It turned out the runout is caused by the ramming sequence in the G-code, which is interpreted before the actual MMU command gets into processing.
The frequency of these errors/runouts differed one machine from another and was really hard to track down.
Therefore - once the MMU is enabled (active and running) we'll disable filament runout handling caused by the printer's filament sensor.
There are still printers which suffer from spurious runouts for no reason.
For testing purposes, runout will be disabled until the real cause is found.
The following situation is now fixed, steps:
1. Load filament through LCD
2. Load to nozzle through LCD
3. Before filament reaches Fsensor => turn off the printer
4. Wait a few seconds and then turn the printer on again
5. Start a print through the SD card
6. Printer will unload and then re-load the filament
Previously in step 6, the printer would start a print without any
filament loaded.
- Dependent on current MMU error being processed
- Includes rename of LEFT_BUTTON_CHOICE -> LCD_LEFT_BUTTON_CHOICE (and right, middle as well) to avoid confusion with MMU buttons' ordering
To render the thermometer and degree symbols
the CGRAM table will instead swap out
the feedrate and clock symbol when
lcd_set_custom_characters_nextpage()
is called.
- We need to take into account additional 35mm move by the MMU.
- While the fsensor is not triggered, reduce the extruder movement each step from 20mm to 5mm. This should help reduce the variance in the filaments position before we run the final load to nozzle sequence.
- Fixed a bug where the feed rate was 60x higher than intended
- Adjusted the sequence via trial and error. The sequence where the filament has not exited the nozzle is 35mm shorter than the previous implementation
- I reduced the feed rate slightly (by 1mm/s) on the fast load sequence
- combined lcd_show_three_choices_prompt_P and lcd_show_two_choices_prompt_P into one function called lcd_show_choices_prompt_P
- Removed third_col parameter
- Added an enum to make the button selection result more readable
- Rename lcd_show_multiscreen_message_two_choices_and_wait_P to lcd_show_multiscreen_message_with_choices_and_wait_P
- Move MMU error screen from ReportError() to ReportErrorHook()
- Fix the menu selection to work for menus with either two choices or three choices
- The buttons shown on the MMU error menu are now determined by the definition of btnOperation[]
getFilamentPresent returns the state of the IO pin of the sensor.
So we should use AT_FSENSOR instead of IN_NOZZLE
since the fsensor does not know if the filament is actually in the nozzle.
Such that:
MMU2:E=32766ErrorTitleTextDescription
Becomes:
MMU2:E=32766 ErrorTitle TextDescription
Also simplified the process of combining ErrorTitle and TextDescription
into the msg buffer by using snprintf. This is saver since we only
use dstSize in one place instead of two.
This could cause the printer to crash. The Serial Stream would show
a garbled string.
Also included in this commit:
* The buffer size is increased from 64 bytes to 192 bytes. We need to
take into account the length of the ErrorTitle and ErrorDescription.
* Fix overwrite in ErrorCode message buffer.
When snprintf() if called we need to read the return value to see how
many bytes were written. Then when we call strncpy_P() through
TranslateProgress(), we need to tell the code to start writing
at byte 'len', or &msg[len]. Also we need to update the byte size
which strncpy_P() is allowed to write (64 - len).
This could cause the printer to crash. The Serial Stream would show
a garbled string and the same corruption would appear on the Status
Screen's status line.
makes the code a bit shorter as well, because the compiler can be sure
that the variables are not accessed from the outside of the
fsensor.cpp compilation unit.
Update fw-build.sh
- Output used space of each translation (easier to troubleshoot)
- Read config.h max size per translation
- output variant .map files (easier to troubleshoot and finding missing/unused messages)
This message is never actually shown on the LCD (it's immediately
overwritten by other content). It also include g-code in a translatable
string, which is not a good idea. It's also otherwise identical to
MSG_V2_CALIBRATION, which we should have used.
Just remove it and save some space in the process.
Having the original PRINTER_ACTIVE macro copied at multiple spots doesn't make sense.
Refactoring it into a non-inline function saved ~400 bytes of code.
It should be safe in terms of performance, all occurrences are at non-time critical spots.
Add an additional parameter to control the position of second choice
prompt position (while defaulting to the old).
This allows Yes/No prompts to be equally spaced.
Modify lcd_show_multiscreen_message_two_choices_and_wait_P to also
handle single-screen or empty (no-clear) prompts, making other functions
redundant. Saves 76 bytes.
Change existing functions to simply call
lcd_show_multiscreen_message_two_choices_and_wait_P with the correct
arguments.
This changes the prompt of existing Yes/No messages: the previous prompt
would use the last two lines of the LCD, while the new prompt is using
just the last line of the LCD instead.
Translation do not require updates, since the Yes/No translation was
already the same in both implementations.
Allow running the model checking during autotuning, with the only
exception being the parameter estimation stage where we alter the same
value which is used by the checker (done to conserve memory/code).
With previous changes the model checker will disable/enable itself when
passing through an unavailable R vector entry, allowing to start the
calibration by checking only the stages where the part fan is disabled.
The C/R0 values should be stable enough to provide a fail-safe mechanism
for printers of the same variant right from the factory.
