The relative_reference_index will now refer to a coordinate that is
static and cannot be changed at runtime. If new mesh parameters
are specifed and the reference lies outside of the mesh then the
reference location will be probed.
Additionally this introduces a 'zero_reference_position' option which
accepts a specific X/Y coordinate. This may be used in place of the
relative_reference_index.
Signed-off-by: Eric Callahan <arksine.code@gmail.com>
When mesh parameters change the substituted indices need to be
updated, even when using the default point set. Always generate
new points to perform this update rather than cache
the "orig_points".
Signed-off-by: Eric Callahan <arksine.code@gmail.com>
It seems opening the canbus socket can sometimes throw an os.error (in
addition to can.CanError). Catch that error to avoid Klipper
reporting an internal error.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
If the stepper already executed some steps with the basic
stepper_kinematics, subsequent creation of a new one for input_shaper
would result in stepcompress errors unless position and time tracking
flags are also copied over.
Signed-off-by: Dmitry Butyugin <dmbutyugin@google.com>
Sample the BME680 every .8 seconds, matching that of the
BME280. This resolves an issue where `get_report_time_delta()`
returns a value that doesn't match the actual delta.
Signed-off-by: Eric Callahan <arksine.code@gmail.com>
In hybrid_corexy and hybrid_corexz, the update_limits function is (only) called by the DualCarriage implementation, whenever the carriage changes.
Unfortunately, these limits also keep track of homing status, when they're unhomed they are set to 1 to -1 (invalid range).
As a fix, if the limit was set to "unhomed", we keep it that way, and only update it with the new rail limits if it was already properly homed before.
Signed-off-by: Frans-willem Hardijzer <fw@hardijzer.nl>
Check for untracked files in the klippy/extras/ and klippy/kinematics/
directories and report those files in the log. This helps identify
code modifications when inspecting a log.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The control bytes for the AHT20 and AHT21 are identical to the AHT10, but I had not been able to test just yet to ensure the sensor code supported the other sensors. I've wrapped up testing and updated the comments/docs to reflect this additional support.
Signed-off-by: Scott Mudge <mail@scottmudge.com>
Fixes common MPU-9250 accelerometer issues for RPi Linux MCU and
improves reliability on all other architectures by adjusting the MPU-*
reading algorithm to only read whole Klipper messages' worth of
data from MPU-* and eliminating many unnecessary checks of the MPU FIFO
fill-level that consumed bus bandwidth needed for data transfer. Fixes
intermittent "Lost communication with MCU 'rpi'" due to "Timer too
close" and transposed / corrupted data due to FIFO overrun/data loss
when using MPU-* accelerometers. In addition FIFO overrun checks are
performed by testing the MPU interrupt flag vs. inferring from the FIFO
fill level.
Stress tested for 13hrs with two MPU-6500 attached to one I2C bus on RPi
and one on a PR2040
Stress tested for 23hrs with two MPU-6500 attached to one I2C bus on RPi
and one on a ATmega328P (Seeduino Nano)
Signed-off-by: Matthew Swabey <matthew@swabey.org>
In some condition ，“drvstrength 2” does not work. According to page 17, some mosfets need “drvstrength” set to 0 .
From datasheet rev1.15, the DRVSTRENGTH reset default is 0 . (instead of 2)
Signed-off-by: Albert Lin <vcore85@gmail.com>
The stalled_bytes counter refers to the number of bytes that are not
yet eligible for transmission. However, the naming leads to confusion
as it could be interpretted as an inability to transmit data. Rename
to upcoming_bytes to try to avoid that confusion.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
This commit adds support for the I2C-based AHT10 ambient temperature sensor. The sensor comes pre-calibrated and has a fairly simple command interface, so the interface class is in-turn relatively simple. So far this has been tested on a BTT GTR 1.0 MCU board (STM32F4), and seems to work fine.
Signed-off-by: Scott Mudge <mail@scottmudge.com>
Add an optional REGISTER parameter to DUMP_TMC so that the output is more filtered/cleaner for manual TMC calibration.
