Flush_handler error on a toolchange | VORONDesign | Page 1

halcyon bloom Mar 22, 2024, 11:41 AM

#

On a toolchange with the ERCF running HH I get 'flush_handler' errors with the printer shutting down.

Internal error on command:"MMU_CHANGE_TOOL"
Traceback (most recent call last):
  File "/home/pi/klipper/klippy/gcode.py", line 211, in _process_commands
    handler(gcmd)
  File "/home/pi/klipper/klippy/gcode.py", line 137, in <lambda>
    func = lambda params: origfunc(self._get_extended_params(params))
  File "/home/pi/klipper/klippy/extras/mmu.py", line 5417, in cmd_MMU_CHANGE_TOOL
    if self._change_tool(tool, skip_tip, next_pos):
  File "/home/pi/klipper/klippy/extras/mmu.py", line 5187, in _change_tool
    self._unload_tool(skip_tip=skip_tip)
  File "/home/pi/klipper/klippy/extras/mmu.py", line 5141, in _unload_tool
    self._unload_sequence(skip_tip=skip_tip, runout=runout)
  File "/home/pi/klipper/klippy/extras/mmu.py", line 4210, in _unload_sequence
    self._initialize_filament_position(dwell=None)    # Encoder 0000
  File "/home/pi/klipper/klippy/extras/mmu.py", line 3141, in _initialize_filament_position
    self._set_filament_position()
  File "/home/pi/klipper/klippy/extras/mmu.py", line 3149, in _set_filament_position
    self.mmu_toolhead.set_position(pos)
  File "/home/pi/klipper/klippy/extras/mmu_toolhead.py", line 165, in set_position
    super(MmuToolHead, self).set_position(newpos, homing_axes)
  File "/home/pi/klipper/klippy/toolhead.py", line 456, in set_position
    self.flush_step_generation()
  File "/home/pi/klipper/klippy/toolhead.py", line 371, in flush_step_generation
    self._advance_flush_time(self.step_gen_time)
  File "/home/pi/klipper/klippy/toolhead.py", line 310, in _advance_flush_time
    m.flush_moves(flush_time, clear_history_time)
  File "/home/pi/klipper/klippy/mcu.py", line 954, in flush_moves
    raise error("Internal error in MCU '%s' stepcompress"
mcu.error: Internal error in MCU 'sb2040' stepcompress

📎 klippy35.log

halcyon bloom Mar 22, 2024, 2:07 PM

#

This is a bit further back in the file

b'stepcompress o=6 i=0 c=720 a=0: Invalid sequence'
Exception in flush_handler
Traceback (most recent call last):
  File "/home/pi/klipper/klippy/toolhead.py", line 447, in _flush_handler
    self._advance_flush_time(min(end_flush, ftime))
  File "/home/pi/klipper/klippy/toolhead.py", line 310, in _advance_flush_time
    m.flush_moves(flush_time, clear_history_time)
  File "/home/pi/klipper/klippy/mcu.py", line 954, in flush_moves
    raise error("Internal error in MCU '%s' stepcompress"
mcu.error: Internal error in MCU 'sb2040' stepcompress
Transition to shutdown state: Exception in flush_handler

halcyon bloom Mar 22, 2024, 2:38 PM

#

b'stepcompress o=6 i=0 c=720 a=0: Invalid sequence'
This means that the steps have a 'oid' of 6 (whatever that means), an 'interval' and 'add' of 0, and a 'count' of 720

https://github.com/Klipper3d/klipper/blob/master/klippy/chelper/stepcompress.c#L206

static int
check_line(struct stepcompress *sc, struct step_move move)
{
    if (!CHECK_LINES)
        return 0;
    if (!move.count || (!move.interval && !move.add && move.count > 1)
        || move.interval >= 0x80000000) {
        errorf("stepcompress o=%d i=%d c=%d a=%d: Invalid sequence"
               , sc->oid, move.interval, move.count, move.add);
        return ERROR_RET;
    }
    .
    .
    .
    .
}

#

This function states that interval and add can't be 0 if count is more than 1

I just don't know where these values come from and why it's causing an error. My knowledge only goes this far...

worldly reef Apr 3, 2024, 7:46 AM

#

I am also getting this issue

#

@halcyon bloom did you manage to sort this?

halcyon bloom Apr 3, 2024, 7:52 AM

#

worldly reef <@222728308326924298> did you manage to sort this?

Yes! Sort of... In my case it happened when my ERCF running HH failed to load filament and retried to load it. There were two settings in HH that couldn't be combined. Don't enable retry if you disable the encoder readings.

worldly reef Apr 3, 2024, 7:54 AM

#

Do you know the lines of those 2 commands?

