Load all items from the look-ahead queue into the trapezoid velocity
queue, and then flush that queue by time. This prevents the host cpu
from being starved on very long moves (which may require a large
number of steps to be generated). It also improves the overall
performance.
With the batch flushing logic in place, it is no longer necessary to
split homing moves up.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Now that all callers use the trapq system to queue moves, it is no
longer necessary to individually allocate and fill a 'struct move'.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Eliminate the old gen_steps() call now that all callers have been
converted to the newer trapq based generate_steps().
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Now that all kinematics use the trapq system, there is no need to call
into the kinematics on each g-code move.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Support associating a stepper_kinematics with a trapq. Support
generating steps from a time range on the given trapq.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add support for building a list of moves in the trapq.c code. Update
the toolhead code so that moves generated from the look-ahead code are
added to that list.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
If the kinematic position calculation function had an error in it (if
it produces sudden position changes) then it could lead to an infinite
loop in itersolve_gen_steps().
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Break up the connect phase into two phases - mcu_identify and connect.
The first phase will contact all the micro-controllers and obtain the
"identify" data dictionaries for them. Once all the micro-controllers
have been contacted, then they can be configured. This fixes an issue
with FIRMWARE_RESTART not working correctly on multi-mcu setups that
use restart_method=command.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Wait in the M25 command handler until the work_handler() background
timer fully exits. This ensures that subsequent g-code commands (such
as M27 or M24) will correctly report a paused state.
Reported by @foosel.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The initial clock measurement wasn't properly implementing a small
delay between clock queries. Fix the pause() call so there is a small
delay between queries.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Some print start scripts may extrude at a position slightly higher
than the first z layer height. Tweak the backwards z test to reduce
the chance of the tuning test ending prematurely.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
PR #1970 introduced a new parameter, controlling the default behavior of the safe_z_home-module. To avoid unexpected changes, the default of the move_to_previous is now defaulted to False.
Signed-off-by: Nils Friedchen <Nils.Friedchen@googlemail.com>
If '[stepper_x]' and/or '[stepper_y]' have a 'position_min' that is
non-zero, and '[safe_z_home] home_xy_position' is '0,0'; then the 'G28'
command will _not_ move to '0,0'; but stay at 'position_min' during the
Z endstop test.
This fix corrects this issue.
Signed-off-by: Jason S. McMullan <jason.mcmullan@gmail.com>
Below off_below the fan will be turned off. When configured correctly this
can prevent stalling, which may lead to overheating and failed fans. To
simplify calibration and use of this setting, off_below is applied in the
input duty cycle domain, prior to any scaling due to e.g. max_power.
Signed-off-by: Nicholas Seckar <nseckar@gmail.com>
When reducing the maximum speed due to the max_accel_to_decel setting,
move velocity limits must still be propagated. Otherwise, the
trapezoid move planner may produce moves with velocity jumps.
Signed-off-by: Dmitry Butyugin <dmbutyugin@google.com>
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Once a hop is performed, it will only be re-issued if the z-axis has been
moved in the meantime. Usually it is only moved by a z-homing so doing so
will cause safe_z_home to do the hop on the next homing action.
When z-axis is homed, x and y positions are known. When setting this
boolean option, these are set back to their last positions.
Whenever a hop is specified it is re-issued after the Z axis has been
homed. This is especially necessary when a pressure-based probe is used.
Also, the module decides if a hop is necessary, based on either a known
Z position or a flag that is set whenever the motors are disabled.
Signed-off-by: Nils Friedchen <Nils.Friedchen@googlemail.com>
Previously, when a sercom pin specified the mcu (myname:PIN), it would result in a pin enumeration error.
Signed-off-by: John Jardine <john@gprime.net>
R Still missing, also might be somewhat dirty since code is converted
into G1 commands.
Signed-off-by: Aleksej Vasiljkovic <achmed21@gmail.com>
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
During a "drip feed" move, both the move's maximum speed and maximum
acceleration may be limited by the kinematics.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add a helper class for tracking gpio outputs that are synchronized to
bus updates on a particular command queue.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Rework the spi_config commands so that bus configuration and shutdown
message configuration is done separately from the main "config_spi"
command. This makes the spi configuration more flexible. It's now
possible to use software spi without a CS pin. It's now possible to
define multiple SPI messages to send on a shutdown event.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
use SET_GCODE_OFFSET's MOVE=1 parameter in the z offset menu.
Less confusing for people who think nothing happened.
Signed-off-by: John "Fess" Fessenden <fess@fess.org>
It's not necessary to fully build up the look-ahead queue prior to
starting a "drip move" homing operation. Instead, allow the
look-ahead queue to flush normally.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
If the homing move is exceptionally long, it could take a long time
for the host to process the look-ahead queue. Delay the print_time
calculation until moves are ready to be sent.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Rename the commands.
The main reason for this change is to force a micro-controller code
updated (due to important bugfix in 7fefe52d).
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The current z_hop support did not work correctly with common slicers
that change Z height between retract and unretract, and it did not
work correctly with slicers that issue a retract at the end of a print
and then at the beginning of the next print.
