YOU CAN BUILD THE 4,750-DOLLAR MACHINE THAT MAKES RAPID TEST STRIPS, FOR 200 DOLLARS
Build the machine that makes rapid test strips for 200 dollars instead of 4,750 dollars.
by Anh Phuc Hoang Le, Quang Lam Nguyen, Bao Hoai Pham, Thien Hoang Minh Cao, Toi Van Vo, Khon Huynh, Huong Thi Thanh Ha
ScienceOpen-hardware
Built withArduino
- difficulty
- ●●●●○
- time
- a weekend-plus
- license
- CC BY 4.0
- repo
- repo FINISHED0 stars
●●●●○ · a weekend-plus · CC BY 4.0 · 0 stars · repo FINISHED
WHAT YOU’LL NEED
Jump to section
COMPAREE VERDICT
SALAD is a lateral flow reagent dispenser — the machine that lays the antibody line across the membrane in every pregnancy test and covid rapid test you have ever used. A team at International University in Ho Chi Minh City benchmarked it directly against the Claremont Bio ALFRD commercial dispenser (listed at 4,750 dollars) and published the results: no significant difference in line width or antibody signal, for 200.61 dollars in parts. The design is two stepper motors on TB6600 drivers run by an Arduino Uno: one pushes a Hamilton syringe, the other drives a lead-screw actuator that carries the membrane under the needle like a conveyor. Timing and speed are set with potentiometers and shown on a small LCD. The structure is two off-the-shelf C-beam linear actuators joined with galvanized iron corners and a set of CNC-cut aluminium plates and frames. This is not a weekend build. You are assembling precision motion control, and the paper expects you to tune the settings yourself. The BOM is complete with vendors, the assembly instructions are illustrated, and the Arduino code is provided, but you are still building a piece of lab equipment from scratch. The single thing most likely to go wrong is uneven membrane travel: if the actuator does not move smoothly under the syringe tip, your line will be uneven, and the authors themselves note the conveyor stage needs further optimisation. This is for someone who already works in a lab that needs to make lateral flow assays and cannot justify five thousand dollars on a dispenser, or someone who wants to learn how diagnostic manufacturing works at the component level. It is not for someone who wants to make one test strip.
IN THE REPO
GOOD TO KNOW
- —Design files on Mendeley Data (doi.org/10.17632/sv5y27mzjb.2): Fusion 360/STL files for eight custom parts, made from CNC-cut aluminium in the original build, plus the Arduino code. The bill of materials with vendors and the illustrated build instructions are in the paper.
- —Published in HardwareX — peer-reviewed hardware paper with validation data against the commercial Claremont Bio dispenser.
- —No GitHub repository; files are dataset downloads from Mendeley. You will need to navigate an academic data repository, not clone a repo.
- —Licence is CC BY 4.0 — open for commercial use with attribution.
- —This is a research instrument. It is not a certified diagnostic manufacturing device and the paper does not claim it is.
- —You will need nitrocellulose membranes and antibody reagents to test it for real; the paper names the products it used, but they come from lab suppliers.
Parts to buy
9 itemsFrom our check of the build. Exact quantities and part numbers are in the creator’s BOM.
Can I build this?
Build at your own risk. Projects involve tools, electronics and sometimes mains voltage — follow the creator’s safety notes.
More builds like this
All projectsGallery
Start here
Navigation into the creator’s own docs — we don’t rewrite the guide, we route you to the source.
- 1.Read the HardwareX paper (The paper is the manual. It contains the assembly instructions, wiring diagrams, and validation data. You cannot build this without reading it.)
- 2.Download the design files from Mendeley Data (CAD files, BOM with part numbers, Arduino code, and assembly photos are all in the dataset. You will need a free Mendeley account to download.)
- 3.Source the BOM(The paper's BOM lists parts and vendors, mostly Vietnamese shops, so find local equivalents. Most parts are standard (NEMA 17 steppers, TB6600 drivers, Arduino Uno, T8 C-beam actuators). The specialised one is the Hamilton syringe (#HAM80075, 10 µL with a 26G needle). There are no limit switches; the authors list them only as a future improvement.)
- 4.Fabricate the frame and printed parts(Have the needle handle, stepper shelf tops, plate pusher, membrane plate, vertical frames and slide holders CNC-cut in aluminium from the dataset files, then join the two lead-screw actuators with galvanized iron corners and M4 screws.)
KNOWN ISSUES
- Real validation needs nitrocellulose membrane and antibodies. The paper lists the exact Invitrogen products and catalogue numbers, but they come from lab suppliers and are not cheap outside a lab. For first tests of line quality, the authors simply dispensed diluted food dye.
- The Mendeley dataset is not a GitHub repo. You download a zip file, extract it, and find the files. If you have never used an academic data repository, budget time to figure that out.
