μRepRap (RepRapMicron) project reports 1μm positioning and 4μm resin droplets

The wiki home page of RepRapMicron, also written μRepRap, describes an open-source project that aims to do for the micron scale what RepRap did for desktop 3D printing. The project says microelectronics and MEMS (micro-electromechanical systems) sit inside most everyday electronics, yet the fabrication methods behind them have grown complex, esoteric, costly, and in many cases dangerous for hobbyists. Its answer is, in its own words, effectively a very precise and compact CNC machine.

The stated objective has two parts: build a macro-scale, 3D-printed micron fabrication and manipulation system, and, while doing so, develop the techniques needed to reproduce a similar system printed at the micron scale. The project credits microscope platforms with sub-micron resolution, developed by The OpenFlexure Project, for creating the opening to try this.

The current recommended build is the "PIKA" V0.02, listed as complete circa May 2026. The page reports repeatable positioning of approximately 1μm and reliable accuracy of 10μm or better in an approximately 8mm cubic volume. With a fine probe etched by the PIKA mechanism itself, UV resin droplets as small as 4μm have been deposited. With a fibre optic probe, 9μm resin lines have been cured. A 0.5mm "Benchy" test model is shown printed with μRepRap at a 4μm layer height.

An earlier testbed for flexure components, the "MAUS C" V0.04 (essentially a debugged V0.03), is the one with full documentation, including a bill of materials and Maus-C-Build-Process instructions. It achieved approximately 5μm positioning in the best case over a 4mm square, though its motion is not linear. It was later updated to V0.05 for compatibility with PIKA Axis Drivers. Pre-built STL files are on Printables, and day-to-day progress is on the RepRap blog. A MAUS 'L' V0.05 was in development in Feb 2026.

How it works: the axes are driven by 400-step NEMA17 stepper motors, and the page says 200-step motors get quite close to 1μm. Driving hardware and firmware are left to the user; existing developments assume interactive GCODE control over a serial connection, and Arduino GRBL is adequate. A probe serves as Z touch sensor, engraving tool and deposition head. A robust starter probe can be made from a 0.5mm hypodermic needle or a 0.25mm acupuncture needle, enough for scratching resist such as Sharpie marker on a glass slide (an early result was a 1.3mm "Hello World" etched this way in mid-2025). For fine deposition the page recommends an electrochemically etched wire point, and says the PIKA can etch its own probes.

Additive fabrication currently works by transferring minute quantities of "Top Coat" UV nail gel resin. The probe picks resin up by capillary action from a smeared layer, and capillary action again forms fairly consistent droplets when depositing. The page says sub-15μm droplets are achievable, with the number and consistency of consecutive droplets depending on the resin and probe. A UV LED in the print bed, run by the hardware controller, sets the resin. One caveat is physical: thin layers of 3D printer UV resin do not readily set in air, because oxygen neutralises the activated photoinitiator. Other resins, ionic gels and other substances may be used, provided the solvent does not evaporate during deposition; materials that stick to the resin, such as gold leaf, can be added.

An optical fibre can also be attached as the probe. A 125/9 optical fibre patch lead and a 3W UV LED module act as a print head that cures resin by projecting UV down the fibre. One example shows a 100μm square set in resin this way.

Setup needs optics: two USB microscopes (minimum 8Mpixel recommended), one to watch the initial Z probe height and one for XY positioning when a print starts, plus a conventional 100x to 200x bench microscope to examine the output.

The project is open about how much is unfinished. Humans cannot handle micron-scale parts directly, so new systems are needed to detach, transport, rotate, position and operate on printed items, and to do maintenance at that scale. Much work remains on what can be printed and in what materials; if objects cannot be printed tall enough or with overhangs, folded structures, hinges and flexures will have to be developed. The authors hope μRepRap may eventually build useful devices and an ecosystem of fabrication tools at a yet smaller scale, and they describe replication at a smaller scale as a possibility and intent, not a result.

Key facts

The PIKA V0.02, the current recommended build (complete circa May 2026), reports repeatable positioning of approximately 1μm and 10μm or better accuracy in an approximately 8mm cubic volume.
Using a probe etched by the PIKA itself, UV resin droplets as small as 4μm have been deposited; a fibre optic probe has cured 9μm resin lines.
Additive fabrication works by transferring UV nail gel resin on a probe via capillary action, with a UV LED in the print bed setting it; the page says sub-15μm droplets are achievable.
The hardware uses 400-step NEMA17 steppers, GCODE over serial (Arduino GRBL is adequate), two USB microscopes of at least 8Mpixel, and a 100x to 200x bench microscope for inspecting output.
The macro-scale system exists; printing a similar system at the micron scale and replication at a smaller scale are stated goals, and the page says much work remains on what can be printed.

Why it matters

Micron-scale fabrication, the realm of microelectronics and MEMS, has been tied to complex, costly and sometimes dangerous silicon processes. The project's pitch is that RepRap-style open hardware could make it accessible, as RepRap did for desktop 3D printing. It argues that biological sciences would gain from inexpensive, rapidly evolving equipment, and that microelectronics could regain independent communities and move beyond its fixation on silicon. For scale, the page notes the Intel 4004 was made with a 10μm feature size, the same figure as the PIKA's reported accuracy. These are the project's own hopes, not demonstrated outcomes.

Who it affects

The page addresses hobbyists and independent builders who want to experiment at the micron scale, and names biology and microelectronics as fields that could benefit if the approach works. Developers are also a target: the current printer is described as a platform for further hardware and software development on the micron scale.

How to use it

The project publishes the designs. Pre-built STL files for the MAUS C are on Printables, and that build has a bill of materials and Maus-C-Build-Process instructions. The recommended build is the PIKA V0.02. You need NEMA17 steppers (400-step; 200-step motors get quite close to 1μm), a controller running GCODE over a serial connection (Arduino GRBL is adequate), two USB microscopes of at least 8Mpixel, and a 100x to 200x bench microscope to inspect output. Probes start with a 0.5mm hypodermic needle or a 0.25mm acupuncture needle; for fine deposition, use an electrochemically etched wire point, which the PIKA can make itself. Deposition uses 'Top Coat' UV nail gel resin. The page gives no cost or bill-of-materials price for the PIKA or MAUS builds.

How solid is it

The figures (about 1μm positioning, 10μm accuracy, 4μm droplets, 9μm lines) come from the project's own hardware reports on its own wiki. The page gives no test method or sample counts, and no independent verification or third-party measurement. The best-case 5μm figure for MAUS C comes with a note that motion is not linear. The page does not say that a micron-scale printer has been built; only a macro-scale 3D-printed system exists.

Risks and caveats

Much remains undetermined about what can be printed and in what materials, and tall objects or overhangs may require folded structures, hinges and flexures. Thin UV resin layers do not set well in air because oxygen neutralises the photoinitiator. Deposition consistency depends on the resin and probe, and the number of consecutive droplets varies. Parts at this scale cannot be handled directly, so detaching, transporting and positioning them needs new systems. Printing at the micron scale and self-replication at smaller scales are stated intents, and the page gives no timescale for them.

“The solution is effectively a very precise and compact CNC machine.”

— μRepRap project wiki