forked from danorder/Flsun-super-racer-Full-klipper-config-
-
Notifications
You must be signed in to change notification settings - Fork 0
/
delta calibrate instruction-zoffset
48 lines (37 loc) · 5.42 KB
/
delta calibrate instruction-zoffset
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
(page under construction)
SUPER IMPORTANT
#1 check for probe bias due to effector tilt otherwise, visable bed deviations eg wear and tear. This will make delta calibrate unusable or usable only with larger layer heights
eg 0.2-0.15ish. usually running bed mesh will also display a excessive low or high spots. The reason for this is delta calibrate assumes everything is 100 accurate mechanically. It won't know
if the probe is either sitting slightly off and, calibrate based on the probe location vs the real nozzle tip location.
piezo probs recommended if its a problem. or the paper method. this highly depends on machines qc etc. note extended delta calibrate in klipper docs should be run for best results.
super important. when first setting up "zoffset" always set the "babystep" / offset about 2.5mm high in fluidd or mainsail. print a flat square or some print and lower gradually
back down to 0 this will avoid a potential wrong offset scraping or damaging the nozzle / bed. note some paper brands can vary in thicknes. the nozzle must be perfectly clean
a small amount of debris can and will affect "haptic feedbak of the paper" the nozzle should just touch. a very small amount of ressistance / scratching should be felt. the paper should be
perfectly flat to the bed be careful not to angle it up this will skew results.
initial setup.
step 1 put on probe
step 2 probe_calibrate it will lower and probe a few times once thats done temporarily remove the probe and use paper to find the offset. Keep nozzle cold and clean klipper will account for themral expansion.
step 3 CALIBRATE MACO OR DELTA CALIBRATE NOTE calibrate macro is mostly to set endstop phase or run delta calibrate a few times to see if values are consstent. afterward just delta_calibrate is typically fine unless mods were done.
step 4 print a flat circle or cube to check offset. the first ever setup of zoffset its wise to baby step up 2mm then lower back to 0 to avoid scraping due to miss measurment etc.
babysteps/gcode offset values can be applied with
Z_OFFSET_APPLY_ENDSTOP: Take the current Z Gcode offset (aka, babystepping), and subtract it from the stepper_z endstop_position. This acts to take a frequently used babystepping value, and "make it permanent". Requires a SAVE_CONFIG to take effect.
Z_OFFSET_APPLY_ENDSTOP MUST HAVE DELTA CALIBRATE RUN TO APPLY THE NEW VALUE TO THE 3 ENDSTOPS IT WILL NOT BE ACTIVE OTHERWISE.
Note once you know your offset. it doesn't generally need to be changed. 90 percent of the time it may appear to be different but its often flow related or moisture some times model geometry.
generally this happens changing filament
Makse sure to run bed mesh / delta calibrate at temp your using. results will be skewed by temp differences as the rods / effector / basse change.
its recommended to let it prheat the bed for about 5 -10 mins before running a calibration cycle. after initial setup usally only delta_calibrate and bed mmesh need to be run
once in awile or with wide tamp changes. eg going from 50 bed to 80-110 for higher temp materials
OTHER NOTES delta_calibrate doesn't turn off active bed meshs simply delete the mesh in fluidd or main sail etc if using billinear. It will have to be redone after anyway.
To state What should be apparent. before doing delta calibrate in particular ensure the bed is perfectly clean this will skew results.
bed mesh deviation factor will vary printer to printer. after all the steps are complete the bed mesh in theory should be correct eg flat and not wavy along edges or tilted.
reported deviation rangers are between 0.05-0.1 for stock glass bed. note first layer for a volcano should be at least 0.3 layer height this will clear most deviation regardles.
depending on the topography of the textured surface of the glass. bed mesh may or may not be nessary (old and damaged). in most cases delta calibrate and, belt tilt are enough.
if deviation seems to be a problem increase probe count.
PROBE COMMAND DOCS
Probe
The following commands are available when a probe config section is enabled (also see the probe calibrate guide):
PROBE [PROBE_SPEED=<mm/s>] [LIFT_SPEED=<mm/s>] [SAMPLES=<count>] [SAMPLE_RETRACT_DIST=<mm>] [SAMPLES_TOLERANCE=<mm>] [SAMPLES_TOLERANCE_RETRIES=<count>] [SAMPLES_RESULT=median|average]: Move the nozzle downwards until the probe triggers. If any of the optional parameters are provided they override their equivalent setting in the probe config section.
QUERY_PROBE: Report the current status of the probe ("triggered" or "open").
PROBE_ACCURACY [PROBE_SPEED=<mm/s>] [SAMPLES=<count>] [SAMPLE_RETRACT_DIST=<mm>]: Calculate the maximum, minimum, average, median, and standard deviation of multiple probe samples. By default, 10 SAMPLES are taken. Otherwise the optional parameters default to their equivalent setting in the probe config section.
PROBE_CALIBRATE [SPEED=<speed>] [<probe_parameter>=<value>]: Run a helper script useful for calibrating the probe's z_offset. See the PROBE command for details on the optional probe parameters. See the MANUAL_PROBE command for details on the SPEED parameter and the additional commands available while the tool is active. Please note, the PROBE_CALIBRATE command uses the speed variable to move in XY direction as well as Z.
Z_OFFSET_APPLY_PROBE: Take the current Z Gcode offset (aka, babystepping), and subtract if from the probe's z_offset. This acts to take a frequently used babystepping value, and "make it permanent". Requires a SAVE_CONFIG to take effect.