#

I have this disabled
retry_tool_change_on_error: 0

halcyon bloom Apr 3, 2024, 7:55 AM

#

retry_tool_change_on_error: 1    # Whether to automatically retry a failed tool change. If enabled Happy Hare will perform
                # the equivalent of 'MMU_RECOVER' + 'Tx' commands which usually is all that is necessary
                # to recover. Note that enabling this can mask problems with your MMU
print_start_detection: 1    # ADVANCED: Enabled for Happy Hare to automatically detect start and end of print and call
                # _MMU_START_PRINT and _MMU_END_PRINT. Disable if you want to include in your own macros
show_error_dialog: 1        # 1 = show pop-up dialog in addition to console message, 0 = show error in console
encoder_move_validation: 1    # 1 = Normally Encoder validates move distances are within given tolerance (slower but more safe)
                # 0 = Validation is disabled for many moves (eliminates slight pause between moves but less safe)

This is from line 357

halcyon bloom Apr 3, 2024, 7:55 AM

#

worldly reef I have this disabled retry_tool_change_on_error: 0

Hmmm, any clue as to when you encounter the error?

worldly reef Apr 3, 2024, 7:56 AM

#

I just do a toolchange and then the printer shuts down

#

halcyon bloom Apr 3, 2024, 7:56 AM

#

Can you share your [stepper_mmu_gear] and [extruder] sections?

worldly reef Apr 3, 2024, 7:59 AM

#

uart_pin: mmu:MMU_GEAR_UART
run_current: 1.0            # ERCF BOM NEMA14 motor
hold_current: 0.1            # Recommend to be small if not using "touch" or move (TMC stallguard)
interpolate: True
sense_resistor: 0.110            # Usually 0.11, 0.15 for BTT TMC2226
stealthchop_threshold: 0        # Spreadcycle has more torque and better at speed
#
# Uncomment two lines below if you have TMC and want the ability to use filament "touch" homing with gear stepper
#diag_pin: ^mmu:MMU_GEAR_DIAG        # Set to MCU pin connected to TMC DIAG pin for gear stepper
#driver_SGTHRS: 60            # 255 is most sensitive value, 0 is least sensitive

[stepper_mmu_gear]
step_pin: mmu:MMU_GEAR_STEP
dir_pin: mmu:MMU_GEAR_DIR
enable_pin: !mmu:MMU_GEAR_ENABLE
rotation_distance: 22.7316868        # Bondtech 5mm Drive Gears. Will be overriden by `mmu_gear_rotation_distance` in mmu_vars.cfg
gear_ratio: 80:20            # ERCF 80:20, Tradrack 50:17
microsteps: 16                 # Recommend 16
full_steps_per_rotation: 200        # 200 for 1.8 degree, 400 for 0.9 degree
#
# Uncomment the two lines below to enable filament "touch" homing option with gear motor
#extra_endstop_pins: tmc2209_stepper_mmu_gear:virtual_endstop
#extra_endstop_names: mmu_gear_touch```

#

step_pin: shtcan:PB4 #
dir_pin: shtcan:PB3 #
enable_pin: !shtcan:PA15 #
##    rotation_distance = <previous_rotation_distance> * <actual_extrude_distance> / 100 #
rotation_distance: 28.01772 # 
gear_ratio: 50:8 #
microsteps: 16 #
full_steps_per_rotation: 200    #200 for 1.8 degree, 400 for 0.9 degree #
nozzle_diameter: 0.400
filament_diameter: 1.75
heater_pin: shtcan:PA8
sensor_pin: shtcan:PA3
sensor_type: ATC Semitec 104NT-4-R025H42G
#pullup_resistor: 2200
control: pid
pressure_advance: 0.05 #
pressure_advance_smooth_time: 0.020 #
pid_Kp: 22.274
pid_Ki: 1.768
pid_Kd: 70.163
min_temp: -100
max_temp: 325
max_extrude_only_distance: 200
max_extrude_cross_section: 50.0

[tmc2209 extruder]
uart_pin: shtcan:PB5
run_current: 0.4
stealthchop_threshold: 0
interpolate: true```

halcyon bloom Apr 3, 2024, 8:01 AM

#

Try setting microsteps: 8 in the extruder section, and leave the gear at 16. If that doesn't work, swap them around.

worldly reef Apr 3, 2024, 8:25 AM

#

same issue wioth both of those ways

halcyon bloom Apr 3, 2024, 8:30 AM

#

Sorry, Then I'm out of options 😦

worldly reef Apr 3, 2024, 2:41 PM

#

Ok, this error is specific to HHv2 and DangerKlipper. if you get it:

#

add this to line 30 of mmu_toolhead.py (in /home/pi/Happy-Hare/extras)
self.danger_options = self.printer.lookup_object("danger_options")
reboot, should be golden

halcyon bloom Apr 3, 2024, 3:06 PM

#

What does that do?

worldly reef Apr 3, 2024, 4:50 PM

#

No idea

#

😉

#Flush_handler error on a toolchange