For labs relying on precise fluid dispensing, traditional volumetric or pressure-based systems often fall short—especially when switching media or dealing with variable temperatures. Even small changes in viscosity can lead oreientatto waste, inconsistent results, or failed protocols.
At Festo’s Research Hub Boston, we’ve developed a closed-loop, pressure-over-liquid dispensing solution that adjusts in real time using machine learning—offering consistency and confidence with every shot.
Most fluid dispensing systems assume viscosity stays constant—but it rarely does. Whether due to ambient shifts, material changes, or lot-to-lot variation, this single assumption can create substantial errors in high-precision environments like life sciences, diagnostics, and automated lab workflows.
Until now, pressure-over-liquid systems couldn’t self-correct. That gap meant teams had to rely on calibration, conservative dosing, or manual adjustments to stay accurate.
Our new approach uses sensor signals to detect how a fluid behaves during dispensing—effectively inferring its viscosity in real time. A machine learning model, trained on a wide range of viscosities and behaviors, adjusts the valve opening time dynamically for each shot.
This closed-loop system ensures the volume delivered matches the target precisely, even as conditions change.
For teams working in high-throughput environments or complex fluid systems, this unlocks new levels of control and flexibility.
We’re actively looking for pilot partners to test this closed-loop dispensing innovation in real-world environments ahead of its broader release.