Case Study: Johnson Controls achieves faster, higher‑quality magnetic‑bearing chiller controller development with MathWorks

Johnson Controls Accelerates Industrial Controller Development for Magnetic-Bearing Centrifugal Liquid Chillers

Johnson Controls faced a growing design and development challenge for its YORK Magnetic Centrifugal Chiller (YMC2): an increasingly complex controller that was slow and costly to update using handwritten C code and repeated bench and chiller tests. To accelerate development and improve reliability, Johnson Controls adopted MathWorks' Model‑Based Design approach, using MATLAB, Simulink, Stateflow, and Embedded Coder to reengineer the chiller’s capacity control logic.

Using MathWorks tools, the team modeled both controller and plant, ran extensive closed‑loop simulations, and generated about 500,000 lines of C code for processor‑in‑the‑loop and hardware‑in‑the‑loop testing. The results: design iterations fell from months to days, the controller reaches set points faster and is far more stable (99% of customers run the default configuration with no adjustments), most HIL issues were unrelated to the controller, and the generated code required no rework—significantly accelerating development and production deployment.

Johnson Controls

Curtis Crane

Senior Staff Engineer