YOU CAN BUILD THE 8-CHANNEL VIBRATION RIG LABS PAY 500 DOLLARS PER CHANNEL FOR
You can build the eight-channel vibration recorder labs pay thousands for — validated against commercial gear, for 220 dollars.
by Oliver Maximilian Zobel, Johannes Maierhofer, Andreas Köstler, Daniel J. Rixen
ScienceOpen-hardware
Built withESP32
- difficulty
- ●●●●○
- time
- a long weekend
- license
- CC-BY-4.0 (hardware), MIT (software)
- repo
- repo ACTIVELY DEVELOPED ON GITLAB (PAPER RELEASE FROZEN ON ZENODO)0 stars
●●●●○ · a long weekend · CC-BY-4.0 (hardware), MIT (software) · 0 stars · repo ACTIVELY DEVELOPED ON GITLAB (PAPER RELEASE FROZEN ON ZENODO)
WHAT YOU’LL NEED
Jump to section
COMPAREE VERDICT
This is a research-grade vibration measurement system for people who already know what experimental modal analysis is and why commercial hardware costs what it does. The build is validated: the authors compared it against a Siemens LMS SCADAS system on a stiff aluminium beam excited by an automatic impact hammer, matching frequency response functions and mode shapes, with no significant differences up to about 3 kHz. The 220 dollar figure (about 200 euros) is real, and the paper notes that IEPE channels on National Instruments' CompactDAQ cost approximately 500 dollars per channel. The upgrade from the previous version is the AD7606C-18 ADC, which gives higher resolution, eight simultaneous channels, oversampling and software-adjustable input ranges, plus SD-card caching on the ESP32-S3 to hit 36 kHz on all channels at once. The Python GUI handles acquisition and conversion; for modal analysis the authors use the open-source pyFRF and pyFBS packages. The single biggest trap is that this is not an introductory project — you need to understand IEPE sensor interfacing, signal conditioning, and what you are actually measuring before you start soldering. If you are in a structural dynamics lab and tired of waiting for budget approval on a commercial DAQ, this is exactly what it claims to be. If you have never done modal testing before, start with something simpler.
IN THE REPO
GOOD TO KNOW
- —Full HardwareX paper with design files, BOM, assembly instructions and validation data.
- —Gerbers, firmware and Python software package on Zenodo (DOI 10.5281/zenodo.13763226).
- —Hardware CC-BY 4.0, software MIT — both allow commercial use with attribution.
- —OSHWA certified DE000150.
- —The paper's frozen release is on Zenodo; ongoing development (hardware, firmware, GUI) lives in the public GitLab group oasis-acquisition, where the firmware has since moved to a newer ESP-IDF release line.
- —Requires IEPE accelerometers (not included in the 220 dollar figure) and assumes familiarity with modal analysis concepts.
Parts to buy
6 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 end to end (Section 2 covers the hardware and firmware, Section 4 is the BOM, Section 5 is assembly, Section 6 is operation and Section 7 is validation.)
- 2.Download the design files from Zenodo (Gerbers, firmware, Python package, CAD files and calibration data are all here.)
- 3.Order the PCB and components(The BOM in the paper lists every part with supplier codes. Key components: AD7606C-18 ADC, ESP32-S3, eight LT3092 current sources, MC34063 boost converter and Amphenol BNC connectors.)
- 4.Assemble the board following Section 5(Flash the ESP32-S3 first, then solder the ADC, boost converter and current sources, and calibrate each channel's IEPE current (Section 5.5).)
KNOWN ISSUES
- The 220 dollar cost covers the acquisition board only. IEPE accelerometers, cables and an impact hammer are not included, and you need at least one sensor to measure anything.
- The board supplies the IEPE excitation current itself. Until the board is fully checked, use a shorted BNC plug as the load instead of a real sensor, so an assembly mistake cannot damage an expensive accelerometer.
- Because of the size of some components, the authors recommend ordering the PCB at least partially assembled from JLCPCB using their manufacturing files; you then solder the ADC, current sources and BNC connectors yourself.
- The OASIS software only records and converts raw voltages. To get FRFs, mode shapes or MAC values you need separate tools (the authors use the open-source pyFRF and pyFBS packages) and enough modal-analysis background to use them.
