aboutsummaryrefslogtreecommitdiffstats
path: root/docs
diff options
context:
space:
mode:
authorKevin O'Connor <kevin@koconnor.net>2019-09-21 14:16:02 -0400
committerKevin O'Connor <kevin@koconnor.net>2019-10-23 19:28:03 -0400
commitc794e11c67d7f48061cb56d5c5c0a2e98ef28865 (patch)
tree6ebd28ef2f241d5d2a17c664e3391d74e88c53ca /docs
parentb8a616064674926cec857acdc99e5636b0e30b21 (diff)
downloadkutter-c794e11c67d7f48061cb56d5c5c0a2e98ef28865.tar.gz
kutter-c794e11c67d7f48061cb56d5c5c0a2e98ef28865.tar.xz
kutter-c794e11c67d7f48061cb56d5c5c0a2e98ef28865.zip
docs: Update Pressure_Advance.md to use tuning tower
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Diffstat (limited to 'docs')
-rw-r--r--docs/Pressure_Advance.md112
-rw-r--r--docs/img/tune_pa.jpgbin0 -> 24502 bytes
-rw-r--r--docs/img/tuning_tower.jpgbin0 -> 20509 bytes
3 files changed, 57 insertions, 55 deletions
diff --git a/docs/Pressure_Advance.md b/docs/Pressure_Advance.md
index 8a849c88..d882b68d 100644
--- a/docs/Pressure_Advance.md
+++ b/docs/Pressure_Advance.md
@@ -13,58 +13,64 @@ uses the second feature (reducing blobbing during cornering) as a
mechanism for tuning.
In order to calibrate pressure advance the printer must be configured
-and operational. The tuning test involves printing objects and
-inspecting the differences between objects. It is a good idea to read
-this document in full prior to running the test.
+and operational as the tuning test involves printing and inspecting a
+test object. It is a good idea to read this document in full prior to
+running the test.
Use a slicer to generate g-code for the large hollow square found in
-[docs/prints/square.stl](prints/square.stl). Use a high speed (eg,
-100mm/s) and a coarse layer height (the layer height should be around
-75% of the nozzle diameter). It is fine to use a low infill (eg, 10%).
+[docs/prints/square_tower.stl](prints/square_tower.stl). Use a high
+speed (eg, 100mm/s), zero infill, and a coarse layer height (the layer
+height should be around 75% of the nozzle diameter).
-Prepare for the test by issuing the following G-Code commands:
-`SET_VELOCITY_LIMIT SQUARE_CORNER_VELOCITY=1 ACCEL=500` and
-`SET_PRESSURE_ADVANCE ADVANCE_LOOKAHEAD_TIME=0`. These commands make
-the nozzle travel slower through corners and they emphasize the
-effects of extruder pressure.
+Prepare for the test by issuing the following G-Code command:
+```
+SET_VELOCITY_LIMIT SQUARE_CORNER_VELOCITY=1 ACCEL=500
+```
+This command makes the nozzle travel slower through corners and to
+emphasize the effects of extruder pressure. Then for printers with a
+direct drive extruder run the command:
+```
+TUNING_TOWER COMMAND=SET_PRESSURE_ADVANCE PARAMETER=ADVANCE START=0 FACTOR=.005
+```
+For long bowden extruders use:
+```
+TUNING_TOWER COMMAND=SET_PRESSURE_ADVANCE PARAMETER=ADVANCE START=0 FACTOR=.020
+```
+Then print the object. When fully printed the test print looks like:
-For the first print use a pressure advance of zero by running
-`SET_PRESSURE_ADVANCE ADVANCE=0.000`. Then print at least 10 layers of
-the test object. While the object is printing, make a note of which
-direction the head is moving during external perimeters. What many
-people see here is blobbing occurring at the corners - extra filament
-at the corner in the direction the head travels followed by a possible
-lack of filament on the side immediately after that corner:
+![tuning_tower](img/tuning_tower.jpg)
-![corner-blob](img/corner-blob.jpg)
+The above TUNING_TOWER command instructs Klipper to alter the
+pressure_advance setting on each layer of the print. Higher layers in
+the print will have a larger pressure advance value set. Layers below
+the ideal pressure_advance setting will have blobbing at the corners,
+and layers above the ideal setting can lead to rounded corners and
+poor extrusion leading up to the corner.
