Crash (Hard Reset) when issuing G2/G3 command #3729

manianac · 2016-05-11T00:27:59Z

I am using a RUMBA board with a 300x300x300 build area. I'm running RCBugFix commit 735f1f5.

I've found when running this code, I lose all comms with the board, and the graphical LCD shows the splash screen (full reset).

G28
G1 X130 Y130 Z5 F9000
G2 X130 Y130 I50 J50

Still experimenting to see if its only large moves, or any arc moves causing this

[edit] Found pretty much any move with I + J over ~50 crashes the board.

The text was updated successfully, but these errors were encountered:

jbrazio · 2016-05-11T01:21:12Z

Could you please attach a serial log ?

fbarcenas · 2016-05-11T06:40:34Z

What slicer you use?

thinkyhead · 2016-05-11T22:12:26Z

Also…

What compiler (Arduino version) are you using?
Are you using a DELTA or a cartesian robot?
Do you have software endstops enabled?

manianac · 2016-05-12T01:49:03Z

Here is the serial log:

Send: G28
Recv: X:0.00 Y:0.00 Z:0.00 E:23.79 Count X: 0 Y:0 Z:6539
Send: G1 Z5
Recv: ok
Send: G1 X130 Y130 F9000
Recv: ok
Send: M105
Recv: ok T:29.5 /0.0 B:29.5 /0.0 B@:0 @:0
Send: G2 X130 Y130 I50 J50
Recv: start
Recv: echo:Marlin 1.1.0-RCBugFix
Recv: echo: Last Updated: 2016-04-27 12:00 | Author: (Robert Kirk, custom config)
Recv: Compiled: May 10 2016
Recv: echo: Free Memory: 3027  PlannerBufferBytes: 1232
Recv: echo:V23 stored settings retrieved (453 bytes)
Recv: echo:Steps per unit:
Recv: echo:  M92 X80.00 Y80.00 Z4000.00 E900.00
Recv: echo:Maximum feedrates (mm/s):
Recv: echo:  M203 X250.00 Y250.00 Z2.00 E15.00
Recv: echo:Maximum Acceleration (mm/s2):
Recv: echo:  M201 X750 Y750 Z500 E500
Recv: echo:Accelerations: P=printing, R=retract and T=travel
Recv: echo:  M204 P750.00 R500.00 T750.00
Recv: echo:Advanced variables: S=Min feedrate (mm/s), T=Min travel feedrate (mm/s), B=minimum segment time (ms), X=maximum XY jerk (mm/s),  Z=maximum Z jerk (mm/s),  E=maximum E jerk (mm/s)
Recv: echo:  M205 S0.00 T0.00 B20000 X10.00 Z0.30 E10.00
Recv: echo:Home offset (mm):
Recv: echo:  M206 X0.00 Y0.00 Z0.00
Recv: Mesh bed leveling:
Recv: echo:  M420 S1 X4 Y4
Recv: echo:  M421 X20.00 Y20.00 Z1.80
Recv: echo:  M421 X95.00 Y20.00 Z1.98
Recv: echo:  M421 X170.00 Y20.00 Z1.94
Recv: echo:  M421 X245.00 Y20.00 Z1.76
Recv: echo:  M421 X20.00 Y96.00 Z1.85
Recv: echo:  M421 X95.00 Y96.00 Z1.97
Recv: echo:  M421 X170.00 Y96.00 Z1.97
Recv: echo:  M421 X245.00 Y96.00 Z1.93
Recv: echo:  M421 X20.00 Y172.00 Z1.94
Recv: echo:  M421 X95.00 Y172.00 Z1.97
Recv: echo:  M421 X170.00 Y172.00 Z1.97
Recv: echo:  M421 X245.00 Y172.00 Z1.86
Recv: echo:  M421 X20.00 Y248.00 Z1.85
Recv: echo:  M421 X95.00 Y248.00 Z1.85
Recv: echo:  M421 X170.00 Y248.00 Z1.87
Recv: echo:  M421 X245.00 Y248.00 Z1.66
Recv: echo:Material heatup parameters:
Recv: echo:  M145 S0 H190 B60 F0
Recv: echo:  M145 S1 H220 B110 F0
Recv: echo:PID settings:
Recv: echo:  M301 P35.65 I3.59 D88.49 C1.00 L20
Recv: echo:Retract: S=Length (mm) F:Speed (mm/m) Z: ZLift (mm)
Recv: echo:  M207 S2.00 F900.00 Z0.00
Recv: echo:Recover: S=Extra length (mm) F:Speed (mm/m)
Recv: echo:  M208 S0.00 F660.00
Recv: echo:Auto-Retract: S=0 to disable, 1 to interpret extrude-only moves as retracts or recoveries
Recv: echo:  M209 S0
Recv: echo:Filament settings:
Recv: echo:  M200 D2.90
Recv: echo:SD card ok
Recv: echo:SD card ok