Signed-off-by: Alex Voinea <voinea.dragos.alexandru@gmail.com>
Ability to specify `VELOCITY` as a parameter for SET_TMC_FIELD. Useful for configuring at runtime the TSTEP based fields of the driver.
Signed-off-by: Alex Voinea <voinea.dragos.alexandru@gmail.com>
Implement a helper for calculating velocity based thresholds for tmc drivers. This code was written in such a way that it can be used with more fields than just tpwmthrs.
Signed-off-by: Alex Voinea <voinea.dragos.alexandru@gmail.com>
If tcoolthrs is configured (not the default 0), then do not force
the value of tcoolthrs=0xfffff during homing. This way, tcoolthrs
can be set to a custom value during homing.
`tpwmthrs` and `en_pwm_mode`/`en_spreadcycle` are now also
correctly restored if they were changed after startup.
Signed-off-by: Alex Voinea <voinea.dragos.alexandru@gmail.com>
`multistep_filt` was not configured at all on tmc2240 and tmc5160. As such, it would default to the value of 0 when the GCONF field was sent to the driver. However, that field has a default value of 1 in the driver, so klipper was overriding the defaults by accident.
This field improves the stability of velocity based thresholds (it's effectively TSTEP filtering) and it also improves the current regulation stability in stealthchop2 (according to the datasheet).
This field was already set to 1 correctly in the tmc2208/tmc2209 driver code.
Signed-off-by: Alex Voinea <voinea.dragos.alexandru@gmail.com>
Purely esthetic. Make all tmc driver init code similar to one another. The various fields were grouped based on the register which they affect or the function they perform.
Signed-off-by: Alex Voinea <voinea.dragos.alexandru@gmail.com>
Previously, the globalscaler was calculated during the config parsing and set to a fixed value. If the current was changed for any reason after the initialization, only IRUN and IHOLD would be changed. This however caused issues:
- If the new current was lower, then the resolution of the possible current values would be low since there are only 32 IRUN/IHOLD steps.
- If the new current was higher, it wouldn't actually work since IRUN and IHOLD are capped at 31, so it wouldn't be possible to increase the current without increasing globalscaler.
With this commit, the globalscaler is recalculated whenever necessary in order to ensure the correct range of IRUN/IHOLD is used.
Signed-off-by: Alex Voinea <voinea.dragos.alexandru@gmail.com>
Use mcu.lookup_command().get_command_tag() instead of
mcu.lookup_command_tag(). This improves error reporting on a protocol
mismatch. It also enables support for a msgtag that is negative.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Some users may not desire this behavior and others
may wish to initialize to a profile not named default.
Signed-off-by: Eric Callahan <arksine.code@gmail.com>
The pca9685_pwm set_pwm() code can be called from a background thread
(when the pin controls a heater). Propagate updates to the main
thread for enable tracking.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Sort using comparison on only first item in list to avoid greater/less
than comparison of stepper objects (which causes an error on Python3).
Reported by discourse user salbang.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
If note_kinematic_activity() has a time far in the future it could
result in a single flush attempt of that time range. Be sure to break
up that range into small chunks using the normal _update_move_time()
mechanism.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
It was possible a note_kinematic_activity() call could increase
last_kin_move_time, but _process_moves() could reset it. Fix by
making sure _process_moves() only ever increases last_kin_move_time.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Issue specific to Python 3, NoneType is being used to compare heartbeat time (actually caused by invalid condition operator), and by returning NoneType in timer functions.
Signed-off-by: Clifford Roche <clifford.roche@gmail.com>
Existing code uses very restrictive build volume constraint checking
with just narrow cone on top of fully cylinder for delta printers.
Code here implements more permissive and still safe build volume
constraint checks.
Signed-off-by: Jan Herich <jan.herich@gmail.com>
Make all the microstep lookup table registers configurable via the
config file. It also loads the default values.
TMC220x and TMC2660 do not support this feature.
Signed-off-by: Alex Voinea <voinea.dragos.alexandru@gmail.com>
Current Z_OFFSET_APPLY_ENDSTOP command only works for printers
with cartesian architecture which have separate Z axis defined.
