FY326 Q7 "red board"

I just switched about 10 times (in the garage because dark outside) back and fort between Dyntrim on and off. The rudder seems / looks a tiny bit faster or perhaps it's just my impression or imagination because I want it to be faster;) . But really really noticeably greater with dyntrim on, no I can't say that.
Aileron / Elevator seems / feels a little more responsive but like the above surely not in way that I can say look what a difference.
What I really know is (on mine) the rudder is faster to the left compared to the right but I have the same behaviour with the stock tx so no problem.
But if Dyntrim=on can do no arm for something else, I would certainly like to have it in place because it is only my 'test' input only for the moment.

I don't mind leaving it in place for now. Best would be for someone who both has an FY326 and does their own builds to play around with the trim range and see what works.

I've changed the test build to use a random txid. If you could test that with a few different fixed id values I think we'll be at a stopping point. Hopefully someone will be able to test the headless and RTH functions.

hexfet wrote: I've changed the test build to use a random txid. If you could test that with a few different fixed id values I think we'll be at a stopping point. Hopefully someone will be able to test the headless and RTH functions.

Really cool... not much happened for a few days, then I go on vacation, and by the time I'm back we have a working FY326 red board protocol

I updated my firmware and used the model provided by Alias_Hendrik. In the latest version the Throttle doesn't need to be reversed anymore.

Everything works: binding, flying in normal and expert mode, flips. My model doesn't seem to have headless mode nor rth, so can't test those. It's truly a pleasure flying with a real radio and not the crappy toy remote it comes with. I can't thank hexfet and everyone else enough...

I'm not sure I understand what you mean by " If you could test that with a few different fixed id values", sorry... simply trying to bind using different fixed IDs?

I tested with different id's and it binds no problem. Fixed id= "none" works too.
Already thanks for all the work and also a big thanks to everyone involded. Had just another couple test flights and flies great.

Have to check my ini for "left overs" from testing and after that I will post it so others who want to use it can download it.

I made a video from one flight . (I'm not used to make flight video's). If you turn the volume up, you can hear the "behaviour" from the motors during some circumstances

Edit; Below is the zip file for the FY-326 Red Board. Included is also an Ini (without return to home for now) for the Headless FY-326. The Headless mode is untested and needs... testers.
And because Durete is also involved (thanks) perhaps he can check out the "Headless" ch9 ini part. I copy and pasted it from another ini. (Found in 2 or 3 other quad ini's the same code for ch9 so i believe it should work).

Nice Alias

Congratulations for all people involved to develop this protocol.
Good job guys

Does that mean I can pull all the wires out of my transmitters now?

Not that they'll be used very often from now on! (If I ever get my act together with the 7e, that is:-)

Alias_Hendrik wrote: Edit; Below is the zip file for the FY-326 Red Board. Included is also an Ini (without return to home for now) for the Headless FY-326. The Headless mode is untested and needs... testers.
And because Durete is also involved (thanks) perhaps he can check out the "Headless" ch9 ini part. I copy and pasted it from another ini. (Found in 2 or 3 other quad ini's the same code for ch9 so i believe it should work).

File Attachment:

File Name: FY326RedBo...vo7E.zip
File Size:2 KB

The model file seems to be configured right to me(right yaw trim enable/disable headless mode)
Now we need someone testing this part of the code in real life, but I'm pretty confident in Hexfet's work

One thing that hasn't been mentioned hitherto is gyro calibration. It's not something us FY326 users think about very often:-) It is accomplished by moving both sticks to the lower LH limit of their travel, and causes fast flashing of the quad's LED's and a transmitter beep.

Is that something that would be beneficial to have a capture of before I start pulling wires out?

S.Giles wrote: One thing that hasn't been mentioned hitherto is gyro calibration. It's not something us FY326 users think about very often:-) It is accomplished by moving both sticks to the lower LH limit of their travel, and causes fast flashing of the quad's LED's and a transmitter beep.

Is that something that would be beneficial to have a capture of before I start pulling wires out?

Typically the receiver just detects that position of the sticks and starts the calibration - nothing special is sent from the tx. Should work the same with deviation if the mixer is at 100 scale, but would be good to test.

That's likely accelerometer calibration, and the quad should be level for that. Typically gyro calibration happens on every power-up which is why the quad should be motionless until bound.

