FY326 Q7 "red board"

hexfet wrote: Max channels in the current test build goes to 11

My bad. I flashed the DFU, and stupidly forgot to update the protocol directory

I just tested it with the right protocols, and indeed it works well. as soon as I flip the switch I set for CH11, it properly recalibrate. Well done, hexfet!

(and thanks for the immortal "goes to 11" reference )

I don't seem to be able to get any of the trims to work.

I'm still a novice with the 7E, so can others confirm this?

Have the FY326 throttle dropouts fully gone away with Deviation?

Check the Dyntrim settings and turn it on/off to see if it makes any difference.

What do you mean by "the trims to work", btw?

It depends on your definition. The Red board interprets the range between -100 and -70 (more or less) as motors off, from -70 to +100 as motor on with different speed. The problem is on the quad side, not the Tx side

You can program your throttle channel to go from -100 to -70 at the beginning of the throttle joystick movement, and then progress more linearly from there. In that case, there will be no "dead range" in the lower part of the throttle movement. I.e. the problem de facto disappears

robca wrote: Check the Dyntrim settings and turn it on/off to see if it makes any difference.

Toggling Dyntrim on or off makes no difference.

robca wrote: What do you mean by "the trims to work", btw?

The LCD display responds to trim settings as one would expect, but the behaviour of the quad is unaffected.

Check the Channel Monitor (channel monitor is your best friend )... trims have very small effect on Deviation, usually you need 10-15 clicks in one direction to see a real change (especially for a low end quad like the FY326). Try looking at how much the neutral position (0 for anything but the throttle) changes when you click. Usually it takes at least 2-3 clicks for the 0 to become -1 or +1 (depending on the direction you choose)

Also make sure that in your model menu you have Trims enabled (the last entry in the Mixer screen)

I noticed the same as S.Giles. That is, trying to correct drift (if it would be necessary) with the trim buttons for pitch and roll did not work. I did not tested it with this build because it was working with one of the previous builds. But after S.Giles mentioned it for this build I tested it with this build too and it did not work for me either.
Setting Dyntrim to off made no difference (but I will check that again).

Regarding Dyntrym, as I think what it does, trim values are added to the stick values or something to extend their range.
Perhaps a few stupid questions but when Dyntrim is implemented and enabled;


1) Because of added extension and sending full stick and full trim values some quads can react different to it regarding speed? for example a difference between forward and backward pitch and hence the reason for "cutting the extension in half" so the "extension" is equally added for forward and backward pitch?

2) with Dyntrim iimplemented and enabled, do we have the extended range even without setting a scalar above 100? Will anything above 100 not be necessary and we still have the extended range?

3) can the trims still do their job to correct drift? Or is that not possible because the range (full or half) from the "trims" is added to the stick range?

I know, probably stupid questions but a clear answer to the above could help me and i'm sure others too to understand how it works in practice and what to expect when testing test builds.
Writing In English all the above is rather difficult to write for me but I hope my questions are readable and already a big thanks to anyone who can shed some light on the above questions

Edit; the max value variation i can see changing in the channel output monitor using the trims for pitch and roll are 10 and -10

robca wrote: Also make sure that in your model menu you have Trims enabled (the last entry in the Mixer screen)

Problem solved with your help robca. Thank you.

The 'step' settings were at a default of 0.1, which was apparently too subtle to have a noticeable effect. Changing all the trims to 0.5 has made them work as I would expect. I will do a little more experimenting, but it's looking good.

I still don't fully understand this aspect of setting up the 7E, but I know more now than I did ten minutes ago!

It's a confusing subject because "trim" here means two different things. If this explanation doesn't answer your questions feel free to ask more.

The trim controls in Deviation are part of the mixer. Any non-zero trim is added to the channel value calculated by the mixer (based on source value, scaling, expo, etc). After the final channel output value is calculated the protocol transfers the value to the aircraft. In Deviation the trimming is independent of the protocol, and as far as I know all hobby-grade equipment works this way.

Some of the toy quad protocols handle trim settings differently. Instead of being combined with the channel value in the tx the trim settings are sent to the aircraft as separate data values in the radio packets. In this context "trim" refers to the trim packet data. Deviation doesn't work this way and we want to maintain the hobby-grade interface, so the trim data in the toy quad protocols is unused in the sense of trimming drift.

Dynamic trim is the name that emerged to describe the technique of employing the unused trim packet data to extend the control range of the aircraft. It has nothing to do with trimming drift. The idea of dynamic trim is to set both channel and trim data in the packets sent to the quad based on the mixer output. So at full stick travel on the tx the values in the protocol packet are max channel and max trim. Theoretically that should produce the same flight characteristic as having the stock tx trim set fully in one direction, and moving the stick in that same direction.