The z_hop support is therefore being removed until these issues can be
fixed.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Don't default to a Z location of 10, as that could cause damage if the
probe's z_offset is greater than 10. Instead, use the "retract
distance" method that is used for normal multi-sample probing.
Update the PROBE_ACCURACY command parameter names to use the same
parameter names as the PROBE command.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Now that homing is implemented via "drip moves", it is no longer
necessary for the stepcompress code to have special homing logic.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Now that homing is implemented via "drip moves", it is no longer
necessary to move the print_time backwards.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Now that homing is implemented via "drip moves", it is no longer
necessary to round the homing speed and it is no longer necessary to
add a delay for cpu processing time.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Rework the low-level implementation of homing movement. The existing
mechanism buffers all homing movement into the micro-controller prior
to starting the home. Replace with a system that buffers all movement
into the host look-ahead buffer and then "drip feed" those moves to
the micro-controllers. Then clear the host look-ahead buffer when all
endstops trigger.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Finish Moves needs be a hard requirement for the event gcodes, as it eliminates the possiblility of an event firing while the gcode is executing.
Signed-off-by: Eric Callahan <arksine.code@gmail.com>
It really helps to see a snapshot of available printer variables when building menus and macros. The list of variables always depends on what printer config you have.
Signed-off-by: Janar Sööt <janar.soot@gmail.com>
This moves skew_correction configuration out of printer.cfg. It is now dynamically set by the SET_SKEW gcode.
Signed-off-by: Eric Callahan <arksine.code@gmail.com>
Add support for enabling the stepper via the communication channel.
This improves support for boards with a shared enable line.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
A print_time of zero may translate to a negative clock on a secondary
micro-controller, which would cause an internal error. Change the
code to pass a real print_time or None if it is not needed.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Only the tmc2208 driver needs special read register translation logic.
Rework the code so a default implementation is available for all the
other drivers.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Move the serial command wrapper class from serialhdl.py to mcu.py.
This will allow that class to better support higher level
functionality.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Introduce a main send_with_response() helper function and use it
during identify and the setup of clocksync.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
An extra to execute gcode when a hardware button is pressed or released. Uses the jinja2 templating system.
Inspired by pull request #545 from Miguel Moitinho miguel@moitinho.net and pull request #1098 from Paulo Drugos paulodrugos@gmail.com
Signed-off-by: Alec Plumb <alec@etherwalker.com>
Add PROBE_SPEED, SAMPLES, SAMPLE_RETRACT_DIST, SAMPLES_TOLERANCE,
SAMPLES_TOLERANCE_RETRIES, and SAMPLES_RESULT parameters to the PROBE
command (and to commands that indirectly invoke a probe).
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Make sure to configure DRVCONF with SDOFF=0 prior to writing DRVCTRL
as that field controls access to that register.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Support retrying Z_TILT_ADJUST a configurable number of times
to a configurable tolerance both in the config or as parameters.
- By default keeps original behavior of no retries.
- Adds parameters RETRIES and RETRY_TOLERANCE to QUAD_GANTRY_LEVEL gcode.
- adds config options retries and retry_tolerance to uad_gantry_level]
- issues an error if we are getting worse intead of approaching tolerance
- issues an error if retries were requested but we did not reach the tolerance
in the specified number of retries
the minimum change should be a single z step for those
probing 2 points for 2 stepper motors and 3 for 3 stepper motors.
at one point it was suggested to use the amount of z adjustment instead of the
range of the probed points as a trigger for retry. I've chosen not to do this.
using z adustment in these cases means the minimum unit of change is related to
the angle created by the probed points and the distance to stepper motor and
can be more than a couple steps which is rather unintuitive.
for the case when someone is using more probed points than z steppers the
probed points range will have some fixed minimum value that can't be reduced
which is also unintuitive but that case should idealy be the rarer case, and
the user can learn to set a higher tolerance that matches their probing setup.
Signed-off-by: John "Fess" Fessenden <fess@fess.org>
Support retrying QUAD_GANTRY_LEVEL a configurable number of times
to a configurable tolerance both in the config or as parameters.
- By default keeps original behavior of no retries.
- Adds parameters RETRIES and RETRY_TOLERANCE to QUAD_GANTRY_LEVEL gcode.
- adds config options retries and retry_tolerance to `[quad_gantry_level]`
- issues an error if we are getting worse intead of approaching tolerance
- issues an error if retries were requested but we did not reach the tolerance
in the specified number of retries
Signed-off-by: John "Fess" Fessenden <fess@fess.org>
List all the configurable parameters in the tmc5160 example block and
use similar wording as the other tmc drivers.
Also, don't allow DEDGE to be configured as it would just break the
current code.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Introduce TMCMicrostepHelper and TMCStealthchopHelper helper code that
uses the mcu_tmc interface. Update the callers to use these new
helpers.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Perform the register initialization directly from TMCCommandHelper and
perform it during the "connect" phase. This unifies the register
setup for the TMC2130, TMC2208, TMC2660, and TMC5160.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Use a get_register() method instead of get_response(). This makes the
tmc2660 class more similar to other TMC drivers. It also enables
DUMP_TMC to dump all the readable content.