But this functionality (persisting Z babystep value to endstops)
is exactly as convinient for delta printers, therefore this
PR implements it.
Signed-off-by: Jan Herich <jan.herich@gmail.com>
add G17, G18 and G19 commands to select arc planes
enhance G2/G3 to support arc moves in XY, XZ and YZ planes
Signed-off-by: Andrew Mirsky <andrew@mirsky.net>
After a G28 z-axis homing, there is a final z hop.
It was hoping to height z_hop as an absolute z height rather than relative.
If the z-axis home leaves the head at a z height higher than z_hop,
e.g. because you were using a probe to do z homing,
this meant that it was z hopping in the negative z direction, which
could result in crashing the toolhead.
Signed-off-by: Joshua Redstone <redstone@gmail.com>
This will hide features in the Klipper Display menu that isn't applicable
for the machine, therefore we can add extra alignment tools in the Control
menu as well.
Also conditionally displays Setup/Calibration options.
Signed-off-by: Nickolas Grigoriadis <nagrigoriadis@gmail.com>
According to ADXL345/ADXL343 datasheets, at 3.3V supply voltage,
which is most frequent in the various boards, the sensitivity of
X and Y axes changes to 265 LSB/g from 256 LSB/g at 2.5V.
Signed-off-by: Dmitry Butyugin <dmbutyugin@google.com>
If a maximum acceleration is changed between two consecutive moves,
this allows to correctly compute the junction velocity between them.
Signed-off-by: Dmitry Butyugin <dmbutyugin@google.com>
This adds a gcode command that can be used insight the slicer to pass the total layer count and current layer information.
Signed-off-by: Stefan Dej <meteyou@gmail.com>
M105 expects get_temp method. Module crashes when gcode_id parameter
is set and M105 called. Add methods as hotfix.
Signed-off-by: Robert Pazdzior <robertp@norbital.com>
Introduce a new handle_fault() method to handle processing of sensor
faults. This simplifies the calc_temp() methods.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Send the fault information explicitly in the query_thermocouple fault
field for max6675, max31855, and max31865.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
If a tmc driver can't find the stepper config section it reports a
"missing microsteps" error which can be confusing. Provide a more
explicit error message.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Added a SET_DIGIPOT command to the mcp4018 implementation.
Previously the mcp4018 was read only, and set at the time of
configuration. This allows you to change the value during a
print, which is needed for some older printers that need to
lower the stepper current during preheating.
Signed-off-by: Jake Bordens <jake@allaboutjake.com>
Also added status report for extruder_stepper objects with the
current pressure advance parameters values.
Signed-off-by: Dmitry Butyugin <dmbutyugin@google.com>
Removes a call to gcmd.respond_info which writes 'VARIABLE SAVED' to the
console every time the SAVE_VARIABLE command is called.
Signed-off-by: Andre LeBlanc <mapleleafmakers@gmail.com>
Raised from issue #5645, UTF-8 encoded symbols or other unexpected symbols on the UART raise an exception which causes klipper to stop. This change support UTF-8 encoded characters (from file names) as well as ignoring unexpected bytes.
Signed-off-by: Clifford Roche <clifford.roche@gmail.com>
An mcu device acting as an "mcu bridge" should only be reset after
other normal devices are reset - otherwise the bridge wont be able to
pass along the reset message to the downstream mcus.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Starting with nodeid 4 instead of nodeid 0 can reduce bitstuffing of
the id field in common configurations.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The `logging.basicConfig` does not reconfigure default logger.
This results in printing only warnings/errors to stderr
instead of also info (or debug).
This fixes the issue by setting log level on root logger.
Signed-off-by: Kamil Trzciński <ayufan@ayufan.eu>
Calling python functions can have high overhead. Inline the
twos_complement code in the _extract_samples() inner loop.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The MPU9250CommandHelper() class is nearly a duplicate of
ADXL345CommandHelper(). Rename ADXL345QueryHelper() to
AccelCommandHelper and remove user facing references to "adxl345". Use
it directly from mpu9250.py .