The effect of dynamic trim on the aircraft depends on how the receiver adds the trim to the channel values. Generally if the trim on the stock tx has a large effect then dynamic trim will also. However some receivers seem to have issues with the channel and trim values changing rapidly simultaneously, which is why the testing is needed. Ideally in the final version there will be no dyntrim option and it won't be visible to users. On the quads where using the full trim range caused issues, testing found that dyntrim would work if the range was reduced.

Thank you hexfet. That explains how trim can be set in 0.1 increments, even though the protocol supports a value change of 1 for each trim button click. That's what was confusing me.

Thanks Hexfet I really appreciate your help and although I have no H8 mini or MJX I also follow those threads to learn as much as possible

I've updated the test builds to remove the dyntrim protocol option. I left the dynamic trim enabled as it seemed to do no harm, though didn't seem to make much difference in the controls.

Unless problems are reported this will be the pull request version. I left the "expert" protocol option, mainly so everything in the protocol can be controlled.

Hi Hexfet, thanks I was waiting to see if other testers noticed a difference between Dyntrim on or off.
Last friday I received a Headless mode model and sinds then it's bad weather so could not test much.

However I can confirm Headless and one key return works with the Devo 7E.

I have Headless / one key return on the left vertical trim buttons like on stock tx and "calibrate" on the left horizontal trim button with 2 conditions both need to be met before calibration is possible; Fmode switch needs to be in position 1 and the throttle stick position have to be at least - 96 before one can calibrate to avoid bad things during flight when pushing buttons randomly.

With stock tx during "sport" flying I reached sometimes the "both sticks to down left corner" which resulted in crashing lol.
With Devo trying buttons to see what button does what, one could enter callibrate during flight. Hence the reason I spend hours in the mixers lol yep still have to learn the advanced mixers but it does work finally for the conditions to be met before i can "calibrate".

One key return works separately from Headless mode just like with stock TX. The only difference, with stock tx you can also stop the "coming back to you" by moving the right stick on stock TX but not with the Devo. With the Devo you need to push the trim button again.

What i noticed on stock tx and Devo, when entering Headless mode, the rates are reduced too as long as one is in headless mode. The reduced rate is slow and because of that it's not a good idea to try Headless mode in the wind. Don't ask me how I know

Big thanks to you to make it possible for us to fly the FY-326 red board, headless or non headless with the Devo

Just a couple of questions:

1/ Would it be possible to make the TX beep when arming the motors with the LH stick, like the stock TX does? It's maybe not that important, but it would be reassuring to have an audible confirmation that arming is happening. (I notice the lack of a beep every time I do this.)

2/ Is documentation of the various extra functions produced as part of the protocol development? As a newcomer to deviation and the 7E, who already has three different models of quadcopter stored in the 7E's memory, it would be helpful to have a quick reference of how to arm, calibrate, enter flip mode, select different rates, etc for each protocol used. Maybe there's something already available, but if it is, I haven't seen it.

Thanks.

The "throttle up/down to enable" is a safety function of the tx, not part of the protocol. The equivalent on deviation would be a throttle hold switch. There've been a couple recent threads about making a sound based on switch movement.

The channel assignments for the toy quad protocols are listed in the user manual in the protocol section. Still need to add the newer protocols. But all options may not apply to all quads that use the protocol.

S.Giles wrote:
...

2/ Is documentation of the various extra functions produced as part of the protocol development? As a newcomer to deviation and the 7E, who already has three different models of quadcopter stored in the 7E's memory, it would be helpful to have a quick reference of how to arm, calibrate, enter flip mode, select different rates, etc for each protocol used. Maybe there's something already available, but if it is, I haven't seen it.

Thanks.

It would be nice to have a legend of some sort within a model file that could be displayed on the radio to remind you of this model's switches and features.

Just tried out the FY326 this evening in the yard. Very windy and gusty. Quad seemed to have the old dropouts, until I remembered to dial up the power. Very solid! Very snappy. Fought heavy gusts nicely. No dropouts. Solid responsiveness.

Fixed! Made a nice quad out of an unusable one.

Nice job guys! Thanks.

Something that has just occurred to me...

The red-board FY326's PCB has pads that look like they're meant to control an on-board camera. Is there any chance that this facility could be made operative in Deviation? It looks like there's a +ve and -ve terminal, and one marked 'S' (IIRC, - I don't have my FY326 apart at the moment so can't check).

The stock transmitter doesn't support any camera functions, so I'm not sure how you'd figure out what the TX payload would have to include to make it work.

Apparently, the FY326 has the ability to easily haul a camera aloft, probably due to its 8.5mm motors, and it seems a pity not to utilise any latent camera functions that may be lurking unused on the PCB.

Is this code now merged to Deviation firmware? Tried to pair my fresh FY326 with a red board and it keeps flashing lights at me (devo 7e still trying to connect press ENT to stop...) Am I missing something?