You can add Octoquad under VTOL in the Firmware Builder for that FC if other stuff is disabled. Think that would work?
I wouldn’t know as I wouldn’t use those Flight Controllers for any purpose.
No, I run Pixhawk 2.4.8 controllers. It’s not a hardware issue, I didn’t experience it until I updated the firmware to 4.6, then when I rolled back to 4.5.7 it disappeared. Something was changed in the firmware update.
Maybe for those less than capable FC’s such as these.
So for anyone interested, I used the custom firmware builder.
After selecting my flight controller, I clicked the drop-down box in the lower left dialog box (titled VTOL FRAME) and lo and behold, OCTAQUAD (X8 quadcopter) was NOT checked.
I checked the box, and a couple minutes later, a custom 4.6.1 firmware was ready. I installed that firmware, and my problem is solved.
Thanks everyone.
2 Likes
Interesting, you mention 4.6.1, thanks, I’ll try it. As mentioned several times now, the only trouble I’ve had is with 4.6.0.
Such as which? Nothing ‘less than capable’ about Pixhawk controllers.
Any F4 based controller. They are all feature limited.
What “pixhawk controllers”? Ones from Holybro?
The idea of ‘limited’ depends on what ‘features’ you want. A controller that has all the features I want cannot be limited, it does the job I want it to do. Might as well argue a family sedan is ‘feature limited’ when you want a wank tank or F1 racer, of course it is. The argument is pointless. Still fails to address the question of the bug in 4.6.0 failing to recognise the frame, when all previous versions happily did. Seems there is another update, 4.6.1, since I first raised the matter, I’ll have a look at that next before making any further comment. Thanks.
1 Like
@gilbo211_25, let’s stick to facts: F4 autopilots are literally feature limited because they have to be. By design, ArduPilot firmware limits the feature set to a subset that fits in flash memory and runs in available runtime memory space. Whether that limits your ability to use them is rather immaterial to the discussion at hand (other than the fact that recently imposed limitations have impacted frame type selection).
That said, your specific issue appears separate from those attempting OCTAQUAD configurations on F405 boards. You are using a hex frame which remains enabled on feature minimized boards (and the OCTAQUAD limitation doesn’t hit any F427 boards until another firmware release).
Did you ever send a log to @rmackay9? Your issue is more serious than the explainable ones that followed your original post.
1 Like
My opinion is that your and my problems are different.
My problem with the Speedybee F405 Wing is that “octaquad” was not included in the published firmware - I checked back to 4.5.7 and octaquad was not included then either - maybe/probably never has been - EVEN THOUGH THERE IS ROOM to include it.
I wonder why certain features are left out of the firmware build for certain flight controllers (the 1MB flight controllers, which includes the 2.4.8 Pixhawk), even though there was/is room to include those features?
Anyway, since you are not running an octaquad, (right?), then your frame issue is something else.
Hi @wsalopek,
Thanks for helping get to the bottom of the issue.
We will review whether we can put back the octaquad in tomorrow’s dev call. One limitation we have though is that all F4 boards need to have enough space. So even if it could fit on one board, it might not on another board of the same class. We support so many boards that we can’t realistically have a list of included/excluded features for each board.
I have access to no more here than words you deploy in online text-boxes, whatever ‘feature limited’ is supposed to mean. Aside from my comment that all the features I want to use are available cannot possibly make Pixhawk feature limited; another way of saying this is other autopilots have enhanced features.Aside from my long background in electronics and in service and maintenance requiring me to communicate quickly and effectively with other people, for that reason I also have an Honours degree in English. To ‘stick to the facts’ with me you need to post comparative feature lists allowing me to read through the differences for myself. Then I will know what you’re talking about. The other issue I previously raised has yet to be resolved beyond QGroundControl reporting a bug. In the event I simply reverted to 4.5.7 pending another update beyond 4.6.0 which hopefully resolves the issue. Thanks.
Thanks for this explanation. All you’re saying is each board has different features, not that one board or other has limited features. Limitation can only refer to limits on use of a particular board, which simply does not emerge as an issue when that’s all someone wants to use it for.
Ok, by “feature limited,” I mean F4 autopilots do not contain the full set of features offered by ArduPilot.
The phrase was intended as an objective term to attempt quantify the issue at hand. Here is the source code. Of note in your case, AP_MOTORS_FRAME_HEXA_ENABLED is set to 1, which specifically enables it for “feature minimized” boards (to use the filename’s term), thus making your issue more concerning than the missing OCTAQUAD frame type mentioned by other users above.
To fully analyze the issue you presented, a log file was requested. If you have not provided it, we cannot proceed farther.
It’s probably a good thing that Octaquad (X8) is an optional feature for F4-based autopilots.
I can say for a fact that even an autopilot with an F7 can struggle with the Octaquad configuration once you add in harmonic notch filters, telemetry and a couple of other standard features. You’ll be looking at disabling an IMU (redundancy) to reduce the CPU load.
Of course over the years the firmware has increased in size, features and complexity. This can increase load which would make some configurations out of bounds or optional.
2 Likes
Problem is that octa is already in there - so its a discrepancy
1 Like
Reporting back as promised earlier in the week. I’d mentioned that I would update 4 machines to 4.6.1 which I did, then went back to 4.5.7 in each to double check, updated again to 4.6.0 for comparison, and finally back to 4.6.1, but unable to replicate the fault. Airframe type and style is now recognised without further ado, they all arm and run OK, so today while it’s still sunny I’ll take them out for a test flight. I’m learning, these flight controllers are hell smart and with patience will sort themselves out. Thanks anyway for all your help, it’s a learning curve.
2 Likes