5 axis printing

Hi to all

I want to make 5d printing with marlin firmware and ramps 1.4 board. can anyone help with this

Thanks advance

What do you mean by 5d? (as marketing people have corrupted the meaning)

Marlin only support x,y,z e0 and e1 (I hear rumors there is work being done on adding another axis, but havent looked into it)

Other firmware Teacup for eg do support more axis on a ramps board.

Hi Dust,

5d means 5 dimensional where revolving and rotational axis will be there. And I want to add to marlin firmware with respect to arduino and ramps.

... the actually fastest way for this is a "conventional" 5-axis robot - [robodk.com]

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

No firmware I know has support for any machine type with 5 axis kinematics, and no slicer I know supports it either.

I suspect 5 axis 3D printing will make the CAM process much more similar to what CNC router currently require, where the operator will need to choose and select different toolpaths (or in this case an angle of attack for the nozzle) manualy.

Quote
Trakyan
No firmware I know has support for any machine type with 5 axis kinematics, and no slicer I know supports it either.

The current release of RepRapFirmware supports up to 9 axes on Duet 2 and up to 6 axes on other electronics. Even before we added support for additional axes, some students used RRF to run a 5-axis 3D printer - I think it was XYZ plus tilt in two directions. There is at least one current user driving a 4-axis printer using RRF.

But as you say, the real issue is slicing for more than 3 axes. The user I know with a 4-axis machine post-processes the GCode to add the coordinates for the 4th axis.

We also have several users who configure additional axes for other purposes, such as putting the extruder drive on its own axis to allow a very short Bowden tube, or driving a filament changer.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I wasn't so much referring to the ability to drive more than 3 axis. I was talking more about the kinematics for a 5 axis machine. For instance slicers output cartesian co-ordinates, the firmware then processes those into the appropriate step/dir signals for whatever kinematics the printer uses (cartesian, corexy, delta). Does the RRF inherently support XYZ+2 rotary axis, or a 5 axis robot arm's kinematics, where I simply tell it an XYZ co-ord and two angles of attack and it converts those to the step/dir required on its own? That's what I meant by a firmware that supports 5 axis kinematics, not just supports 5 stepper drivers (+extruders). By the sounds of it that work is being done outside of the firmware by the post processor.

Support for the kinematics in firmware is (relatively) trivial and I'm sure the kinematic equations are out there already for a lot of 5 axis machines.

I feel the law of diminishing returns is kicking in though. You'd get a bit more design freedom with fewer geometric constraints. I suppose you could use this to add "layers" perpendicular to each other, creating a crossing grain type of effect like plywood or some fibre composites have. The problems I see here are that there aren't really an overwhelming number of geometric constraints to begin with and if you really need more strength, try looking at a different fabrication method since the increase in strength will be marginal if even applicable (if a part only really sees load along one plane, orient that with your layers). This all comes at the expense of a much more complicated machine and a much less simple CAM process for the end user. Imagine getting into 3D printing and the first thing you have to do is sort out what orientation you want the layers in for all the different surfaces of your model, how and where the different layer orientations meet (voronoi patterns?)...

I agree, the benefits of using more than 3 axes probably outweigh the additional complication. One possible benefit could be to reduce the need for printed support. To do it well I think you would need 3 Cartesian axes and a bed that both rotates and tilts. By using rotation instead of tilting in two directions, you could use an asymmetric hot end that can print at high angles of bed tilt, because one side of the hot end can be designed with no protrusions that would foul on the bed or the print.

As you say, adding support for the kinematics in firmware is not difficult. RRF uses a class hierarchy for kinematics so that new ones can be added with few or no changes to the rest of the firmware. It's in the slicer that most of the work would need to be done.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Since dc42 dragged 32bit boards into the conversation

I have to point out that smootheware/smoothieboards also supports up to 6 axis + extruders
see [smoothieware.org]

Well then I guess we should all focus our efforts into a new 6 axis slicer since it seems thats what all the big (control)boards on the block are after
I would sure love to help but I didn't pay much attention in my computer algorithms paper, I swear I learnt more during the exam trying to figure out the concepts and algorithms on the fly than I did from any of the lectures.

A lot has been done for Marlin 2.0 towards adding additional non-extruder axes that can be controled by G1 and homed with G28 (have a look at my bf2_6axis_dev branch https://github.com/MarlinFirmware/Marlin/compare/bugfix-2.0.x...DerAndere1:bf2_6axis_dev). Those axes could be used as the basis for introducing future 4 axis, 5 axis or 6 axis kinematics. The additional endstops already work. If someone finds the missing part to make the additional steppers move, please report at https://github.com/DerAndere1/Marlin/issues/1.

If I could make a guess, another change in the files Stepper.h and/or Stepper.cpp has to be made.

BTW: Marlin 1.1.9 has support for the hangprinter (it has 4 kinematic axes) if you enable the HANGPRINTER option. It reduces the maximum possible number of extruders by one, though.

Cheers
DerAndere1


----------------------------------------------------------------------------------------------------
DerAndere
https://it-by-derandere.blogspot.com

Edited 1 time(s). Last edit at 02/22/2019 06:05PM by DerAndere1.

Last I dealt with smoothieware, which was shortly after 6 axis support was added, it was 6 axis OR use extruders, as the board didn't have enough memory for the planning queue. The smoothieware tenders said maybe you could turn off enough other features to fit it, but sounded doubtful. I don't know if things have changed since.

Quote
Dust
Since dc42 dragged 32bit boards into the conversation

I have to point out that smootheware/smoothieboards also supports up to 6 axis + extruders
see [smoothieware.org]

We added support for up to 6 non-extruder axes to official MarlinFirmware/Marlin . The information I posted above is now outdated. Documentation of the feature can be found in this pull request

Edited 1 time(s). Last edit at 08/30/2021 04:35AM by DerAndere1.

... will see -- receiving a pan-tilt module for a big XY-micropositioner, so in need of a controller (and software) with XY+AB with the micropositioner and an external Z-axis or a parallel scara (2 polar axes +Z).

Now looking into proper software (which dind't cost a fortune) to get in touch with (RoboDK 1-2k€, linuxCNC for free but much more preparing needed)

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]