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The mpu9250.ClockSyncRegression() class is a duplicate of
adxl345.ClockSyncRegression(). Remove the duplicate copy and use the
code from adxl345.py .
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The MPU9250QueryHelper() class is a duplicate of ADXL345QueryHelper().
Rename ADXL345QueryHelper() to AccelQueryHelper() and use it directly
from mpu9250.py .
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Adds a save_config_pending_items to the status reported by
configfile reflecting the items and values that a future
SAVE_CONFIG would actually persist.
Signed-off-by: Kurt Haenen <kurt.haenen@gmail.com>
Just log an error on a fault.
Remove the host check for min/max temperature as the micro-controller
code already implements that check.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Allows a limited number of DS18B20 read failures
before stopping the printer. This is designed to
tolerate spurious read errors, while still stopping
for serious issues.
The printer will stop when the sensor
fails to report a value five times in a row.
Implementation works as follows:
The MCU reports any read errors using a new "fault"
parameter in its answers.
The Python code tracks the number of errors
and triggers the shutdown. This paves the way for
more sophisticated error handling in the future,
as well as an example for other sensors to follow.
Signed-off-by: Lorenzo Pfeifer <Lorenzo.Pfeifer+github@googlemail.com>
Adding Klipper functionality to support cancelling objects while
printing.
This module keeps track of motion in and out of objects and adjusts
movements as needed. It also tracks object status and provides that to
clients.
The Klipper module is relatively simple, and only provides one piece of
the workflow. Moonraker already supports processing uploaded files to
insert the required gcode markers for cancelling objects, using
https://github.com/kageurufu/cancelobject-preprocessor. This library is
also available as an executable for use in slicers, and pip
installations also include the script as a callable.
Mainsail has integrated support, and code changes for Fluidd are
available. Support in other interfaces is planned, and we've spoken to
several other developers about integrating frontend support in their
projects.
Signed-off-by: Troy Jacobson <troy.d.jacobson@gmail.com>
Co-authored-by: Franklyn Tackitt <git@frank.af>
Co-authored-by: Eric Callahan <arksine.code@gmail.com>
Add PROBE and CHIP to TEST_RESONANCES
Since it's possible to specify more than one chip
in TEST_RESONANCES the CHIP parameter has been
renamed to CHIPS
Signed-off-by: Mikkel Schmidt <mikkel.schmidt@gmail.com>
Make sure whitespace isn't in the pin description. Also make sure an
alias doesn't map to an invalid pin.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Minor indentation and variable name changes.
Don't remove linebreaks from message_protocol_error lines.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The MCU protocol error message often confuses users, especially after getting bombarded with the config format error line. This PR tries to improve it and has the following goals:
- Put the technical error at the end to prevent confusion and avoid the immediate jump to help channels instead of continuing to read
- Inform the user first what the type of error is (In this case: MCU Protocol error) and not at the end of the long error message
- Give the users a clear instruction what to do (pretty much unchanged), but possibly more approachable because the user is not confused by the technical error anymore
- Shows the version numbers of Klipper separate from all MCU version numbers so that it is more obvious that this is the version of Klipper
- Separates the version numbers between outdated MCUs and up-to-date MCUs, so that it becomes obvious if a MCU flash failed because the MCU doesn't disappear from the outdated list
- Previously, when viewing the errors in a web frontend, additional line breaks made it extremely hard to read the message, sometimes with as little as one word on one line. The error messages are now formatted to not have rogue line breaks, improving readability.
Signed-off-by: Felicia Alexa Hummel <felicia@drachenkatze.org>
If the final update in a series of SET_LED update commands didn't
alter a color than the transmit would not occur - even if prior
commands did make color changes. Refactor the update code and fix.
Reported by @mstansberry.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
This is most probably the result of a copy paste error. All vendor documentation for SI7021 (e.g. https://www.silabs.com/documents/public/data-sheets/Si7021-A20.pdf) points to 21 and it actually only works when adapted to 0x15 (=21).
Signed-off-by: Uli Wolf <github+klipper@wolf-u.li>
Eric Callahan