it get bigger on adding more Mixer to the Channel. This is why i think it is an amplified ( through adding Mixer ) and triggered from the AD converter.
or im total wrong and it comes direct from the Software on the RX.
-> With external Frsky Module and Frsky Hack Module over PPM it does not Jitter
-> With External Flysky it does a little

Im getting grey Hairs on hunting It on Test Bench with my little Knowledge of the Complete Deviation Thing.

may be with the LC filter on the vcc of the TX Modules . The last thing after the LC filter is a shielded Network Cable for all the connections .
Sometimes the Cpu Clock can get into the Data Lines and getting something wrong on the 2nd Cpu. On my Point as an licensed Radio Operator it is a wonder thad 3 Modules with Cpus inside ,on thad wild unshielded Cable Tree over the Motherboard with 2,4 Ghz and thad high Data rate are doing this good job over this long Range . <- my 2 ct.
hard to describe with English words

Alexandro wrote: @cmpang
if module is right connected try to copy the protocol folder from the Test Build into the 7e Drive.

Mixing protocol files/folder with DFU's from different builds is a really bad idea. The code in the protocols knows where things are in the code in the DFU, and tries to use it. I'd consider the transmitter crashing when you tried to load it a good result. Certainly better than a crash at random while in flight or erratic results.

Sorry if I'm nagging about things you know - but I'd rather be a nag than have someone lose a model because we failed to warn them.

d76 works fine.. thx Alex

Still using the "old" 15e8960 version here.. just had a first proper test flight today. Went off perfectly, no issues whatsoever, no RSSI warnings etc.. I didn't go overly far (~300m) since this is on a race copter.
Used a 7e with a Frsky X4R-SB.

Hello,
Done a Flight Test today with Ver.073 ( latest Test Build)
Tx 8s, 100mW, RX X8R with Black Box PCB Antenna.
Location near River, Wet ground, Plane 1,2m Trainer Balsa
Connected Altimeter on Sport

All very nice , no Glitches. RSSI strong no Warning ( Warning set to 45 or less )
Altimeter showed good High , flying Radius 200-250 m, Alt max 50-100 m ( sorry can not look long an the Display )

I think it is ok to fly it on the 8s, but be careful there can be some hidden glitches !

p.s. DSM2 with intigratet Orange RX Gyro had more Problems with Hold ! DSM2 3 Hold - FrskyX 0 , last 5 Days of Testing.

greetings Alex

and

THANKS to HEXFET for his GOOD WORK to bring us the FRSKYX Protocol !

any other Testers ?
7e ?

some results on my 7e:

d76 binds ok. 12 channel seemed work ok. not until today I noticed there is a tiny latency in all channels.. that is some delay (about 0.5 second) upon my stick input.

tested 073, led shows binding ok but no response from all channels

so the frskyX protocol cannot be used on my 7e until there is a fix,

@hexfet been out of the loop for a while. What are the differences between the two 7e builds please?

My D10 brush the test firmware can not boot up, there is no D10 users have tried this X FRSKY mode?

gongye wrote: My D10 brush the test firmware can not boot up, there is no D10 users have tried this X FRSKY mode?

I have 0732b2f on my Devo 10 with no problems. Haven't installed C2500 module yet tho !

@djdazzy
-> 7e

076 was binding with out PA on but worked with 12 Channel set on the Testbench

the next Version was to Big for adding the PA on Code, then hexfet removed some Telemetry to fit it into the 7e

here was some Error with Binding at Channel set to 12 , the last version 4ed was binding with 8 Channel.
After setting to 12 it worked to the Point of RX Power Cycle then it does not work. After switching back to 8 it comes back to Work with out rebinding.

@gongye
copy all files from the Test Buid folder into the Drive of the Devo 10 and over write it .
may be this helps.
My D10 is here to Test it tomorrow, then i can Report ( White Version )



greetings Alex

Hi guys, just wanted to let you know - I tested the 4ed build with my Devo 7e and a new receiver X4R, unfortunately I don't have it connected to anything so I just could check if it bound - all good
I'm really looking forward to this protocol being released in a nightly,

Have a nice day all,
rasqba

Got back to looking at this today. For the 7e would someone please try installing the latest nightly build (full install), then install the latest test build. The issue with the 7e started after merging recent changes in the default branch to frskyx. Perhaps there was a fs change that I didn't notice.

Likely the problem will remain. Haven't spotted any issue inspecting the code, so planning to start a new branch from d76 and add changes one at a time till it breaks.

Thanks all for the test reports.

Hi, I think I can connect my receiver to some FC and check it in cleanflight, just let me know which test build should I flash (4ed or 073?), and after flashing it over newest nightly which files should i copy (all except models, tx.ini and hardware.ini?)

Thanks,
rasqba

OK, I did a clean install of the latest nightly (18/4). Then i installed the latest test build (4ed4de6). The receiver (X8R) did bind but with a flickering green light. No movements on any channel.
Then again i did a clean install of the latest nightly and installed the test build 0732b2f. This time i got a good bind. (no flickering, but stable green light) But i got only movement on channels 6, 7 and 9 (i also had a Sbus to PWM decoder on the sbus for channel 9 to 16).

I still got version d764a03 on my 8s. With that transmitter everything works fine.

Tom

After testing the latest test builds, i again did a clean install of the latest nightly and installed test version d764a03 on my devo 7e again. And now everything works fine again. All 12 channels and no jitter (when using 1 as fixed ID.

Tom

TomPeer wrote: After testing the latest test builds, i again did a clean install of the latest nightly and installed test version d764a03 on my devo 7e again. And now everything works fine again. All 12 channels and no jitter (when using 1 as fixed ID.

Tom

What is the reason to install the latest nightly first? Aren't we surpose to flash the test build's DFU?