arrow_back
All case studies
science
Medical / Bioprocessing
Protein Filtration Control Library
Bioprocessing TFF system
The Windows control library behind a tangential-flow-filtration system for a US bioprocessing-equipment manufacturer — real-time monitoring and control of 50+ parameters, with a PID rewrite that reached steady state 3× faster at 2× lower error.
50+
Parameters in real time
3× faster
To steady state
2× lower
Steady-state error
- Client
- A US-based bioprocessing-equipment manufacturer (name withheld — NDA)
- Delivery
- Delivered by our founder at Lattice Innovations
- Role
- Firmware/software developer — control library, API and control algorithm
- Timeline
- 2019
Stack
Java (multi-threaded) Windows PID control Real-time instrument comms
help The problem
Tangential-flow filtration concentrates and purifies biologics — proteins, vaccines, therapeutics. The system runs filtration, diafiltration and intermediate steps as a profiled process, not a single fixed operation, while continuously watching flow and pressure at both input and output. The instrument needed a control layer the product's backend and frontend could drive.
build What we built
- A multi-threaded Java control library for Windows that talks to the instrument and monitors and controls 50+ system parameters in real time — the engine the rest of the product's UI is built on.
- A clean API surface exposed to the front end, so operators drive the machine through a GUI without touching the low-level instrument protocol.
- Process profiling for filtration and diafiltration, sequencing the stages that turn raw fluid into a purified product.
- Continuous flow and pressure monitoring at input and output — the feedback that keeps the process inside safe operating bounds.
- An advanced PID control algorithm proposed and implemented to replace the original loop: steady state reached 3× faster with steady-state error cut 2×.
check_circle The result
- done Delivered the control library the TFF product relied on for monitoring and control, with measurable improvement over the original algorithm.
- done The same instrument line was later used in COVID-19 testing.