MCU 'EBBCan' shutdown: Rescheduled timer in the past only on z-tilt & bed mesh

General Discussion
1

Basic Information:
Printer Model: Mercury One.1
MCU / Printerboard: Octopus & EBB36
Host / SBC: RPI4b+
klippy.log
klippy (1).log (2.0 MB)

I’m running a Mercury One.1 with RPI4B, Octopus, EBB36, CAN & Tap. I am able to home the printer xyz just fine, but every time I try to do a bed mesh or z-tilt I get the error “MCU ‘EBBCan’ shutdown: Rescheduled timer in the past This generally occurs when the micro-controller has been requested to step at a rate higher than it is capable of obtaining.”

I’ve tried lowering the frame rate of the picam but still get the error every time. All other functions of the printer work fine, just bed mesh and z-tilt cause the issues.

I have confirmed that i’m on the most up to date versions of klipper on all mcu’s as of yesterday, i’ve also confirmed that the printer is able to navigate to the probe locations without causing the error, it only occurs when i run z-tilt or calibrate bed mesh.

Here is my z-tilt config:
[z_tilt]
z_positions: 130,264 #stepper_z lead screw location
2,9 #stepper_z1 lead screw location
256,9 #stepper_z2 lead screw location

points: 235,215 #probe location stepper_z (rear right)
30,215 #probe location stepper_z (rear left)
30,25 #probe location stepper_z (front left)
235,25 #probe location stepper_z (front right)

speed: 100
horizontal_move_z: 4
retries: 4
retry_tolerance: 0.0325

Here is my bed mesh config:
[bed_mesh]
algorithm: bicubic
speed: 200
probe_count: 3,3
mesh_min: 40,40
mesh_max: 220,225
horizontal_move_z: 4
mesh_pps: 2
algorithm: bicubic
bicubic_tension: 0.2
fade_start: 1
fade_end: 10
fade_target: 0

2 
[stepper_z]
...
rotation_distance = 2
homing_speed = 8
microsteps = 32

EBBCan: mcu_awake=0.053 mcu_task_avg=0.000017 mcu_task_stddev=0.000011 bytes_write=12958 bytes_read=25758 bytes_retransmit=0 bytes_invalid=0 send_seq=1129 receive_seq=1129 retransmit_seq=0 srtt=0.001 rttvar=0.000 rto=0.025 ready_bytes=0 upcoming_bytes=0 freq=63999643 adj=63998864
heater_bed: target=0 temp=22.2 pwm=0.000 raspberry_pi: temp=35.0 mcu_temp: temp=34.8 ebb_temp: temp=27.8  sysload=0.22 cputime=4461.081 memavail=7498616 print_time=61.491 buffer_time=0.000 print_stall=0 
extruder: target=0 temp=21.7 pwm=0.000

Can you try to slow down your Z moves to 4 mm/s, maybe?

3

I just tried that and still had the same error only on z-tilt & bed mesh. The printer can home xyz just fine.

I’ve attached the log from after changing to 4mm/s
klippy.log (2.4 MB)

4

Looks like I missed something.
The basic idea is simple: with that Z reduction rate (2 mm rotational distance) and this speed, steppers do too many steps per second.
Toolhead board, on the other hand, tries to follow and query endstop with the same frequency.

So, we want to decrease the Z move speed so the probe will query less frequently.

Unfortunately, I can’t remember where the speed of the Z axis should be limited for probing.

Sent 97 9874.406022 9874.405487 38: seq: 1a, get_clock, trsync_start oid=1 report_clock=3410464598 report_ticks=480000 expire_reason=4, trsync_set_timeout oid=1 clock=3412064598, endstop_home oid=0 clock=3410464598 sample_ticks=960 sample_count=4 rest_ticks=80 pin_value=1 trsync_oid=1 trigger_reason=1
Sent 98 9874.971001 9874.970910 6: seq: 1b, get_canbus_status
Sent 99 9875.390199 9875.390108 6: seq: 1c, get_clock
...
Receive: 96 9875.308090 9874.971001 15: seq: 1c, analog_in_state oid=10 next_clock=3417766584 value=31509
Receive: 97 9875.391148 9875.390199 11: seq: 1d, clock clock=3404319459
Receive: 98 9875.488070 9875.390199 14: seq: 1d, trsync_state oid=1 can_trigger=1 trigger_reason=0 clock=3410497056
Receive: 99 9875.493291 9875.390199 12: seq: 1d, shutdown clock=3410862885 static_string_id=Rescheduled timer in the past

endstop_home oid=0 clock=3410464598 sample_ticks=960 sample_count=4 rest_ticks=80 pin_value=1 trsync_oid=1 trigger_reason=1

rest_ticks - is how often the endstop should be queried.
64_000_000 / 80 = 800_000 - steps per second

One step for Z axis with your config is:
2 / 200 / 32 = 0.0003125

So, the requested Z speed is set to:
0.0003125 * 800_000 = 250 mm/s

I found it:

[probe]
pin = ^EBBCan:PB6
x_offset = 0
y_offset = 0
z_offset = 0
speed = 250
#speed: 5.0
#   Speed (in mm/s) of the Z axis when probing. The default is 5mm/s.
samples = 3

You don’t want to set it to 250 mm/s here.

5

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.