search

MarlinFirmware / Marlin

Marlin is an optimized firmware for RepRap 3D printers based on the Arduino platform. Many commercial 3D printers come with Marlin installed. Check with your vendor if you need source code for your specific machine.
https://marlinfw.org
GNU General Public License v3.0
16.25k stars 19.23k forks source link

[BUG] Support of TMC 2208 in standalone mode is broken #20688

Closed Nalorokk closed 3 years ago

Nalorokk commented 3 years ago

Bug Description

Extruder driver seems to stop during first layers of almost every print, sometimes it can work a little longer. It happens for me only on Marlin, and I never saw this problem on actual Klipper or Chitu firmware which clearly tells this is firmware bug. Seems like here is fix in Klipper firmware git. While you may say that it is problem of driver itself, but it works for me in Klipper with harsh pressure advance and bigger accelerations, I never saw it stopping like that in other firmwares, while it often shut downs even from extruder movement in menu in Marlin.

Configuration Files

config.zip

I tried different MINIMUM_STEPPER_PULSE, MINIMUM_STEPPER_POST_DIR_DELAY, MINIMUM_STEPPER_PRE_DIR_DELAY, LIN_ADVANCE, E0_DRIVER_TYPE. It seems like increasing delays may somewhat little increase driver stability, but it won't solve problem.

Seems like solution described in this quote from someone genius:

When using stealthChop, please always make sure, that you use velocity ramping. A hard stop will cut away motor back-EMF at once. As stealthChop is a voltage based chopper, it cannot respond to this at once, like spreadCycle. The result is an overcurrent, and the motor driver goes to overcurrent switch off, until it becomes disabled / enabled again.

To resolve the problem, please use at least a tiny velocity ramping, when hard stopping the motor, e.g. within a few / a few ten microsteps.

Steps to Reproduce

  1. Start any print on marlin
  2. Soon enough extrusion will stop

Expected behavior:

I would prefer if it would not stop during prints.

Actual behavior:

Extruder motor is powered off completly until printer reboot.

samtny commented 3 years ago

Was just coming here to open the same issue.

If I turn off LIN_ADVANCE in the config then Marlin works okay (at least so far), this seems to be supported by others I have been reading. "No Linear Advance with TMC2208".

However:

I had found this commit where the issue was "fixed" in Klipper: https://github.com/KevinOConnor/klipper/commit/c9cb462f90a68deb73cacb179f7bbcde2cb9aeed

But @Nalorokk 's link above may be a more up-to-date fix, not sure yet:

https://github.com/KevinOConnor/klipper/pull/2597

The new v4.2.7 "Silent" boards on Amazon have these TMC2208 / TMC2249 (drop-in replacement for 2208) in them. Seems many of us got a 3D printer for the holidays and want to compile our own Marlin, but running into this issue on these boards. ;)

Nalorokk commented 3 years ago

If I turn off LIN_ADVANCE in the config then Marlin works okay (at least so far), this seems to be supported by others I have been reading. "No Linear Advance with TMC2208".

Well, for me without LIN_ADVANCE it just works for little longer, but still unreliable even for test cube.

samtny commented 3 years ago

24 hours of prints in and extruder continuing to work for me.

I know if I turn LIN_ADVANCE on it will freeze up immediately.

I also have the S CURVE stuff turned off, junction deviation off, etc.

@Nalorokk Out of curiousity, what are the voltages you have coming out of your stepper drivers (the 2208's)? Mine is at about 1.1V - I thought this was rather high compared to "stock" for my setup (Ender 3, 4020 motors), but I dunno if these TMC standalones are supposed to be set higher than other drivers.

devin122 commented 3 years ago

FWIW, I'm having the same issues with with my 2208s in non-standalone. Some days it works fine, some days it seems to consistently stop extruding during the first few layers. I ended up tracking it down to the drivers built in short-circuit protection, which fits well with the Klipper explanation given

samtny commented 3 years ago

The solution for Klipper looks ... pretty involved to me, unfortunately I don't know enough about how Marlin / stepper logic works to even begin trying to port it.

Two commits from @KevinOConnor seem to be involved in the fix:

The second appears to be an improvement / fix against the first.

It seems he is putting the computed steps into a queue, evaluating the queue for rapid "step + change direction + step" instructions, then pausing the queue at the direction change and inserting a delay before the next step, so as not to overwhelm / trigger voltage overage on the steppers.

ellensp commented 3 years ago

that system is completely incompatible with marlin at this time. Re posting it everywhere is not helping anyone

sjasonsmith commented 3 years ago

Duplicate of #16076

github-actions[bot] commented 3 years ago

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.