Manta m8p v1.1 + CB1 timer too close

General Discussion
1

Basic Information:

Printer Model: custom big printer
MCU / Printerboard: Bigtreetech Manta m8p v1.1
Host / SBC Bigtreetech CB1
klippy.log

klippy.zip (2.1 MB)

Hello everyone, I have the following problem:

Periodic Klipper crashes.

A bit about the printer:

· Z-axis gantry: heavy, 8 kg, balanced with 8 kg counterweights (inertial mass 16 kg)

· Z-axis driven by 4 NEMA 23 motors via belts, with 4 TMC5160T Pro drivers

· Core XY, NEMA 17, TMC2209

· Extruder: NEMA 14, TMC2209

The printer is 1000x1000 with a warped bed (up to 10 mm deviation).

What I’ve tried:

  1. Adjusting currents

  2. Routing power wires away from sensor wires and the mainboard

  3. Improving cooling (host temps now reach up to 40°C)

  4. Disabling all Klipper services not in use

  5. Checking wiring for shorts

  6. 2x headroom on drivers and 8x headroom on main PSU — 48V 10A for the 4 NEMA 23s, and 24V 20A for the board and everything else

  7. Toggling interpolation on/off

  8. Reducing microsteps down to 1/8 to lower precision and reduce command load

  9. Reducing slice resolution down to 0.049 mm (4x worse than the Elegoo Neptune 4 Max)

  10. Reducing speeds down to 150 mm/s travel and 30 mm/s print speed

Firmware:

CB1 image = V3.1.0, clean, unmodified. Main config files untouched — only Klipper settings adjusted.

Issue:

Errors can appear as early as 30 minutes in, or as late as 2–5 hours into a print.

2

As far as I can tell, it seems that the reason for TTC in this case is retries:

Stats 2618.0: gcodein=0 mcu: mcu_awake=0.038 mcu_task_avg=0.000051 mcu_task_stddev=0.000055 bytes_write=6085098 bytes_read=1567946 bytes_retransmit=9
Stats 2619.0: gcodein=0 mcu: mcu_awake=0.038 mcu_task_avg=0.000051 mcu_task_stddev=0.000055 bytes_write=6085949 bytes_read=1568360 bytes_retransmit=372
Stats 2620.0: gcodein=0 mcu: mcu_awake=0.038 mcu_task_avg=0.000051 mcu_task_stddev=0.000055 bytes_write=6086784 bytes_read=1568929 bytes_retransmit=751 bytes_invalid=0 send_seq=142515 receive_seq=142515 retransmit_seq=142496 srtt=0.000 rttvar=0.000 rto=0.025 ready_bytes=53 upcoming_bytes=0 freq=64001032
Stats 6154.6: gcodein=0 mcu: mcu_awake=0.037 mcu_task_avg=0.000054 mcu_task_stddev=0.000055 bytes_write=16390127 bytes_read=3874707 bytes_retransmit=751 bytes_invalid=0 send_seq=369158 receive_seq=369158 retransmit_seq=142496 srtt=0.001 rttvar=0.000 rto=0.025 ready_bytes=18 upcoming_bytes=0 freq=64001043
MCU 'mcu' shutdown: Timer too close
clocksync state: mcu_freq=64000000 last_clock=394170032704 clock_est=(6125.645 392331432355 64001043.050) min_half_rtt=0.000088 min_rtt_time=4910.767 time_avg=6125.645(865.575) clock_avg=392331432355.581(55397697863.315) pred_variance=2994828.759
Dumping serial stats: bytes_write=16391088 bytes_read=3875574 bytes_retransmit=3095 bytes_invalid=0 send_seq=369176 receive_seq=369176 retransmit_seq=369171 srtt=0.001 rttvar=0.000 rto=0.025 ready_bytes=31

Otherwise, it seems that the main MCU is connected by USB: RESERVE_PINS_USB=PA11,PA12
and, as far as I understand, CB1 is installed on the Manta board.

So, I’m not sure what can cause such connection issues.

Regards,
-Timofey

Also, side note: it is not related to the issue, but it is weird that you have an older Klipper host than the MCU firmware:

Loaded MCU 'mcu' 139 commands v0.13.0-734-gfe4eb8650
Git version: 'v0.13.0-636-g293e1e9d'
3

That’s the thing – I have no idea what’s causing it either.

As for the firmware – I’ve also swapped the SD card for the CB1 and tried different images, but the same error keeps appearing. I’m starting to wonder if I’m missing something in the Klipper config.

4

Doubt so, it is either a low-level software issue (kernel) or a hardware issue.

There are no magic config settings that affect this.

So, I guess the only actionable thing you can do is try using UART or something. Basically, attempt to ditch USB, as it seems to have issues in your setup.

5

If you have a camera attached, or any other usb device, remove it and see if that changes anything.

The usb hub chip built into the manta board that splits between the mcu and the onboard usb socket it not the best and i have seen it cause this issue before.

6

I tried it, and the error is still the same, but now I have new data: if I set the speed to 50%, it prints.

Could it be an error in the config, where something causes it to calculate something complex?

7

I got rid of all TTC by just upgrading from CB1 to CM4… it was ok 2 years ago when I bought it, but lately it errored out doing imput shaper(trough Klipper Shake&Tune plugin ).

8

Super long shot…
Check your Moonraker log for unusual activity. I had a seemingly random TTC issue on one of my machines. Turns out I had a leftover service from an uninstalled package (HelixScreen) that would flood Moonraker with connection attempts.

9

Successful

the problem was the manta m8p v1.1 board

If you connect a terminal to VUSB and power it via a Type‑C connector, the problem disappears.