Current source files are here https://github.com/VikOlliver/RepRapMicron/tree/main/UV_head and pre-built STLs here https://www.printables.com/model/1871410-high-resolution-resin-print-head-for-reprapmicron

An existing method¹ for curing UV resin in situ to make 3D structures can readily be implemented on the μRepRap using commonly available optical fibre leads and UV LED modules.

Do not look into UV source with remaining eye.

Optical fabrication techniques are not a primary aim for μRepRap as they do not work well on scales smaller than the wavelength of light accessible to the typical workshop (~380nm). Thus they will not scale to smaller μRepRaps.

However, using UV light patterning to cure resin in situ offers a rapid way to produce structures with members on the order of tens or hundreds of micrometres a side. These would otherwise be tedious to fabricate with dip-pen techniques.

In μRepRap tradition, all parts are standard commercial components and should be readily available in most localities.

UV source and optical fibre parts described below (left), Optic Probe Holder on Z Axis (right)

UV Light is directed down a Single Mode Fiber Optic Cable (A), in this case a 1G/10GB SMF Duplex Fiber Cable 9/125µm Jumper Patch Cord. It has been expediently converted to a Simplex cable with a hacksaw.

UV is generated by a "3W" (dubious manufacturer rating) UV LED module (B). This module uses an aluminium PCB as its heatsink. If not available as component, it can also be found inside many round, gooseneck USB UV lamps sold for hobbies. The module is intended to be driven at 5V.

The LED Module is held in alignment with the Optical Fibre by an adaptor (C) using two M3 x 20mm screws, with M3 washers under the heads and the two M3 nuts.

A screw terminal block (D) is used to put a 56 Ohm 5W resistor in series with the LED module. The resistor is required in this case as the RAMPS 1.4 CNC controller is driving the D9 output pin ("Coolant Mist", GCODE M7) at 12V and the module is rated at 5V. The terminal block is used instead of solder because the 5W resistor gets exceedingly warm.

TBC

## Software

The fibre_process.py Python filter converts GCODE output by Prusaslicer in "Marlin" mode. As there is no extruder, the filter detects any GCODE movement where the extruder A Axis is being driven forward and uses M7/M9 to turn the Coolant Mist driver on and off appropriately. It also scales the print up by 10x. Prusaslicer needs to run at a scale of 1mm to 10μm as it will not permit the use of a "9mm" nozzle, and the filter rescales this to microns (which are millimetres as far as GRBL is concerned - it's complicated).

Take a clean glass microscope slide and cover an area approximately 1/3 the way up the slide with 12mm Kapton^TM Tape (or one of the clones occasionally referred to as "Krapton Tape"). Take care when laying the tape down that no debris or air bubbles are trapped underneath. Smooth vigorously with a lint-free cloth and IPA. Dry with air blast if available. If desired, affix a ring binder reinforcement ring to the reverse of the slide. This can be helpful in defining an area where microscopic prints may be.

Fit the Optic Probe Holder to the Probe Assembly and fit that to the Z Axis. Ignore the Probe Wire, this is not used in the optical configuration. Fit a clean glass slide to the Stage and secure with the magnetic clamps. Run the homing sequence on the controller to reset the X, Y and Z axes.

Adjust the Probe Arm such that the square hole for the Optical Fibre is centred over the centre of the Stage. Insert one end of the Optical Fibre into the Holder. Lower the Probe and adjust the height of the Probe Assembly so that the Optical Fibre contacts the Slide perpendicularly - use the microscope side view to confirm.

Determine the point where the Optical Fibre touches the Slide By raising and lowering the Probe. Raise the Probe by 5μm and set that height as Z=0. Run the intended print and ensure that the Optical Fibre remains clear of the Slide. It is better to leave too much space than too little, and under most circumstances an extra 5μm will not cause the print to fail.

Abort the print when satisfied. Raise the Z Axis to 200-300μm and using a pointed probe or hypodermic needle apply a droplet of standard 3D printer resin as close as possible to the area immediately under the Optical Fibre. Be careful not to move the Probe. It is best to bring the needle alongside the fibre tip rather than try to go under it. Let surface tension do the distribution.

If using a hypodermic syringe, fill the syringe with resin without the needle attached or it will be very slow and will probably foam. Then eject excess air and cap the needle. A 1ml syringe will last several days if stored in the dark and is ample for dozens of prints.

Trail the droplet so that it contacts the side of the Optical Fibre, and forms a continuous blob from the Fibre to the Slide. You should be able to move the Probe 200μm in any direction, and still have it surrounded by copious quantities of resin.

At this point the print can be run. When complete, raise the Z Axis 200-300μm and manually lift the Probe clear of the resin. At this point the resin is not completely cured and is extremely soft.

Undo the magnetic clamps and carefully remove the slide. Place the slide on an absorbent pad under a microscope and gently wash the print with isopropyl alcohol from a hypodermic syringe or fine pipette.

Tilt the slide to remove excess alcohol, and gently wash again. With a fan, gently blow on the alcohol until it has evaporated enough to reveal the outline of the printed object, then repeat the washing process. Completely dry using the fan.

Use a UV lamp or chamber to completely cure the printed structure.

Do not forget to clean the Optical Fibre.

Pristine fibre optic end (left) and end covered with set resin film (right)

The end of the fibre optic should be cleaned with isopropyl alcohol and a cotton bud or lint-free cloth after each use. If this is not done, a film of cured resin will eventually build up on the end of the fibre causing prints to take the form of puddles.

If this occurs, scrape the film from the end of the Fibre with an unadorned fingernail, and then clean as usual.

Work by Diffraction Limited demonstrated how a UV laser directed down an optical fibre could be used by a micromanipulator to print 3D objects on the micron scale. Their technique is more refined, using a 4μm fibre, and adding a readily available dye to the resin to prevent the UV beam from over-curing the depth of the resin. These may be implemented on the μRepRap, but the objective from μRepRap's perspective is to rapidly generate bulk material. Fine detail can be added with probe-based deposition later