- Set both TB6600 drivers exactly as the paper does: microstep 4 (800 steps per revolution) and 1 A, which is switches ON-OFF-OFF for the first three and ON-OFF-ON for the last three. The authors picked this setting after testing for heat, speed and vibration.
- Smooth membrane travel matters. The membrane stage is a lead-screw actuator on V-wheels; if the wheels are loose or the screw binds, your line will be uneven. The authors themselves flag the conveyor stage as the part needing more optimisation.
- The Arduino code is provided but not commented line-by-line. You will need to read it to understand what it does, especially if you want to change dispense speed or line length.
- This is a research instrument. You cannot sell test strips made with it without going through regulatory approval, and the paper does not claim otherwise.
Can I actually make covid tests with this?
Technically yes, but legally no. You can dispense the antibody line, which is the hard part of making a lateral flow test, but you cannot sell diagnostic tests without regulatory approval. This is for research and development in a lab setting.
Where do I get nitrocellulose membranes?
The paper names what it used: Invitrogen 0.45 µm nitrocellulose membrane (catalogue #88018), with Whatman Standard 14 as the sample pad and Whatman CF4 as the absorption pad. These come from lab suppliers, not hobby shops.
How does this compare to the Claremont Bio dispenser?
The paper ran them side by side at the same dispensing rate. There was no significant difference in line width or in mean signal intensity, and SALAD strips showed lower background and a slightly better signal-to-noise ratio after lateral flow. The weak spot was line length: SALAD's varied far more than the Claremont's, which the authors put down to an unsteady conveyor. Setup took 5-10 minutes on SALAD versus 15-20 on the Claremont, with the same 1-2 minute run.
Do I need a laser cutter?
No laser cutter is needed. The custom parts (needle handle, stepper shelf tops, plate pusher, membrane plate, vertical frames and sliding plate holders) were CNC-cut from aluminium in the original build, about 20 dollars from a local shop, and their design files are in the Mendeley dataset. Printing them in plastic is untested by the authors.
What happens if I get the stepper wiring wrong?
Usually the motor will not move, will run backwards, or will skip steps. To reverse direction, swap the two wires of one coil (A+ with A-). Otherwise check the wiring against the paper's schematic (Fig. 3). Always disconnect the 12 V supply before rewiring, because changing motor or driver connections live can damage the driver.
Community builds
No community builds yet — be the first, we feature the best ones.
Discussion1
FROM THE COMPAREE TEAM
About 200 dollars in parts versus a 4,750-dollar Claremont dispenser — and the paper found no significant difference in line width. If you were building this for a lab, what would you tune first: the conveyor stage the authors flagged, or the syringe dispense rate?
Anh Phuc Hoang Le, Quang Lam Nguyen, Bao Hoai Pham, Thien Hoang Minh Cao, Toi Van Vo, Khon Huynh, Huong Thi Thanh Ha
A team at the School of Biomedical Engineering, International University, Vietnam National University in Ho Chi Minh City, built SALAD for point-of-care diagnostics labs that cannot afford commercial dispensers. They published it in HardwareX with full validation data against the 4,750-dollar Claremont Bio machine.
Keep building — related projects
Same category or platform, picked by how close they are to this build.
FULL CADYou can build the 15,000-dollar laser rig that measures flow, for 520 dollars
ScienceFor advanced builders
FULL CADYou can build the €1,330 vacuum gauge for €70
ScienceFor advanced builders
NO CAD FILESYou can build a 16-bit computer out of the switches that ran telephone exchanges
Open-hardwareFor expert builders
FULL CADOpenAstroTracker — 3D printable astrophotography tracking mount for DSLR cameras
ScienceFor advanced builders
FULL CADYou can build the machine that spins thread from scratch
ScienceFor expert builders
FULL CADYou can turn a cheap resin printer into a bioprinter for €500
ScienceFor expert builders
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.
CompareeTEAM18d agoedited
Practical notes from our verification: this is a clean academic hardware publication in HardwareX under CC BY 4.0, with the design files in a Mendeley Data repository (doi 10.17632/sv5y27mzjb.2) — STL files for the needle handle, stepper shelf, membrane plate, vertical frames and slide holder, plus Arduino code and a full bill of materials with suppliers that totals about 200 dollars. The electronics are simple: an Arduino UNO, two NEMA 17 steppers on TB6600 drivers, two T8 lead-screw actuators, an I2C LCD and four potentiometers for settings, with a Hamilton syringe as the single priciest part. The authors did not just claim equivalence: they dispensed the same anti-sheep HRP antibody with SALAD and with the commercial Claremont ALFRD and found no significant difference in line width or mean signal, with lower background noise and slightly better SNR on SALAD strips. They also say plainly that the conveyor stage needs optimisation for stability. No GitHub repo, so you download the files from Mendeley, and you still need your own source of nitrocellulose membranes and antibodies. Correction (4 October 2026): we re-checked this page line by line against the project's own repository, documentation and videos, and fixed errors in earlier versions.