🩹 Fix MKS Gen-L V1 pins, allow more RAMPS overrides

[thisiskeithb]

Description

MKS Gen-L V1.0 moseft pins keep getting changed, with the latest being in #25717 since they assumed MKS' schematic was correct & the pins diagram wrong. I physically traced the pins on my MKS Gen-L V1.0 board and the pins diagram is correct, not the schematic.

They are:

     HEATER_BED = 8
       HEATER_0 = 10
HEATER_1 / FAN1 = 7 (FAN1 when used in a single hotend config / frequently used for E0_AUTO_FAN_PIN)
           FAN0 = 9

I fully defined the pins as they are on the board itself and added silkscreen labels & Marlin's custom MOSFET_*_PIN naming scheme as comments and disabled them for documentation purposes. Some #ifndefs were required in pins_RAMPS.h to prevent a wall of warnings due to pins being redefined.

Also, this board only ships with an ATmega2560, so I updated the board environment to reflect that.

Requirements

MKS Gen-L V1.0

Benefits

This will hopefully be the last MKS Gen-L V1.0 change since I physically traced pins in question to the MCU pins.

Configurations

Configs from #26971 or Artillery Genius BLTouch configs from our Configurations repo.

Related Issues

• [BUG] MKS_GEN_L extruder fan not working #26971
• Fix MKS Gen L V1.0 MOSFET & FAN Pinmap (Artillery) #25717

[thisiskeithb]

After chatting with @ellensp, I reverted to using the default MOSFET_*_PINs as defined in pins_RAMPS.h, but left the correct pins (commented out) in pins_MKS_GEN_L.h for documentation purposes so there's no confusion as to what they should be in the future.

[thisiskeithb]

Merge conflicts caused by 9342dae have been resolved.

[sjasonsmith]

@ellensp would you mind reviewing and commenting on this? I'm a bit confused by the back and forth regarding which incorrect pins to use versus comment out. It sounds like you know what's going on and will be better able to assess whether this is ready to go in than I am.

[thisiskeithb]

The commented out pins are there for documentation purposes. I physically traced out the connectors/mosfets to the MCU pins on my MKS Gen-L V1 and verified they are correct.

This PR is ready to merge.

[sjasonsmith]

I just still don't understand what they mean. They are just commented out pins "for documentation purposes", but do they represent pins that are wrong? Are they from the schematic, from the board? I'm left guessing what a commented out pin names.

Perhaps more explanation is warranted in the comments to describe the errors in the schematic, rather than just the single line saying "there are errors on the schematic". I'm not sure how to interpret that information as I look at the rest of the file.

[thisiskeithb]

I just still don't understand what they mean. They are just commented out pins "for documentation purposes", but do they represent pins that are wrong? Are they from the schematic, from the board? I'm left guessing what a commented out pin names.

I left some review comments, so hopefully that clears things up.

Perhaps more explanation is warranted in the comments to describe the errors in the schematic, rather than just the single line saying "there are errors on the schematic". I'm not sure how to interpret that information as I look at the rest of the file.

I'm not able to review MKS' schematic for all the pin & connector errors, but their pins diagram PDF is correct.

<sarcasm>Maybe there needs to be a STOP CHANGING THESE PINS warning at the top instead?</sarcasm> 😄

[sjasonsmith]

What do you think of the more explicit documentation of these that I suggest in this review?

[thisiskeithb]

What do you think of the more explicit documentation of these that I suggest in this review?

All suggested changes accepted/merged. Should be good to go once CI is done.

[thinkyhead]

Either the MOSFET pin scheme needs to be retained, or this file needs to do the handling of the flags *_EFB, *_EEB, etc., for cases where the Extruder, Fan, Bed arrangement may change. These flags may already be set for the HAS_MULTI_HOTEND and/or HEATERS_PARALLEL and should be preferred over referring to these settings flags directly.

[thinkyhead]

Reminder that the rule for Extruder, Fan, Bed pins is not to always define them according to their silkscreen label, but according to the configuration requirements. The board may say something like "H1, Fan, Bed" which would correspond to the common default FET_ORDER_EFB. But when two extruders are defined, the FET_ORDER_EEB flag gets set, so the pins should then be defined as "H1, H2, Bed" with no fan.

If the FET_ORDER_* flags aren't coming out right for this board according to those flags, either these FETs are just not defined in the correct sequence, or the general FET assignment in pins_RAMPS.h is wonky.

[thinkyhead]

Hmm, I wonder if it work just using the order 10, 9, 8, 7… But we have to look at the machines these ship in also and see if they are using non-standard ordering on shipping units.

[thisiskeithb]

Either the MOSFET pin scheme needs to be retained, or this file needs to do the handling of the flags *_EFB, *_EEB, etc.,

I really don't follow all that FET_ORDER stuff since we don't do that on non-AVR boards, but I have physically traced the pins on the MCU and verified that they now match real hardware.

[thisiskeithb]

Hmm, I wonder if it work just using the order 10, 9, 8, 7… But we have to look at the machines these ship in also and see if they are using non-standard ordering on shipping units.

We host several Artillery configs and the older/non-32-bit printers use a standard Gen-L: https://github.com/MarlinFirmware/Configurations/tree/import-2.1.x/config/examples/Artillery