Download Test Gcode
Shakita Sohm
Here is a piece of the gcode from my salt and pepper shakers from a contest a while back. It is from Fusion 360 using the Carbide 3D post version 43618 from the HSM posts library online. It uses the 3D Adaptive, 3D Contour, 2D Contour, Face, Horizontal and Bore tool paths.
Bottom Inside.nc (1.5 MB)
Hi. I have a mini CNC machine. I have GRBL controller. I downloaded Inkscape to convert images into gcode but for some reason Inkscape is not saving my file. It has an error. What other pogram can I use to save as gcode for a CNC machine? Thank you so much for your help!
Hi Tim. Thank you so much for writing me. I have just reassembly my mini cnc because it drew a strange line after finished to draw a picture. I made the cnc from mdf and I am testing it as a plotter. I think to transform it into a milling machine.
Tony, glad you reached out. Have you tried reading the gcode program you created that is saved on the SD card with your computer? We found that the SD card itself was the issue. I recommend trying another SD card. Let me know if that resolves your issue.
Probes come in many flavors and as such have varying levels of accuracy, reliability, and repeatability, depending on several factors. This command tests the probe for repeatability (precision) and produces a standard deviation based on two or more probes of the same XY position.
Here are two. One is a LS engine motor mount plate and the other is a simple tab.Now these were done without any z axis because they were done for the original crossfire as I am still waiting on Box 2 and my THC.1.75 flange gcode.tap (1.5 KB) LS MM single modified gcode.tap (1.4 KB)
Watch the videos and then work through each tab. I have created a custom gcode generator to assist in making testing towers. This used to be a laborious process and beyond the skills of many users. Other times pre-sliced gcode was used from the internet, but it is impossible to have gcode available for every printer configuration. Until now!
Every attempt has been made to ensure this is safe but ultimately there always is risk in running pre-sliced gcode from the internet. Preview the gcode in your slicer or Zupfe GCode Viewer and print at your own risk.
The gcode generated by this page is originally from Simplify3D. This website then uses Javascript to modify the contents based on user inputs. This site is not a web based slicer, therefore it is limited in some ways.
A 'calibration' slicer profile in S3D is used as the basis of the gcode on this site. S3D offers multiple processes to assist with splitting the towers into segments where the print settings can vary. Apart from this, the only special functionality used is post processing scripts to delete some lines, and to modify others with simple search and replace functions.
You may notice settings related to temperatures, retraction, Z hop, part cooling, etc have set values, but these are altered by post processing scripts and this site to ultimately be set using the user's inputs. Several parameters work like this, please don't be fooled by what is in the slicing profile. Opening the final gcode file in a text editor and searching for 'custom' will show if the user's inputs have been successfully adopted.
The information below is mainly for my reference. However, if you wish to duplicate the tests yourself out of interest or perhaps to develop a new test for the site, then the steps must be followed exactly, including replicating the process names.
First layer test: No changes, although it should be noted that a single square is included which is then duplicated and positioned by this site. Non uniform scaling of the source STL needs to occur to suit certain nozzle/layer combinations. The square should be 25 x 25 mm with it's height scaled to match the target layer height.
It may be preferable to have the printer as close to printing conditions as possible during these tuning procedures. That means having filament loaded and the part cooling fan on for PLA temperatures. If there is no UI button available to turn on the part cooling fan, you can do it manually via gcode with M106 S255.
The bed can be probed at the start of the print with a G29 command, with the resulting mesh immediately used to compensate as the initial layers are produced. Alternatively, the bed can also be probed some other time (while not printing), the mesh stored in the EEPROM and then restored with M420 S1 at the start of a print. In this case the print will start sooner, since we do not need to wait for a new mesh to be probed, although it may not be as accurate if anything has changed since probing. Either of these gcode commands should come after the G28 home command in the start gcode.
This test is intentionally placed before others because it is assumed that your first layer must be reasonable for the later tests to succeed. In some cases, however, if your flow, retraction, etc is way off, it may prevent your first layer from sticking properly and you may wish to try these other tests first.