- Calibration matters. Section 5.5 walks through setting each channel's IEPE current (4 mA recommended) with an ammeter — skip it and your sensors may be under- or over-powered.
- 36 kHz on all eight channels is the upper limit the authors found. Push the sample rate or oversampling factor too high and the firmware aborts the run with an error, so format the microSD card as FAT32 with a 32 kB block size as the paper recommends.
Can I use this for audio or general-purpose data acquisition?
Technically yes, but the IEPE front-end is optimised for accelerometers (AC-coupled, high-pass filtered). For DC signals or non-IEPE sensors, you would need to bypass or modify the front-end.
How does it compare to the previous OASIS version?
OASIS-UROS moves to the AD7606C-18 ADC (18-bit, eight simultaneous channels, oversampling, software-selectable ranges), caches data on an SD card instead of streaming it, and improves the IEPE front-end and power supply, reaching up to 36 kHz on all eight channels. One thing it lost: the original board's WiFi sampling is not yet available in the UROS firmware.
What accelerometers should I buy?
The paper used PCB Piezotronics 356A03 triaxial sensors for validation. Standard IEPE accelerometers work, with 4 mA per channel recommended, and you pick a range from ±2.5 V to ±12.5 V per channel. The inputs are always AC-coupled (0.8 Hz high-pass), so it is not suited to signals below a few hertz.
Is the calibration procedure mandatory?
Yes. The paper's calibration step sets the IEPE excitation current for each channel, and the board should not be used before it is done. Sensor sensitivity still comes from your accelerometer's datasheet.
Can I use this for operational modal analysis on a running machine?
The hardware can record long continuous runs because data is cached to the SD card, but the OASIS software only handles acquisition. For modal identification you need separate tools — the authors use the open-source pyFRF and pyFBS packages, and the paper only validates impact-hammer EMA.
Community builds
No community builds yet — be the first, we feature the best ones.
Discussion1
FROM THE COMPAREE TEAM
The paper validated this board against a Siemens LMS SCADAS on an impact-hammer test. If you have used both commercial and DIY DAQ hardware for modal testing, what was the real difference — not the spec sheet, the thing that actually mattered in the lab?
Oliver Maximilian Zobel, Johannes Maierhofer, Andreas Köstler, Daniel J. Rixen
Three of the authors are researchers at the Chair of Applied Mechanics, TUM School of Engineering and Design, Technical University of Munich; Johannes Maierhofer runs Maierhofer-Technology. They built OASIS-UROS as an open, low-cost IEPE acquisition system for structural dynamics research and teaching, upgrading their earlier OASIS board.
Keep building — related projects
Same category or platform, picked by how close they are to this build.
FULL CADYou can build your own super-resolution microscope from magnetic LEGO-like cubes
ScienceFor advanced builders
FULL CADYou can build the machine that spins thread from scratch
ScienceFor expert builders
FULL CADOpenAstroTracker — 3D printable astrophotography tracking mount for DSLR cameras
ScienceFor advanced builders
NO CAD FILESTinyGS — open source global satellite ground station network
Open-hardwareFor beginners
FULL CADYou can build a laser scanning microscope from a dead Blu-ray player
ScienceFor advanced builders
FULL CADElectrochemistry labs pay thousands for a rotating disc electrode. This one is under 100 dollars
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.
CompareeTEAM23d agoedited
We verified the HardwareX paper, the Zenodo repository and the OSHWA certification (UID DE000150). The KiCad hardware project, JLCPCB manufacturing files, Arduino firmware, the Python acquisition GUI and a printable bottom case are all published on Zenodo, and development continues in the authors' public GitLab group (oasis-acquisition). The hardware cost of about 220 dollars is stated in the paper, which also quotes roughly 500 dollars per IEPE channel for one commercial system (NI CompactDAQ). Validation was done on an aluminium beam with an automatic impact hammer and seven triaxial accelerometers, compared against a Siemens LMS SCADAS. One step you cannot skip: before first use, the constant current source of each channel must be calibrated with an ammeter. Correction (4 October 2026): we re-checked this page line by line against the paper, the design files and the authors' GitLab, and fixed errors in earlier versions.