RepRapMicron Delta mods
This page is a development stub. Please enhance this page by adding information, cad files, nice big images, and well structured data!
//identifies pages under construction. remove when the page is relatively stable.
Contents
Creating A Micron Accurate Delta Stage
While the OpenFlexure Delta Stage is excellent for microscopy, it requires modification to become an R&D platform for μRepRap. There are a lot of unknown unknowns though, so it makes sense to push it as far as possible until we know what more of them are. A key difference is that microscopy is in 2D while μRepRap requires motion and observation in 3D. So rather than observing a slide from underneath, μRepRap needs to view the precision work area in several dimensions. This was initially done by extending a beam from the slide platform which can reach a microscope stage, and while this allowed initial testing to proceed it had unwanted vibration and angular movement. Multiple microscope views will probably be required, and coloured multidirectional lighting may make visual alignment easier.
Suggested modifications so far are:
Implemented
- Replace OpenFlexure Raspberry Pi package with RAMPS system running GRBL
- Improve robustness of motor mounts
- Adaptors for NEMA17 motors
- Extended beam to position probe over free-standing microscope
- Rigid probe arm mounted to stage body
- Axis limit switches
Unimplemented
- 'O' Ring accessibility improvement, possible replacement similar to Darwin Z axis anti-backlash nuts
- Illumination with different colours/directions to improve 3D visibility
Practical Progress - Delta Stage
To commence experimentation an OpenFlexure Delta Stage was used, as it had a larger working volume than the Block Stage. It was constructed with two modifications: A simple beam extension (two 85mm lengths of No. 12 fencing wire soldered at an angle for bracing) to allow the probe to be moved in the field of view of a conventional microscope capable of sub-micron resolution, and the driving of a single axis of the stage with a NEMA17 stepper motor. By controlling the stepper with a standard RAMPS board and 3D printer software, a repeatable motion with a step accuracy of approximately 3 microns was observed. Model files required can be found at https://www.printables.com/model/797699
Probe tips were held in standard hypodermic needles bent approx. 4mm from the tip at approx. 80 degrees. A croc clip fits in the cavity in the needle base, and allowed replacement and positioning in the early experiments. Hypodermic needles themselves are reasonably robust substitutes for delicate probes, and the author plans to make a variety of tips.
As the Marlin 3D printer software is theoretically capable of operating a delta stage, though configuration was problematic and the OpenFlexure stage is not a traditional RepRap delta configuration. The solution was to replace the Marlin delta firmware with standard GRBL-servo XYZ-axis software and use a portable Python program to implement the kinematics and a control panel. This is expected to provide an experimental platform capable of reliably moving a probe in 3 dimensions under the view of a suitable microscope. Assistance with porting the model to Marlin or any other RepRap firmware would be appreciated, as this is a more flexible solution long-term.
Subsequently a second experimental platform was built based on the OpenFlexure indirect lighting stage. This allowed a more robust and adjustable probe holder to be fitted to the lighting stage. The same upgrade added microswitches to the delta stage feet to facilitate automatic levelling in GRBL.
Initial tests showed the probe can make marks in a coloured substrate (Sharpie marker on a microscope slide) and deposit controlled dots and streaks of a viscous fluid (motor oil with soot mixed in) at intervals on the slide. Later tests showed good alignment for 0.1mm squares arranged in a cross, and the Sharpie is starting to tear unpredictably losing edge definition with lines thinner than 10μm. Initial tests show deposited dots correspond closely to the tip size, and that multiple dots can be created with one dip of the probe in the fluid. XY movement is reasonably well controlled, and effort is being put in to automate movement using gcode files. Once this is well tested, an attempt will be made to deposit UV-sensitive photopolymer resins, cure them, and progress to multiple layers.
Redesign
These are potential features for a redesign of the Delta Stage, or indeed a Block Stage:
- Integrate NEMA17 motor mounts
- Alternative to 'O'-rings for backlash control
- Increase gear ratio for finer control over the same work area
- Reduce gear backlash with pre-loaded flexures in split driven gears.
- Anchor points for cameras viewing multiple angles
- Multiple attachment points on a 20mm grid for prototyping extensions.
