YOU CAN BUILD THE 1,000-DOLLAR SENSOR THAT WATCHES A TREE BREATHE, FOR 260 DOLLARS
A research-grade sensor that resolves half-micrometre stem movement with zero friction, published by Oregon State and built for about a quarter of the commercial price.
by Cameron Clonch, Mark Huynh, Bryson Goto, Alexander Levin, John Selker, and Chet Udell
ScienceGarden
Built with3D printing
- difficulty
- ●●●●○
- time
- a weekend-plus
- license
- GPL-3.0
- repo
- repo ACTIVE5 stars
●●●●○ · a weekend-plus · GPL-3.0 · 5 stars · repo ACTIVE
WHAT YOU’LL NEED
Jump to section
COMPAREE VERDICT
This is a published research instrument that measures stem diameter changes down to half a micrometre across weeks in the field, removing the friction problem that plagues band and contact dendrometers. The design is genuinely clever: a multipole magnetic strip moves past an AS5311 linear magnetic encoder across a gap of about 0.3 mm, so nothing in the measurement path rubs. The frame is carbon fibre, chosen for its near-zero thermal expansion, cut and drilled into parts held together with screws, with a spring keeping the grip on the stem; the electronics go into a Pelican 1120 case mounted nearby, and one 2000 mAh battery is estimated to last about six months. The paper puts commercial dendrometers at around 1,000 dollars each against roughly 260 dollars for this one. The catch is that this is an instrument build, not a weekend electronics project. There is no hand-holding beyond the paper: the repository gives you firmware and PCB files and the paper gives you drawings and machining steps, but you are stitching those together yourself. You will need machine-shop access — a bandsaw, a milling machine with a drill chuck and a tap set — to make the carbon fibre parts, and you will need to source a long parts list including the AS5311, a matching 2 mm-pole magnetic strip, an Adafruit Feather M0 and the OPEnS Hypnos board. The single part most likely to go wrong is the sensor and magnet pairing: the AS5311 needs a strip with 2.0 mm pole pairs and 10 µm/m resolution or better. The published version clamps stems 23 to 40 mm across and was field-tested on a grapevine. If you are setting up a multi-site vineyard or orchard study and need ten dendrometers, this saves thousands. If you want to see whether your garden apple tree is water-stressed, this is a hard way to find out.
IN THE REPO
GOOD TO KNOW
- —Full HardwareX paper with design rationale, validation data, and comparison to commercial units.
- —CAD files on Zenodo (carbon fibre frame flat geometry), Arduino firmware on GitHub.
- —Bill of materials with part numbers and estimated unit cost of 260 dollars.
- —Licences are GNU GPL v3 (firmware) and CERN OHL-S v2 (hardware) — both allow commercial use with share-alike.
- —No step-by-step assembly guide; you are expected to read a 13-page academic paper and work from the CAD.
- —The AS5311 magnetic position sensor is the single critical part and costs about 17 dollars; substituting it will ruin the measurement.
Parts to buy
10 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 front-to-back, especially the Design Files Summary and Bill of Materials table. (The paper is the build guide; there is no separate manual.)
- 2.Download CAD files from Zenodo and firmware from GitHub. (Zenodo DOI is 10.5281/zenodo.4482948 (linked in the paper).)
- 3.Source the AS5311 sensor and verify it is the AMS AS5311 with 28-TSSOP package before ordering anything else.(This part is make-or-break; substitutes will not work.)
- 4.Machine the carbon fibre parts from the part drawings — or hand the drawings to a machine shop.(Material is 1/16-inch and 1/4-inch carbon fibre sheet; it dulls tools and fractures easily, so cut slowly and drill with pecking.)
KNOWN ISSUES
- The AS5311 is a specific 12-bit magnetic position sensor; buying a generic Hall effect sensor or a different AMS part will give you a useless reading.
- There is no step-by-step assembly guide. The paper shows exploded views and the CAD shows geometry, but you are expected to figure out the order yourself.
- The AS5311 has to sit about 0.3 mm from the magnetic strip. If the sensor plate or magnet mount shifts, readings become unreliable — use the LED indication system after assembly and after any knock to confirm the spacing.
- The Pelican 1120 case is not optional — this thing sits on a tree outdoors for months and needs to be weather-sealed and rodent-proof.
