Thanks to the tireless efforts of Robert C. Nelson, Machinekit uSD images for the BeagleBone are now part of the regular Debian image build cycle. The first image is available now, with updates expected about every two weeks.
I will update my Machinekit page as soon as I get a chance, but for now just follow along with any of the BeagleBone getting started guides to put the xz compressed image onto a uSD card. Once you're running the image, you can easily perform updates using the standard Debian package management tools.
Many thanks to Robert for building the images (and the required Xenomai kernel), everyone working on the Machinekit project for the packages, and to all of you who use these tools to drive your machines.
Wednesday, January 21, 2015
Monday, January 19, 2015
3D Printing With Machinekit
While it is possible to 3D print with a "traditional" CNC controller like Mach3, LinuxCNC, or Machinekit, there are a lot of small problems that get in the way. There is a lot of ongoing work in the Machinekit project to make this process much easier.
Alexander Rössler has been working on gcode remapping and a standard set of signal names that will reduce the amount of gcode post-processing required and improve print quality (migrating "queue-busting" custom M1xx codes for setting things like heater temperatures and fan speed to inline analog output values that keep the hot-end moving).
Bas de Bruijn has been reworking the documentation files and build scripts, allowing documentation to exist in a separate (and easier for non-programmers to modify) repository. Very soon you should see online documentation built automatically just like the buildbot packages for the executable files.
Jean-Paul Moniz has been working on getting some of the fancier bed-probing and auto-calibration features common in Arduino firmware working on Machinekit. I (Charles Steinkuehler) helped this effort slightly by exporting some homing settings to HAL that previously required a shutdown and restart of the linuxcnc application to change (traditional machine tools don't need to change their homing details very often).
In addition, work is continuing with remote interfaces, velocity based extrusion, simplifying the example HAL files, and other changes to help make Machinekit more accessible for new users looking for a solution with more "upside" than an AVR or Cortex M4. Machinekit scales from systems like the BeagleBone and Raspberry-Pi to multi-core x86 systems with custom FPGA hardware interfaces.
I'm sure I missed several folks who've reported progress on the mailing list, but it's not intentional. There's just so much happening with Machinekit already this year it's hard to keep track of it all!
Remember, the best way to keep up to date is to monitor the Machinekit Google group. Drop in and say hi!
Alexander Rössler has been working on gcode remapping and a standard set of signal names that will reduce the amount of gcode post-processing required and improve print quality (migrating "queue-busting" custom M1xx codes for setting things like heater temperatures and fan speed to inline analog output values that keep the hot-end moving).
Bas de Bruijn has been reworking the documentation files and build scripts, allowing documentation to exist in a separate (and easier for non-programmers to modify) repository. Very soon you should see online documentation built automatically just like the buildbot packages for the executable files.
Jean-Paul Moniz has been working on getting some of the fancier bed-probing and auto-calibration features common in Arduino firmware working on Machinekit. I (Charles Steinkuehler) helped this effort slightly by exporting some homing settings to HAL that previously required a shutdown and restart of the linuxcnc application to change (traditional machine tools don't need to change their homing details very often).
In addition, work is continuing with remote interfaces, velocity based extrusion, simplifying the example HAL files, and other changes to help make Machinekit more accessible for new users looking for a solution with more "upside" than an AVR or Cortex M4. Machinekit scales from systems like the BeagleBone and Raspberry-Pi to multi-core x86 systems with custom FPGA hardware interfaces.
I'm sure I missed several folks who've reported progress on the mailing list, but it's not intentional. There's just so much happening with Machinekit already this year it's hard to keep track of it all!
Remember, the best way to keep up to date is to monitor the Machinekit Google group. Drop in and say hi!
Subscribe to:
Posts (Atom)