I am unable to send anything until I disconnect and reconnect (OctoPrint).
Compiler: Arduino IDE 1.6.8
Cartesian Robot, Makerfarm Prusa i3v.
Software endstops are enabled, I will attach my Configuration.h

A slicer did not generate this, I have gcode I run to test new config options or features.
Configuration.h.txt

thinkyhead · 2016-05-12T02:12:12Z

Do you get the same result with:

G1 X130 Y130 F9000
G2 I50 J50

It's the same idea, and should make a complete circle with a radius just over 70mm, centered around the point 180, 180.

thinkyhead · 2016-05-12T02:20:08Z

Let's debug, shall we? Add the following lines to the plan_arc function in Marlin_main.cpp:

    SERIAL_ECHOPAIR("Segment ", segment);
    SERIAL_ECHOPAIR(" at ", arc_target[X_AXIS]);
    SERIAL_ECHOPAIR(",", arc_target[Y_AXIS]);
    SERIAL_EOL;

…immediately after the lines…

    // Update arc_target location
    arc_target[X_AXIS] = center_X + r_X;
    arc_target[Y_AXIS] = center_Y + r_Y;
    arc_target[Z_AXIS] += linear_per_segment;
    arc_target[E_AXIS] += extruder_per_segment;

…and re-flash and test. This will tell us how far it gets before it blows up.

manianac · 2016-05-12T10:23:56Z

70mm arc @ 180 did crash it, I will update the firmware tonight (long work hours).

manianac · 2016-05-12T23:22:01Z

Ok, ran the updated firmware (segment wasn't a defined variable, ended up using "i" from the loop).

Send: G28
Recv: X:0.00 Y:0.00 Z:0.00 E:23.79 Count X: 0 Y:0 Z:6539
Send: G1 Z5
Recv: ok
Send: G1 X130 Y130 F9000
Recv: ok
Send: G2 I50 J50
Recv: Segment 1 at 129.30,130.71
Recv: Segment 2 at 128.61,131.44
Recv: Segment 3 at 127.92,132.17
Recv: Segment 4 at 127.25,132.91
Recv: Segment 5 at 126.59,133.66
Recv: Segment 6 at 125.94,134.42
Recv: Segment 7 at 125.30,135.19
Recv: Segment 8 at 124.67,135.97
Recv: Segment 9 at 124.05,136.76
Recv: Segment 10 at 123.45,137.55
Recv: Segment 11 at 122.85,138.36
Recv: Segment 12 at 122.27,139.17
Recv: Segment 13 at 121.70,139.99
Recv: Segment 14 at 121.14,140.82
Recv: Segment 15 at 120.59,141.66
..
[lines removed for length of post]
..
Recv: Segment 363 at 204.97,113.84
Recv: Segment 364 at 204.03,113.50
Recv: Segment 365 at 203.08,113.16
Recv: Segment 366 at 202.14,112.84
Recv: Segment 367 at 201.18,112.54
Recv: Segment 368 at 200.23,112.24
Recv: start
Recv: echo:Marlin 1.1.0-RCBugFix

I do wonder though, am I the only one duplicating this issue? If so, I'll focus on it on my own board, I just wasn't sure if anyone else was experiencing this.

manianac · 2016-05-13T01:15:30Z

Alright, life would've been easier if the 2560 bootloader didn't zero out MCUSR, but I'm 90% positive its a watchdog reset. Disabling watchdog temporarily solves the issue, and WATCHDOG_RESET_MANUAL causes a bootloop that otherwise would've pointed out the issue.

You may want to consider another location for wdt_reset other than Temperature.cpp

Blue-Marlin · 2016-05-13T08:28:04Z

Uiiii.
That means your temperatures and heaters are not updated for that long.
Calculating the next arc sub move to plan, takes longer than it's execution in the stepper-ISR. For that, there is always a free place in the planner buffer and

  // If the buffer is full: good! That means we are well ahead of the robot.
  // Rest here until there is room in the buffer.
  while (block_buffer_tail == next_buffer_head) idle();

idle() is not called, what should call the heater manager.

As a workaround try to increase MM_PER_ARC_SEGMENT and/or N_ARC_CORRECTION.

