✨Mellow Fly E3 Pro V3 (STM32F407VG)

[jaysuk]

Description

Based on the Fly-E3-v2 mainboard that was merged into Marlin, this is for the Fly-E3-Pro-V3

Requirements

A Fly-E3-Pro-V3

Benefits

Adds support for the Fly-E3-Pro-V3 board

Configurations

Related Issues

None

[thisiskeithb]

Since this is a duplicate PR, it'd be better to send PRs to #25894 / https://github.com/IoakeimSogiakas/Marlin/tree/FLY_RRF_MARLIN

[jaysuk]

i've sent a message to the author but not heard anything back.
can we not just merge this one and close the other PR?

[thisiskeithb]

can we not just merge this one and close the other PR?

That’s not how it generally works. Users collaborate via a single PR/fork, not open multiple PRs.

[ellensp]

PR #25894 closed, superseded with this on of this one

[thisiskeithb]

I guess that settles that.

[ellensp]

original pr was not finished and seems to be abandoned. this is a step forward

[jaysuk]

@ellensp thanks

[jaysuk]

As far as I'm concerned this is ready to merge

[jaysuk]

anything else i need to do with this?

[ellensp]

be patent, main developer is taking a break for an unknown period of time

[sjasonsmith]

@jaysuk can you update this to work with the latest bugfix-2.1.x and push new changes? If so we should be able to do a final review of this and get it in soon.

[jaysuk]

@sjasonsmith updated to the latest. only had to change the boards file

[thisiskeithb]

I also just pushed 55a109e to fix a few small things in pins_MELLOW_FLY_E3_PRO_V3.h.

[jaysuk]

thanks, didn't even think to check those. Once this is merged I will move on to the next board. I still have the Fly-CDYv3 and Super8 boards to do in the F4 range. Then its on to the H7 ones

[sjasonsmith]

I noticed that this supports a lot of peripherals with custom pinouts, such as motor expanders and various temperature sensors.

How confident are you that all of these are correct for this board?

[jaysuk]

i checked all the pins against the pinouts of the modules etc.
i maintain all the docs for the RRF port so i know the boards "fairly" well

[thisiskeithb]

PR has been brought up to date & I've pushed some more updates to pins_MELLOW_FLY_E3_PRO_V3.h. See my question above about flash drive support.

[thisiskeithb]

I just discovered this note about the Mellow Fly E3 Pro V3's thermistor pull-ups:

Important

The Fly-E3-Pro-v3 uses a 2k2 resistor on the thermistor inputs rather than 4k7. Please make sure you add R2200 to each M308 in your config

I ran into this while working on LDO Leviathan board support and it means that users will have select thermistor 1000 for any thermistor they want to use and will have to define all the custom thermistor 1000 parameters found in Configuration_adv.h since Marlin assumes 4.7K as the default pull-up type for most thermistors.

I also saw the note about needing to add jumpers on Z2's output (I assume they are driven in series?), so adding a warning to the pins file would be a good idea.

[thosoo]

When building this now I get the following error message:

`Marlin\src\HAL\STM32\timers.cpp:328:40: error: static assertion failed: One or more timer conflict detected. Examine "timers_in_use" to help identify conflict.
  328 | static_assert(verify_no_timer_conflicts(), "One or more timer conflict detected. Examine \"timers_in_use\" to help identify conflict.");
      |               ~~~~~~~~~~~~~~~~~~~~~~~~~^~
Compiling .pio\build\STM32F407VG_mellow_fly_e3\src\src\HAL\shared\backtrace\unwinder.cpp.o
*** [.pio\build\STM32F407VG_mellow_fly_e3\src\src\HAL\STM32\timers.cpp.o] Error 1
======================================= [FAILED] Took 46.96 seconds =======================================

Environment                Status    Duration
-------------------------  --------  ------------
STM32F407VG_mellow_fly_e3  FAILED    00:00:46.957
================================== 1 failed, 0 succeeded in 00:00:46.957 =================================

[mlewis-everley]

When building this now I get the following error message:

`Marlin\src\HAL\STM32\timers.cpp:328:40: error: static assertion failed: One or more timer conflict detected. Examine "timers_in_use" to help identify conflict.
  328 | static_assert(verify_no_timer_conflicts(), "One or more timer conflict detected. Examine \"timers_in_use\" to help identify conflict.");
      |               ~~~~~~~~~~~~~~~~~~~~~~~~~^~
Compiling .pio\build\STM32F407VG_mellow_fly_e3\src\src\HAL\shared\backtrace\unwinder.cpp.o
*** [.pio\build\STM32F407VG_mellow_fly_e3\src\src\HAL\STM32\timers.cpp.o] Error 1
======================================= [FAILED] Took 46.96 seconds =======================================

Environment                Status    Duration
-------------------------  --------  ------------
STM32F407VG_mellow_fly_e3  FAILED    00:00:46.957
================================== 1 failed, 0 succeeded in 00:00:46.957 =================================

I just merged the latest bugfix-2.1.x into this branch locally and then built this and it has all compiled OK. Maybe try that?

Just wondering, what might need doing to get this added and if there is anything I can do to expedite things?

[mlewis-everley]

So been trying out Marlin with this board, both on bugfix and stable, can get it to compile and install, but I am getting some odd behavior.

1. When stepper motors are enabled, H0 begins heating and carries on until max temp is triggered.
2. Trying to run dual independent Z, the second Z motor runs backwards when increasing the height of the gantry and then runs in the correct direction when decreasing height. Basically, the second Z stepper always moves down. Trying to change INVERT_Z2_VS_Z_DIR seems to have no effect.

This behavior is consistent on both bugfix and stable.

[mlewis-everley]

OK, so I noticed that @thisiskeithb added some amends to the initial pull request, so I have applied those to my fork and the Z-stepper issue has gone. Though now all the steppers are REALLY loud, even in stealthchop mode. I am not sure if this is because I had to change the X_RSENSE value (it was 0.15)?

Just got to try and work out what is causing the heater issue...

[mlewis-everley]

Slowly chipping away at this, looks like the pin definitions for FAN2_PIN and FAN3_PIN are incorrect. They should be:

#ifndef FAN2_PIN
  #define FAN2_PIN                            PB6
#endif
#ifndef FAN3_PIN
  #define FAN3_PIN                            PB7
#endif

Currently FAN2_PIN is mapped to HE0 :-/.

Not sure what the preferred approach is here, does @jaysuk need to update their pull request?

[thisiskeithb]

Though now all the steppers are REALLY loud, even in stealthchop mode. I am not sure if this is because I had to change the X_RSENSE value (it was 0.15)?

I'm not sure why that would be different.

Slowly chipping away at this, looks like the pin definitions for FAN2_PIN and FAN3_PIN are incorrect. They should be:
[...]
Not sure what the preferred approach is here, does @jaysuk need to update their pull request?

I've merged in the latest bugfix-2.1.x and updated the fan pins, so give this PR another try.