I did a try a moment ago and won't work. In another try with Dyntrim off same result.
Now that Robca is back, perhaps he can also give it a try.

I tried calibrating with my Devo 7e, after checking that on the channel monitor I was actually sending 100% of the signal... doesn't work for me either (i.e. even if I move both sticks to the lower left, nothing happens)

It works with the crappy FY326 Tx, so it's possible that the Tx sends a different packet when it detects a calibration event.

I captured the calibration twice in the "FY326 calibration.zip" file copied to my share 1drv.ms/1FU1Tzs (i.e. I started sampling with the radio/quad already bound, moved both sticks lower left, waited for the confirmation, returned to center, then repeated once more)

Please let me know how else I can help...

BTW: where's the Deviation Tx utility that you folks are using to interpret the protocol? I can't find it in the download section

Thanks for the data. There is a separate bit for the calibration. I've added a control for it on channel 11 (>0 is calibrate on). The test build is updated. Please give it a try. I don't know if you'll still have to put the sticks into the calibrate position or not - depends on the receiver.

Look in the repo utils/spi directory for format_spi.pl. It will take the hex csv file and interpret it according to the radio chip (default is nrf24l01). Run it with 'perl format_spi.pl <csv filename>' and it will print to stdout.

@ Hexfet

Thanks, it works. I have put ch11 on SW A1 and as soon as I switch from SW A0 to SW A1 and back it calibrates. I did not needed the sticks.
I tried also with only the sticks (and SW A1 off) => no calibration.
Would be great if it can be accomplished with both sticks to the lower left like with stock tx if that is possible?

Thanks.

hexfet wrote: Thanks for the data. There is a separate bit for the calibration. I've added a control for it on channel 11 (>0 is calibrate on). The test build is updated. Please give it a try. I don't know if you'll still have to put the sticks into the calibrate position or not - depends on the receiver.

Look in the repo utils/spi directory for format_spi.pl. It will take the hex csv file and interpret it according to the radio chip (default is nrf24l01). Run it with 'perl format_spi.pl <csv filename>' and it will print to stdout.

I just updated the test build and was going to try, but it looks as if the FY326 protocol is limited to 10 channels ... do you need to change a setting in your code to enable channel 11? I think that the max number of channels available s defined by the protocol

Alias_Hendrik wrote: @ Hexfet

Thanks, it works. I have put ch11 on SW A1 and as soon as I switch from SW A0 to SW A1 and back it calibrates. I did not needed the sticks.
I tried also with only the sticks (and SW A1 off) => no calibration.
Would be great if it can be accomplished with both sticks to the lower left like with stock tx if that is possible?

Thanks.

@ Robca
Change the channels in the ini to 11 or to 11 in the "model setup" in the Devo.

Alias_Hendrik wrote:

Alias_Hendrik wrote: @ Hexfet

Thanks, it works. I have put ch11 on SW A1 and as soon as I switch from SW A0 to SW A1 and back it calibrates. I did not needed the sticks.
I tried also with only the sticks (and SW A1 off) => no calibration.
Would be great if it can be accomplished with both sticks to the lower left like with stock tx if that is possible?

Thanks.

@ Robca
Change the channels in the ini to 11 or to 11 in the "model setup" in the Devo.

You actually can't. and that's my point if I choose another protocol (say, DSM2), I can change the "# of channels" in model setup to 11 or 12. If I choose the FY326 protocol, it maxs out at 10 channels. I have not tried editing the INI file yet...

In the Deviation code, the protocol defines the max number of channels supported. That needs to be addressed for the protocol to be fully functional

Strange because in my 7E if I look in "model setup" it is set at 11 channels. With previous test builds the max. I could set was 10 channels. With this testbuild the max I can set is 11 channels.

In my main menu => "about deviation" the version is 4.0.1 50ad99f

Max channels in the current test build goes to 11

Alias_Hendrik wrote: Thanks, it works. I have put ch11 on SW A1 and as soon as I switch from SW A0 to SW A1 and back it calibrates. I did not needed the sticks.
I tried also with only the sticks (and SW A1 off) => no calibration.
Would be great if it can be accomplished with both sticks to the lower left like with stock tx if that is possible?
Thanks.

Thanks for the testing. The nice thing about deviation is it can be accomplished with the mixers. You can set up any stick combo you'd like.

How could I forgot about that i'm still on a learning curve