Also, use the tmc2130.get_config_microsteps() helper.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Reduce the delay in send_with_response() by introducing a reactor
mutex to wake the main thread when the response is obtained.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Allow the callers of process_batch() to directly inspect the gcode
mutex. Those callers can then directly invoke run_script().
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
If commands are slowly fed to the printer it could cause the number of
queued commands to grow without bound. Be sure to pause the incoming
command stream even if the lookahead queue is empty.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Now that EndstopError and gcode.error use the same underlying
Exception object, it is no longer necessary to translate an error.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Introduce a homing.CommandError and use that as the basis for both
gcode and EndstopError exceptions.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Factor out _calc_mean() and _calc_median() functions and call from
both run_probe() and cmd_PROBE_ACCURACY().
This also fixes a subtle error in the run_probe() median function - on
some kinematics the x and y position can change on a z move so the x
and y should be taken from the z probe values actually used.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
This changes the config file so that the configuration of
multi-samples is now done in the [probe] (and [bltouch]) section
instead of the various delta, bed_mesh, z_tilt, etc. config sections.
With this change, PROBE and PROBE_CALIBRATE commands now also utilize
multi-sampling.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Now that the extruder, display, heater_fan, output_pin, and servo
classes have get_status() methods, it is no longer necessary to
override them in the menu code.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add "gcode_position" named tuple to gcode.get_status(). Add a
"position" named tuple to toolhead.get_status().
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add action_respond_info(), action_respond_error(), and
action_emergency_stop() callables to the get_status() return
dictionary. This allows gcode macros to directly invoke these
actions.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
These action commands prevent Octoprint from sending extra gcodes that can interfere with Klipper's pause/resume control.
Signed-off-by: Eric Callahan <arksine.code@gmail.com>
The SET_GCODE_OFFSET command could cause unwanted behavior when an
offset is applied to the following g-code command. (In particular,
when the following command is an extrude only move - as in
issue #1289.)
Don't apply the offset immediately. Instead, add support for a MOVE=1
parameter which will schedule a move to apply the given offset.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Prior to the BLTouch v3, the recommended command times were 700us
(pin_down), 1200us (touch_mode), 1500us (pin_up), 1800us (self_test),
and 2200us (reset). However, the recommended Marlin timing (via servo
"angles") was 647.111, 1162.667, 1472, 1781.333, and 2193.778us.
As of the BLTouch v3, the recommended times are now 650, 1165, 1475,
1780, and 2190us. The v3 continues to recommended Marlin timings of
647.111, 1162.667, 1472, 1781.333, and 2193.778us.
Update Klipper to use the new BL-Touch v3 recommended timing. The new
timings are required for the BL-Touch v3 and they are closer to what
the Marlin firmware has historically used.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Improve the precision of soft pwm pulses by allowing the host to
directly specify the clock duration of the pulse. This improvement in
precision may be noticeable when controller servos (and bltouch).
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Mark methods internal to gcode.py with an underscore so that it is
easier to see which functions are purely helper functions.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
"The z_tilt_adjust was coded to require 3 points because the original
intent was for railcore-xl and voron. Since then, a number of users
have started using z_tilt_adjust for prusa i3 style printers with dual z
steppers. That's fine, but we just haven't gotten around to removing
the superfluous check in the code." -- Kevin O'Connor
remove that check.
Signed-off-by: John "Fess" Fessenden <fess@fess.org>
Factor out minimum required points check for ProbePointsHelper
in prep for setting different values for quad_gantry_level
and z_tilt_adjust
Signed-off-by: John "Fess" Fessenden <fess@fess.org>
"Feed Filament" menu item value was off-screen that made this item useless. So I changed its name to just "Feed". Users shall easily guess that this item is for filament feeding because this item is located in "Filament" menu.
Signed-off-by: Alexander Fadeev <alfsoft@gmail.com>
safety feature that prevents quad_gantry_level from trying to do a
correction that might break things - if your probe fires early for
whatever reason and tries to do an 8mm correction in one corner instead
we abort if a correction is over a configurable limit by default 4mm
configurable via `max_adjust` parameter in the config
Signed-off-by: John "Fess" Fessenden <fess@fess.org>
In particular trying to keep the output from a single quad_gantry_level
so that it fits in the octoprint terminal window without the need to
scroll
Signed-off-by: John "Fess" Fessenden <fess@fess.org>
Adding gcode command for setting the target temperature of temperature_fan.
Example gcode: SET_TEMPERATURE_FAN_TARGET temperature_fan=chamber target=35.0
Signed-off-by: Martin Nowroth <madnoth@gmail.com>
The parameter to the M221 command should be the ratio of the nominal to
measured filament area, rather than the ratio of the diameters. Since we
are taking the ratio, most of the area calculation cancels out.
Fixes#1535.
Signed-off-by: Len Trigg <lenbok@gmail.com>
Kevin O'Connor