Lab and Research Instrumentation for Universities and R&D Labs

Who builds custom lab instrumentation for university research?

inteRFace R&D designs and builds custom measurement rigs and lab instrumentation for university research groups, student and faculty projects, and R&D labs in companies. We design the electronics, write the acquisition firmware and software, build the fixtures that hold the experiment still, and deliver a rig with a written measurement procedure. We are an engineering firm in Silicon Valley with 150+ products delivered and 200+ boards shipped, working with teams across the United States, and the first conversation is a free 30-minute call.

The typical project is one where the science is clear and the apparatus is not: a group knows what it wants to measure but has no electrical engineer to design the front end, a student has inherited a rig nobody can explain, or a lab wants to reproduce a published setup and the methods section leaves out the parts that matter.

What kinds of research rigs do you build?

• Energy-harvesting measurement apparatus: triboelectric, piezoelectric, electromagnetic and thermoelectric harvesters, with controlled excitation and the electrical measurement around them.
• High-voltage and low-current measurement front ends, from hundreds of volts down to nanoamps, with guarding and the right input impedance.
• Data acquisition and logging systems: sensors, signal conditioning, ADCs, timestamps and software that saves raw data with its metadata.
• Instrument automation: bench supplies, meters, scopes and source-measure units controlled over SCPI from Python, so a sweep runs unattended and the same way every time.
• Test fixtures and mechanical setups: holders, alignment jigs and enclosures, printed or machined, so the sample sits in the same place for every run.
• Custom boards for the experiment: sensor interfaces, drivers, power, isolation and a microcontroller or Linux board to run it.

How do you measure high voltage and very low current in the same rig?

By designing the measurement so the instrument does not change what it measures. Many harvesters, especially triboelectric ones, produce hundreds of volts at nanoamps to microamps from a very high source impedance, so an ordinary 10-megohm meter or scope probe loads the source and reports a voltage that was never there. The rig therefore uses an input matched to the source: an electrometer-class input or a high-impedance divider whose loading is known and corrected for, a charge measurement through a known capacitor, or a transimpedance stage for current. Around it go the details that decide whether nanoamp readings mean anything: guarding, low-leakage insulation and cabling, a single grounding point, shielding from mains pickup, and a measured leakage baseline recorded before every run.

High voltage also means safety design: interlocks, current-limited sources, discharge paths and labeling, because a student will operate the rig alone at some point. For a first sizing of a divider ratio, the free Voltage Divider Calculator picks standard resistor pairs and shows the output impedance the measuring input will see.

How do you measure the output power of an energy harvester?

With a load sweep under controlled excitation, not a single open-circuit voltage. The harvester is driven by a repeatable mechanical, thermal or vibration source, its output is measured across a range of load resistances to find the load that delivers peak power, and power is computed from the voltage and current actually measured on that load. When the application stores energy, the rig also measures charging of a capacitor or battery through the rectifier, because that, not the peak instantaneous power, is what a device downstream can use. Excitation amplitude, frequency, temperature and humidity are logged next to every electrical reading, since they often move the result more than the harvester design does.

What data acquisition and logging do you set up?

The simplest system that captures what the experiment needs at the resolution and rate it needs. Sometimes that is your existing bench instruments automated from a Python script; sometimes a commercial DAQ; sometimes a custom board with an ADC and a microcontroller, when the signal needs conditioning no off-the-shelf unit provides or the rig has to run outside the lab. In every case the output is raw data with timestamps, units, calibration constants and run conditions stored together, in a format your analysis code can read, so a figure in a paper can be traced back to the files it came from.

Can you turn the setup in a paper into a working bench rig?

Yes, and it is one of the most useful things an engineering team can do for a lab. We read the methods section, list what it specifies and what it leaves out (instrument models, input impedance, excitation details, grounding, sample mounting), propose the missing pieces with the reasoning written down, build the rig, and check it against a figure the paper reports before you trust it with new samples. Where the paper's setup has a measurement problem, such as a probe loading a high-impedance source, we tell you and show the corrected measurement next to the original method.

How do you work with student and faculty projects?

In phases sized to a research budget, with the student kept in the loop. A typical project starts with a short design review of the experiment, then the rig is built in stages so each one produces data you can use, and it ends with a handover session and documentation the next student can run without us. Students can sit in design reviews and bring-up sessions on video, so the knowledge stays in the group. If your institution pays through a purchase order or needs particular paperwork, tell us on the first call and the quote will be prepared accordingly. Fixtures and enclosures can come from our 3D printing and mechanical design work; custom boards follow our PCB design process, and acquisition code our firmware development practice.

How much does a custom research rig cost?

Priced in engineering hours; you get a written quote with phases before any work starts. The range is wide because rigs are: automating instruments you already own is a small job, while a rig with custom high-voltage or low-current electronics, firmware, fixtures and software is a full project. For calibration from figures already published on this site, a simple 2-layer board typically runs 15-50 engineering hours of design, a standard 4-layer mixed-signal board 50-140, and a Raspberry Pi 5 GPIO adapter board, schematic through released production files, took about 55 hours. The drivers are the measurement range and accuracy you need, how much custom electronics the rig needs, safety requirements and the documentation you want. The free project cost estimator gives a planning range, and a free call turns it into a scoped quote.

How does a lab instrumentation project start?

With a free 30-minute call with a senior engineer: book a time and bring what you want to measure, the ranges and accuracy you need, the instruments the lab already has, and the paper or sketch of the setup if there is one. A mutual NDA can be signed before unpublished work is discussed, and the lab keeps 100% of the IP and the data. We are based in Silicon Valley and work with teams across the United States, mostly remotely, with rigs and boards shipped by carrier and in-person days in the Bay Area by arrangement.