- The paper's 260-dollar figure is the per-unit share of each part. Many parts are sold in packs of 5, 10, 20 or 100, so building a single unit means paying for the whole packs — your first unit will cost noticeably more, and the economics only work if you build several.
- Installation is checked with the built-in LED indicator, which shows whether the magnet sits at the right distance from the sensor. Re-check it after strong winds, crop work, or every couple of weeks, as the authors advise, and expect occasional bad readings that jump to 0 or 4095, which the authors filtered out of their data.
Can I use a different microcontroller?
Yes, as long as it can read SPI (for the AS5311) and I²C (for the SHT30D) and log to SD. The Feather M0 is convenient because it has all three and fits in the case, but an ESP32 or Teensy would work if you rewrite the firmware.
How often does it need new batteries?
The paper estimates about six months on one 3.7 V 2000 mAh LiPo battery when logging every 15 minutes (its power-budget calculation gives 190.7 days; a summary list says 277 days). The authors still advise a quick check with the LED indicator every couple of weeks and after strong winds or crop work.
Can I use it on a plant stem instead of a tree trunk?
It was designed for woody stems such as grapevines: the published version fits stems about 23 to 40 mm across, and it was field-tested on a 26 mm vine. The authors say it can be scaled for other sizes by widening the rod spacing and lengthening the rods. Soft herbaceous stems were not tested, so the spring force would be your experiment.
Do I need to recalibrate it every season?
The paper describes no recalibration routine. In lab tests from 10 to 42 °C and in the field next to a control unit on a Pyrex cylinder, the device was mostly insensitive to temperature, with the control drifting at most about 30 µm a day. Comparing against a control unit like that is the simplest way to check your own build over a season.
Community builds
No community builds yet — be the first, we feature the best ones.
Discussion1
FROM THE COMPAREE TEAM
The paper tracked daily stem swings of about 150 to 200 micrometres on a grapevine over a week, while the control unit on a Pyrex cylinder stayed within about 30 micrometres a day. Have you used contact or band dendrometers before, and did drift actually ruin your data?
Cameron Clonch, Mark Huynh, Bryson Goto, Alexander Levin, John Selker, and Chet Udell
Built at the Openly Published Environmental Sensing Lab and the Department of Horticulture at Oregon State University to measure plant water stress in the field at research-grade precision without research-grade budgets. The lab's focus is open instrumentation for environmental science.
Keep building — related projects
Same category or platform, picked by how close they are to this build.
FULL CADYou can weld the heated press that polymer labs test plastic with
WorkshopFor expert builders
FULL CADOpenAstroTracker — 3D printable astrophotography tracking mount for DSLR cameras
ScienceFor advanced builders
FULL CADA 300-dollar kit turns a decades-old lab microscope into an automated scanning system
ScienceFor intermediate builders
FULL CADYou can build an NMR machine - and it needs no liquid helium at all
ScienceFor expert builders
FULL CADYou can 3D print the 6,200-dollar lab pump for 362 dollars
ScienceFor intermediate builders
FULL CADBuild the robot that tests how much cream gets through your skin
ScienceFor advanced 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.
CompareeTEAM15d agoedited
Practical notes from our verification: this is a published HardwareX instrument with real field data, which is rare for open hardware at this price point. Validation covered lab temperature tests, a one-week deployment on a grapevine at Oregon State's Woodhall III Vineyard in October 2020 alongside a control unit on a Pyrex cylinder, and a comparison of the vine's daily swings against vapour pressure deficit. The GitHub repository holds the firmware plus PCB files and has kept receiving updates since the 2021 paper; the mechanical CAD is on Zenodo under CERN OHL-S v2. The single part you cannot substitute is the AS5311 magnetic linear encoder with its matching magnetic strip; a generic Hall sensor will not give you micrometre readings. The paper puts the cost at about 260 dollars per unit, built on an OPEnS Hypnos board with a custom PCB. Fabrication is machine-shop work: the carbon fibre parts are cut on a bandsaw, then drilled and tapped on a milling machine, and a few mounts are 3D printed. There is no assembly video. Against commercial dendrometers at around 1,000 dollars each, the real case is deploying many of them across a field or vineyard. 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.