If one side looks too close, but the other too far, adjust the levellng knobs to correct this. It is worth printing this gcode more than once after making adjustments to make sure the result is accurate and repeatable.
The cube should look similar to those at the top of this page. If there are no major issues, please continue to the next step. If there is a significant defect, the culprit will likely be found by working through the frame tab. Minor issues will hopefully be resolved with the subsequent tests.
Once you have determined the correct value, it must be saved to the firmware to take effect on subsequent prints. Although it can be hard coded into the firmware by recompiling Marlin, it is far easier to use gcode to achieve this.
Unfortunately, I can't provide pre-sliced gcode for this process. It is vital to use gcode generated by YOUR slicer. Setting up your slicer to print the cube in the right way should be simple by following these steps:
To overcome this, you may scale up the X and Y dimensions of the cube. As long as the file is sliced as described above, the wall thickness will not alter from this change in scale and the test will be valid.
You may also consider holding the cube up to a bright light source, to see if there are any gaps in the extrusion. Small gaps may indicate the need to margially increase flow rate. If this test cube is too small to do this effectively, most models completed in vase/spiral mode will be suitable.
The gcode generators on this site work by using javascript to modify source gcode originally created by Simplify3D. However, when you completed the calibration test above, you sliced your own gcode, making your own baseline and then making a flow adjustment relative to that. Therefore, this test is unique from the others on this site which is why the flow rate doesn't necessarily translate.
Let's say your old flow rate was 100% and you have tested and corrected this to 96%. The gcode on this site originally had a flow rate of 90% when sliced, so applying your 96% to that gives a final result of 86.4%, not 96%. Your slicer profile settings will also be different in other ways, which further complicates matters. Therefore, there is not a straightforward correlation between your slicer and my gcode generators.
The aim of the site is to discover ideal settings you can apply to your own slicer profile, not to optimise the gcode created by the generators. Keep this in mind and focus on the aim of each test, rather than the general print quality.
If you are experiencing significant over or under extrusion that prevents you from using the tests properly, by using the custom start gcode function on this site you can optionally issue an M221 to override the values in the generatored gcode. For example, using M221 S90 would tell the firmware to only extrude 90% of what the gcode asks for. This is an easy method for making a quick correction that will alow the tests to complete successfully.
On TMC drivers, the current is set directly with gcode commands. This can be set in the firmware, via a terminal or by using the printer's LCD. This value should then be saved to EEPROM to stay persistent.
TMC drivers connected via UART or SPI serial can easily have their current set via gcode. This is not peak current, but rather RMS (root mean square) current. Rather than the maximum, think of this as more a typical/average current, where the driver will be operating mostly. To convert the peak current from stepper motor specs to RMS, divide it by 1.41.
The following form will create a retraction tower to conveniently test back to back parameters in the same print. Of the three available parameters, it is best to change only one per test print. For example, keep the retraction speed and extra restart distance the same, but vary the retraction distance over each segment. Changing more than one parameter makes is hard to tell what made the difference. The print is quick, so repeat the test varying other parameters until you are happy with them all.
Here is the STL if you would like to slice a similar test yourself: retractiontestv2.stl. This file has been updated to V2, which changes the external shape from circular to pentagonal. It is also prints slightly faster. The original file is still available here: retractiontest.stl
I would then repeat the test, setting the same retraction distance for each segment and instead altering the retraction speed to dial that in. A third test could then take place to test extra restart distance, a fourth for Z hop, etc.
If you would like to be able to customise additional parameters for a retraction test, Prahjister has made a great tool: Retraction Calibration Tool. It has a higher degree of difficulty due to needing more parameters but is ultimately more powerful. Warning! This is an external website and beyond my control. Some users have reported success and others have had issues with the gcode generated. As with the gcode made by this website, monitor your printer during printing with a view to cutting the power if needed.
df19127ead