GCS Module: 1wire protocol 4w-if SK/BL ESC flash, settings

[fujin]

It may be viable to form a partnership with either the team behind Rapidflash, or the BLHeliSuite team, and implement a GCS pane for ESC configuration. If the configuration/flash protocol are open, or simple enough, such a partnership might not even be required, but some consideration must be given to the delta between upstream updates / new features and the TL UI lagging behind.

I have no doubt that we would be able to do enough to flash and re-configure ESCs with no change to wiring required.

Thoughts?

cheers

[fujin]

#1687 -- rapidflash Chrome configurator reference may come in handy.

[fujin]

@Brotronics kindly pointed me at https://github.com/DTFUHF/baseflight/blob/usblinker/src/usblinker.c

[dustin]

I would love this. I've had a couple builds where getting to ESCs is just hard enough to make this unpleasant. Not only for changing direction, but also experimental reconfiguration (e.g. comp_pwm) and builds.

[peabody124]

I also think this would be an awesome feature. As a suggested implementation method, I would write a PIOS_COM interface that bit bangs the serial protocol. I did something very similar in my ESC code using a soft USART that sat on the PWM lines: https://github.com/peabody124/TauLabs/blob/esc/flight/PiOS/STM32F10x/pios_softusart.c

[fujin]

https://www.youtube.com/watch?v=2N6L8Igcgc4

[fujin]

https://github.com/nathantsoi/cleanflight/blob/serial1wire/docs/1wire.md

[dustin]

IMO, the Cleanflight way is a really dumb cop-out solution. OpenPilot did a much better job (though there's more work to do) and I think Tau Labs can do even better.

I want "install image on all my ESCs" and potentially "motor 3 is backwards." Options, presumably, should be consistent across all of them.

Using my flight controller as a serial adaptor feels very lame and platform specific.

[Trex4Git]

I think for the CF way much less work would be needed.
As BlHeli is also under active development, and a lot of work is already done in BlHeliSuite especially as a first step implementing an "interface passthrough" in Taulabs should be fine.

Otherwise it would be a lot more work to get the BlHeliSuite functionality into GCS, and also someone must do the maintaining when there are changes in new BlHehli Versions.

[dustin]

Doesn't OP's work on more than just Windows? Because this
proposal/cleanflight's is Windows only.

On Tue, Jul 28, 2015 at 12:24 AM Trex4Git notifications@github.com wrote:

I think for the CF way much less work would be needed.
As BlHeli is also under active development, and a lot of work is already
done in BlHeliSuite especially as a first step implementing an "interface
passthrough" in Taulabs should be fine.

Otherwise it would be a lot more work to get the BlHeliSuite functionality
into GCS, and also someone must do the maintaining when there are changes
in new BlHehli Versions.

—
Reply to this email directly or view it on GitHub
#1690 (comment).