Instead of assuming the model state is always valid, allow NAN values to
pass-through the various check/estimation stages.
This allows running the model checker with incomplete parameters (for
example, missing entries in the R vector) and resume automatically.
flash: -22
RAM: 0
menu_lcd_lcdupdate_func() takes care of starting lcd_timeoutToStatus if the knob is clicked.
When the knob is clicked we only want to back out of the menu.
This changes makes the menus slightly more conistant behind the scenes.
flash: -8
RAM: 0
When the menu is entered and left the function
menu_lcd_lcdupdate_func() takes care of resetting the timer.
Currently the firmware will reset the timer twice when the lcd_babystep_z menu is entered. This commit fixes that.
Split the planner sync code out of planner_abort_hard() so that we can
independently resync the planner position from the counters.
This is needed in xyzcal as we directly modify the stepper counters
(bypassing both planner and stepper).
Call this new function instead of planner_abort_hard() when leaving, so
that motion can resume in the middle of the gcode_M45 instruction.
We already disable the heaters upon entering, and the new temperature
isr doesn't perform any direct movement until we return to the main
loop.
This allows us to remove direct control of the soft_pwm interrupt from
the header, which is dangerous.
printer_smodel_check was incorrectly substituting the final " with a
null in the command to simplify the model string comparison, but in
doing so was also corrupting the next pop from the cmdqueue.
We can modify the current strchr_pointer as long as we *don't* change
it's length. This can cause an incorrect extra read from the queue,
resulting in the last command to be completely ignored.
cmdqueue will run commands when EOF is reached without returning to the
main loop, which is already incorrect.
However, since it needs to ensure the queue is empty, an st_synchronize
call can result in a re-entrant call to get_command, which will
reprocess EOF again. Even if we removed st_synchronize, another command
could be picked by an unsuspecting manage_inactivity() somewhere else.
Short-circuit EOF processing by closing the file early and checking for
the file state early in get_command.
This should fix#3549
Call this variable menu_block_mask instead. We don't need to know the
exact reason of why we're locking the menu.
We will be able to reuse this to prevent menu entry during more
activities in a cleaner way than testing for each condition as it's
currently done for both menu entry and longpress.
This allows us to use "M310 B0 W0.01" as a way to report the current
error continuosly on the serial without 1) more code and 2) without
preventing regular usage.
Use OCTOPRINT_ASK_* for the present form of actions. In these cases the
host will perform the pausing manouvers for us.
Use OCTOPRINT_* instead for the past tense forms when we are in charge.
Also always emit the action, whether we are or not sd-printing. This is
due to the new Stopped handling behaving correctly in either case.
If the user accepted to resume, do not wait for bed temperature when
recovering. This only adds a pointless delay when recovering a short
pause and doesn't really improve the recovery quality after a long
pause.
Do not allow resuming until all thermal and fan errors are clear.
Call the appropriate resume function when resuming a printing depending
on the saved_print state (is saved_print is available, then we always
need to resume from the saved state even when printing via usb).
Clear the Stopped state when resuming, so that commands can be accepted
again.
Ensure we never fall into the boundary values provided by the min/max
limits.
Save/restore the initial guess value, so that a convergence failure
restores the initial model state.
Force processing of the pause and stop "parking" commands _after_ the
main loop completes.
This was/is currently done in lcd_commands, which is a poor place to
continue processing, since it can be called already within an aborted
command. This requires checking for planner_aborted before any action
can be performed.
After calling planner_abort_hard() no motion command can be
scheduled until we return to the main loop since the call can
potentially be scheduled inside a nested process_command call.
Despite previous fixes, bugs keep creeping in due to nested calls not
being obvious to detect at all.
Stop allowing motion _completely_ for the entire processing loop by
default. That is, instead of aborting the current plan_buffer_line call,
abort the entire command until we can actually schedule motion safely
again.
This benefits handling of pretty much all g/m-codes, since this flag
(now "planner_aborted" for clarity) becomes a general "command aborted"
call.
This also now ensures that the flag prevents _any_ new block (including
blocks partially planned while servicing an interrupt) are scheduled
after planner_abort_hard is called.
There are only two exceptions where it's safe to resume in this context:
- Within uvlo_, where we never return to the main processing loop
- When we're intentionally scheduling a new process_command loop for a
MK3 filament recheck (which is *bad*)
Handle those two cases as exceptions.
Do not inhibit motion when Stopped is set.
We actually do need to move to move away the extruder from the bed, and
setting Stopped breaks it without adding any sort of security (M*
commands, such as M600 could still perform moves and still pass
through, while M104 would still set heaters).
During a hard error the internal queue is cleared (and sd file closed,
if any), so no new "unforeseen" command can be read.
Handle "Stopped" instead as a flag to inhibit serial processing and
automatically switch to "paused for user" state. While in this state
simply drop any input without incrementing the processed gcode line
number, behaving as-if the last command was still being processed.
This allows "Stopped" to correctly handle a printer-initiated paused
state and recover as expected by requesting a resend when resuming.
D.R.racer
Guðni Már Gilbert
VintagePC
Alex Voinea
3d-gussner
Yuri D'Elia
DRracer