FQ777-124 Pocket Drone

Devo 8S with discreet NRF24L01 test #1

goebish, thanks for working on this!

I tested it with an 8S with a discreet NRF24L01+PA+LNA with mixed results. The user interface is messed up, if you use the touch screen to bring up the model menu it goes right to the mixer page with no icons to go to other pages. I found that if I use the transmitter menu icon it brings up a menu that will allow me to select the model menu or transmitter menu. It looks more like a Devo 10 interface than an 8S interface. I was using your DFU and file system on a transmitter that I wiped clean before I loaded the test build. I used the Walkera DfuSE_V2.0 to load the DFU. I also tried using the file system from the latest nightly along with your DFU and got the same results.

The good news, I did get it to fly with the new protocol using the advanced GUI. It flew great in basic flight, I don't know if it has channels for flight modes so I didn't test them yet. When I switched to the standard GUI then the channel numbers got changed around or reversed, I couldn't keep it in the air long enough to figure out if it was a channel order or reversed channel problem. But this may be an existing bug in the standard release. Switching GUIs has never worked very well. Unfortunately I don't think you can re-order the channels in the standard GUI. I've always used Notepad++ to do it. Also in the Standard GUI I didn't get the touch screen icons when I selected the model menu, instead I got a series of equivalent gray function boxes down the left side.

I'll try the new protocol in a Devo 10 with a 13 pin 4-in-1 next.

Edit: I cleared the transmitter and loaded V5.0.0 DFU & file system and the menus worked fine. I then loaded your DFU and the menus are messed up again and the background is black instead of the blue sky. Then I cleared the transmitter again and loaded nightly deviation-devo8-v5.0.0-b289612 from 08-05-16 DFU and File system and it also has messed up GUI without your DFU. I verified on boot up that it did display version "Devo8-v5.0.0-b289612" on power up. Then I tried the DFU & file system from deviation-devo8-v5.0.0-dd199b1 from 07-13-16, it also had a messed up GUI.

Devo 10 with 4-in-1 test #1

goebish

I used your DFU with the file system from the 07-13-16 nightly on a Devo 10 with a 13 pin 4-in-1 board and the FQ777 worked great and all menu items seemed normal (limited test). Then my TX battery died so I didn't get a chance to load your file system and test it with that yet. I also haven't tried anything but basic flight because I don't recall if this quad has channels for different flight modes.

Edit: The FQ777-124 is supposed to have flip mode, headless mode and return to home mode.I tried channels 5-10 and channel 6 was flip mode when set to +100. Channel 9 was headless mode when set to +100. I think channel 10 may be return to home, It looks like just reversing the channel 10 value either way sets it to return to home mode, and giving any Ail or Ele stick input cancels it. I need more room & daylight to test it further. The other channels seemed to do nothing as far as I could tell.

Attached is a preliminary FQ777-124 Devo 10 Model file. It seems to work fine but could probably use more testing, especially the return to home mode.

I placed the following comments in the model.ini file:
; ==CHANNELS==
; ch 1 - Ail
; ch 2 - Ele
; ch 3 - Thr
; ch 4 - Rud
; ch 6 - Flip mode = +100
; ch 9 - Headless mode = +100
; ch 10 - Return to Home - reverse value to enable - use Ail or Ele stick to cancel
;
; ==SWITCH ASSIGNMENTS==
; FMod - Throttle Hold
; Aile D/R - Ail & Ele D/R, up = high rates
; MIX - Flip mode, Mix1 & Mix2 = enable
; Gear - Headless mode, down = enable
; ELEV D/R - Rud D/R, up = high rate
; RUDD D/R - Return to home, flip either way to enable, Ail or Ele stick to cancel

goebish wrote: Those test builds are based on the latest nightly, it may have bugs, but if the FQ777 protocol is working fine then that's Ok
"Flight mode" is set to highest rate, setup dual rate if you need to tame it.
Channel 6 = flip.
Channel 9 = headless
Channel 10= RTH

It looks like the 8S nightlies have been bad for at least 2 releases. But I did get the FQ777 protocol to work on the basic 4 channels, the GUI was too messed up to spend time doing a full model file setup.

The Devo 10 version seems to work great! This little quad is like a CX10 on Nitro and a good TX sure makes it a lot more fun to fly.

Thanks a lot to all who worked on this.

Goebish,

I see that you've used my implementation. Be carefull with rx_tx_addr[0] and [1] are not initialized in your code. They are initialized at powerup for multi. They should be random so it works but...
I've looked at this file: github.com/goebish/deviation/blob/protoc...col/fq777_nrf24l01.c

- Pascal

Yes it works with any random number. I've tried myself a FQ777-124 from a friend. Changing the TXID does not affect the frequencies.
- Pascal

By the way, I think you can also implement quickly/easily the ASSAN protocol. It works fine with any TXID since I've cracked the ID/frequency relationship. Some people on the forum are eager to try it out.
- Pascal

Hello all, i'm new to this forum and first of all i want to thank you for your job and the incredible ability to make our favorite drone to fly with devo radio!
I have this drone and devo 7e, i have tryed to test the last goebish build but i have a problem with the nrf module, i have installed it on my radio and work perfect with other drone like cx-10, h8 mini ect..
If nrf is enabled in hardware.ini (i have it in port no. 13) and selecting fq777 protocol the radio crashes into reboot looping, if i disable nrf module the model with fq777 work fine.
I don't understand nothing of programming, but hope my experience can help you!

sorry for my bad english

goebish, i love you!!!
Now works perfect!! Maybe just a little latency but is already quite good result!!!
Thanks soo much for your great job!!!

Ok, I cleared the Devo 10 and loaded deviation-devo10-v5.0.0-4f78a00.dfu and it's file system. All functions worked great including RTH, Flip, Headless and there seemed to be no lag in binding or flying. I tested changing the fixed id. I still needed to bind it every time I powered up the quad and the tx went into auto-bind when powering it up with this model active. I don't know if this protocol should do this or not. Some protocols with fixed id only need to bind once.I also tried a fixed id of "None" and it had the same behavior.