One of the possible real solutions could be to call at least manage_heater() better idle() every ~200ms in plan_arc. Something similar is probably needed for the BEZIER_CURVE_SUPPORT (G5)

AnHardt · 2016-05-13T11:54:37Z

Someting like

    float feed_rate = feedrate * feedrate_multiplier / 60 / 100.0;

+   millis_t previous_ms = millis();

    for (i = 1; i < segments; i++) { // Iterate (segments-1) times

+     millis_t now = millis();
+     if (now - previous_ms > 200) {
+       idle();
+       previous_ms = now;
+     }

      if (++count < N_ARC_CORRECTION) {
        // Apply vector rotation matrix to previous r_X / 1

AnHardt · 2016-05-13T17:19:13Z

I can reproduce the problem with setting MM_PER_ARC_SEGMENT smaller and smaller. First it begins to stutter (~0.5). When further decreasing i get watchdog resets (<0.33). Also depends on F (5000) - faster -> erlyer stuttering and resetting..

AnHardt · 2016-05-13T18:21:49Z

#3751 prevents from the watchdog reset but increases the stuttering a bit. Using thermalManager.manage_heater(); instead if idle() would have a bit less stuttering but all the jobs in manage_inactivity() should be done from time to time.

thinkyhead · 2016-05-15T00:33:33Z

How about something like this…

uint8_t idle_counter = 0;
millis_t next_ping_ms = millis() + 200UL;

for (i = 1; i < segments; i++) { // Iterate (segments-1) times

  millis_t now = millis();
  if (ELAPSED(now, next_ping_ms)) {
    next_ping_ms = now + 200UL;
    if (idle_counter++ & 0x03)
      thermalManager.manage_heater(); // ...and reset watchdog timer
    else
      idle();
  }

. . .

jbrazio · 2016-05-24T12:13:17Z

@AnHardt @thinkyhead what is the status of this ticket, should we consider it a bug that needs a patch or can it be closed ? Thanks.

AnHardt · 2016-05-24T16:25:45Z

Fixed by #3751 and follow ups.

github-actions · 2022-04-08T01:27:55Z

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.

jbrazio added the Bug: Potential ? label May 11, 2016

jbrazio added this to the 1.1.0 milestone May 11, 2016

AnHardt mentioned this issue May 13, 2016

Idle during long arcs #3751

Merged

jbrazio added Bug: Confirmed ! and removed Bug: Potential ? labels May 24, 2016

jbrazio closed this as completed May 24, 2016

github-actions bot locked and limited conversation to collaborators Apr 8, 2022

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Crash (Hard Reset) when issuing G2/G3 command #3729

Crash (Hard Reset) when issuing G2/G3 command #3729

manianac commented May 11, 2016 •

edited by thinkyhead

Loading

jbrazio commented May 11, 2016

fbarcenas commented May 11, 2016

thinkyhead commented May 11, 2016 •

edited

Loading

manianac commented May 12, 2016 •

edited

Loading

thinkyhead commented May 12, 2016

thinkyhead commented May 12, 2016 •

edited

Loading

manianac commented May 12, 2016

manianac commented May 12, 2016 •

edited

Loading

manianac commented May 13, 2016

Blue-Marlin commented May 13, 2016 •

edited

Loading

AnHardt commented May 13, 2016

AnHardt commented May 13, 2016 •

edited

Loading

AnHardt commented May 13, 2016 •

edited

Loading

thinkyhead commented May 15, 2016 •

edited

Loading

jbrazio commented May 24, 2016

AnHardt commented May 24, 2016

github-actions bot commented Apr 8, 2022

Crash (Hard Reset) when issuing G2/G3 command #3729

Crash (Hard Reset) when issuing G2/G3 command #3729

Comments

manianac commented May 11, 2016 • edited by thinkyhead Loading

jbrazio commented May 11, 2016

fbarcenas commented May 11, 2016

thinkyhead commented May 11, 2016 • edited Loading

manianac commented May 12, 2016 • edited Loading

thinkyhead commented May 12, 2016

thinkyhead commented May 12, 2016 • edited Loading

manianac commented May 12, 2016

manianac commented May 12, 2016 • edited Loading

manianac commented May 13, 2016

Blue-Marlin commented May 13, 2016 • edited Loading

AnHardt commented May 13, 2016

AnHardt commented May 13, 2016 • edited Loading

AnHardt commented May 13, 2016 • edited Loading

thinkyhead commented May 15, 2016 • edited Loading

jbrazio commented May 24, 2016

AnHardt commented May 24, 2016

github-actions bot commented Apr 8, 2022

manianac commented May 11, 2016 •

edited by thinkyhead

Loading

thinkyhead commented May 11, 2016 •

edited

Loading

manianac commented May 12, 2016 •

edited

Loading

thinkyhead commented May 12, 2016 •

edited

Loading

manianac commented May 12, 2016 •

edited

Loading

Blue-Marlin commented May 13, 2016 •

edited

Loading

AnHardt commented May 13, 2016 •

edited

Loading

AnHardt commented May 13, 2016 •

edited

Loading

thinkyhead commented May 15, 2016 •

edited

Loading