-This blobbing is the result of pressure in the extruder being released
-as a blob when the head slows down to corner.
+One can cancel the print early if one observes that the corners are no
+longer printing well (and thus one can avoid printing layers that are
+known to be above the ideal pressure_advance value).
-The next step is to increase pressure advance (start with
-`SET_PRESSURE_ADVANCE ADVANCE=0.050`) and reprint the test object.
-With pressure advance, the extruder will retract when the head slows
-down, thus countering the pressure buildup and ideally eliminate the
-blobbing.
+Inspect the print and then use a digital calipers to find the height
+that has the best quality corners. When in doubt, prefer a lower
+height.
-If a test run is done with a pressure advance setting that is too
-high, one typically sees a dimple in the corner followed by possible
-blobbing after the corner (too much filament is retracted during slow
-down and then too much filament is extruded during the following speed
-up after cornering):
+![tune_pa](img/tune_pa.jpg)
-![corner-dimple](img/corner-dimple.jpg)
+The pressure_advance value can then be calculated as `pressure_advance
+= <start> + <measured_height> * <factor>`. (For example, `0 + 12.90 *
+.020` would be `.258`.)
-The goal is to find the smallest pressure advance value that results
-in good quality corners:
-
-![corner-good](img/corner-good.jpg)
+It is possible to choose custom settings for START and FACTOR if that
+helps identify the best pressure advance setting. When doing this, be
+sure to issue the TUNING_TOWER command at the start of each test
+print.
Typical pressure advance values are between 0.050 and 1.000 (the high
end usually only with bowden extruders). If there is no significant
-improvement after gradually increasing pressure advance to 1.000, then
-pressure advance is unlikely to improve the quality of prints. Return
-to a default configuration with pressure advance disabled.
+improvement with a pressure advance up to 1.000, then pressure advance
+is unlikely to improve the quality of prints. Return to a default
+configuration with pressure advance disabled.
Although this tuning exercise directly improves the quality of
corners, it's worth remembering that a good pressure advance
@@ -72,8 +78,8 @@ configuration also reduces ooze throughout the print.
At the completion of this test, update the extruder's pressure_advance
setting in the configuration file and issue a RESTART command. The
-RESTART command will also return the acceleration, cornering speeds,
-and look-ahead times to their normal values.
+RESTART command will clear the test state and return the acceleration
+and cornering speeds to their normal values.
Important Notes
===============
@@ -91,20 +97,16 @@ Important Notes
[nozzle temperature](http://reprap.org/wiki/Triffid_Hunter%27s_Calibration_Guide#Nozzle_Temperature)
prior to tuning pressure advance.
-* It is not unusual for one corner of the test print to be
- consistently different than the other three corners. This typically
- occurs when the slicer arranges to always change Z height at that
- corner. If this occurs, then ignore that corner and tune pressure
- advance using the other three corners.
-
-* Check for warping at the corners during the test prints (the corners
- detaching from the bed and rising a small distance upwards during
- the print). If one corner appears warped then ignore that corner
- when tuning. If significant warping is seen throughout the test then
- typical solutions are to reduce the slicer's first layer speed,
- adjust the bed temperature, and/or to use the slicer's brim feature.
- Pressure advance itself is unlikely to impact warping, but this
- tuning test is sensitive to it.
+* It is common for the test print to show different behavior on each
+ corner. Often the slicer will arrange to change layers at one corner
+ which can result in that corner being significantly different from
+ the remaining three corners. If this occurs, then ignore that corner
+ and tune pressure advance using the other three corners. It is also
+ common for the remaining corners to vary slightly. (This can occur
+ due to small differences in how the printer's frame reacts to
+ cornering in certain directions.) Try to choose a value that works
+ well for all the remaining corners. If in doubt, prefer a lower
+ pressure advance value.
* If a high pressure advance value (eg, over 0.200) is used then one
may find that the extruder skips when returning to the printer's
diff --git a/docs/img/tune_pa.jpg b/docs/img/tune_pa.jpg
new file mode 100644
index 00000000..f8f4d18e
--- /dev/null
+++ b/docs/img/tune_pa.jpg
Binary files differ
diff --git a/docs/img/tuning_tower.jpg b/docs/img/tuning_tower.jpg
new file mode 100644
index 00000000..694b9a50
--- /dev/null
+++ b/docs/img/tuning_tower.jpg
Binary files differ