HOW DIY TECHNOLOGIES ARE DEMOCRATIZING SCIENCE
A fully motorized microscope you print at home and clinics in Tanzania actually use for malaria diagnosis.
by OpenFlexure Project, University of Bath
ScienceOpen-hardware
- difficulty
- ●●●●○
- time
- a weekend-plus
- license
- CERN-OHL-S
- repo
- repo ACTIVE0 stars
●●●●○ · a weekend-plus · CERN-OHL-S · 0 stars · repo ACTIVE
WHAT YOU’LL NEED
- 3D printer + filament — printable parts — files are in the repo
- Electronic parts — full list with part numbers in the repo BOM
- Dev board / microcontroller — runs the project firmware
Partner
COMPAREE VERDICT
OpenFlexure is the rare open-hardware project that moved from university research to real clinical trials. The University of Bath design uses printed flexure joints — thin plastic hinges that flex rather than slide — to position the stage in sub-100-nanometre steps, which is why it does not need bearing rails. A Raspberry Pi drives the motors and camera, so the scope can scan a whole slide autonomously. In Tanzania, BongoTech manufactures them locally and the Ifakara Health Institute is evaluating them for malaria diagnosis. The current v7 build is labelled beta, and rightly so: you are assembling a precision instrument from plastic, cheap steppers, and friction-fit optics, then asking it to hold sub-micron focus. Most of a weekend will be tuning backlash and getting the illumination even. If you want a ready-to-go USB microscope, buy one. If you want to understand how a microscope works — and contribute data to an open hardware project with published clinical trials — this is the one to build. The single hardest part is not the print or the assembly; it is accepting that your first calibration will drift and you will need to go back in.
IN THE REPO
NOT IN THE REPO
- —CAD files for the body and stage are in the GitLab repo; printed flexures replace all metal rails.
- —Bill of materials lists the Raspberry Pi, camera module, stepper motors, optics, and fasteners — sources included.
- —Python-based server software runs on the Pi; autofocus and whole-slide scanning are built in.
- —Build guide, assembly videos, and user documentation are all present; the project labels v7 as beta.
- —Licensed CERN-OHL-S (strongly reciprocal); commercial manufacture allowed but derivative designs must share-alike.
- —Peer-reviewed in multiple journals including Biomedical Optics Express 2020 (best paper award).
Can I build this?
Build at your own risk. Projects involve tools, electronics and sometimes mains voltage — follow the creator’s safety notes.
Partner · KickstarterHeyGears G1: 10M+ colors and transparent parts in one print, plus UV printing on flat objects. Figures, parts and labels — no painting.
Gallery
Start here
Navigation into the creator’s own docs — we don’t rewrite the guide, we route you to the source.
- 1.Read the build guide on openflexure.org and verify your Pi and camera module are on the compatibility list (The v7 beta docs call out known Pi and camera combinations that work; starting with untested hardware wastes the weekend.)
- 2.Clone the GitLab repository and slice the STL files with 0.2 mm layers and 40% infill or better (The flexures need clean layer adhesion — poor infill or wet filament will cause them to crack under load.)
- 3.Order the bill of materials early; objective lenses and stepper motors often have two-week lead times(The BOM is in the repo docs folder.)
- 4.Assemble the stage and flex it by hand before installing motors — you are checking for binding(If a flexure binds or snaps during dry assembly, reprinting one part is easier than debugging it later.)
KNOWN ISSUES
- Printing the flexures in PLA that has absorbed moisture or with too little infill — they crack under repeated motion and you will not know why the stage drifts.
- Buying a non-RMS-thread objective or a camera module not listed in the compatibility table — the optics module is designed around specific thread pitches and sensor sizes.
- Expecting plug-and-play autofocus on the first boot — the motor calibration and backlash compensation need tuning, and the docs are honest that v7 is still beta.
- Tightening the M3 screws that clamp the flexures until they crack — the build guide gives torque guidance but it is easy to overshoot on the first assembly.
- Skipping the optical alignment step and wondering why one corner of the field is sharp and the other is not — the condenser height and lens spacing both matter.
- Assuming clinical-grade performance out of the box because it is being evaluated in Tanzania — those units are tuned and maintained by people who built dozens; your first one will be a learning build.
What magnification can this achieve?
Depends entirely on the objective lens you install — the optics module accepts standard RMS-thread objectives, so 10×, 20×, 40×, and oil-immersion 100× are all possible. The docs recommend starting with a 10× air objective for alignment.
Can I use a Raspberry Pi Zero or an older Pi 3?
The Pi Zero works but is too slow for real-time autofocus; the Pi 3 works but the Pi 4 is strongly recommended for whole-slide scanning. The compatibility notes are in the build guide.
Is this actually being used in clinics?
Yes — BongoTech in Tanzania manufactures them locally and the Ifakara Health Institute is evaluating them for malaria diagnosis. The Nature and Tom's Hardware articles confirm this, and the openflexure.org case studies page names the partners. It is not WHO-approved or universally deployed, but it is past the proof-of-concept stage.
Do I need to calibrate it every time I use it?
No — once the backlash compensation is tuned and the optics are aligned, it holds calibration across sessions. But the first-time setup is not five minutes.
Can I sell microscopes built from this design?
Yes, the CERN-OHL-S licence allows commercial manufacture, but any modifications you make to the design must be released under the same licence — it is strongly reciprocal.
Community builds
No community builds yet — be the first, we feature the best ones.
Discussion1
FROM THE COMPAREE TEAM
Sub-100-nanometre stage steps from printed plastic flexures — and clinics in Tanzania are using these for malaria diagnosis. If you built one, what would you point it at first?
OpenFlexure Project, University of Bath
Richard Bowman and the Bath Open Instrumentation Group developed OpenFlexure to make precision microscopy accessible in low-resource settings. The design has been peer-reviewed in journals including Biomedical Optics Express (2020 best paper) and is now manufactured locally in Tanzania for clinical trials.
Keep building — related projects
Same category, picked by how close they are to this build.
FULL CADOne homemade board that can listen to almost every radio signal in the air around you
Open-hardwareScience●●●●○advanced
FULL CADA lab positioning stage costs 10,000 dollars. This one steps 50 nanometers and is open source
ScienceOpen-hardware●●●●●expert
Partner · KickstarterCodeeBot: build the robot from Lego-compatible bricks and program it by hand; a GPT tutor explains each step.
FULL CADAllsky: a Raspberry Pi camera in a weatherproof dome that films the whole night sky by itself
ScienceOpen-hardware●●●○○moderate
NO CAD FILESTinyGS — open source global satellite ground station network
Open-hardwareScience●●○○○starter
DISCLAIMER
- Comparee is not the author of the projects featured here. All rights to each project belong to its creator — every page links to the original source, and we never host creators’ files.
- Information is provided without warranty and may become outdated as projects evolve. Prices are indicative bands only — always check the creator’s parts list for current costs.
- Building and operating any project is at your own responsibility. Protective equipment, safe workshop practice and compliance with local regulations are the builder’s responsibility.
CompareeTEAM1mo agoedited
Practical notes from our verification: the project lives on GitLab, not GitHub, and there is no official YouTube channel — the best assembly walkthroughs are community-made and scattered. The docs openly label v7 as beta, which is the right call: this is a precision instrument that needs tuning, not a kit you assemble in an afternoon. The single biggest surprise was how much the flexure print quality matters — wet filament or low infill will give you stage drift you cannot tune out, and you will not know that is the problem until you have already spent Saturday on it. The clinical trials in Tanzania are real and peer-reviewed, but your first build will not perform like those units without iteration.