Reducing physical testing by making optimized design and material choices earlier via simulation results

Making the perfect cup of coffee

Coffee plays a big part in many people’s lives, with millions having their own coffee machine at home to ensure they can have their favorite cup whenever they want.

Coffee drinkers, however, expect the same consistent taste every time, and at a reasonable price.

BSH Home Appliances (BSH), a subsidiary of Bosch, is one of the world’s leading home appliance manufacturers. It has an extensive portfolio of small and large devices that perform tasks from cooking, dishwashing, cooling, freezing, washing and drying.

However, defining what makes good coffee is difficult. “It’s a complex system and it’s not always clear which inputs, like the amount of coffee, the pressure profile or the water temperature, are the most important for flavor,” says Ronny Leonhardt, research engineer at Bosch Corporate Research. “Testing all of these can be extremely time consuming and costly.”

On top of getting the taste and price right, BSH also needs to consider the environmental impact of each cup of coffee. Most countries’ regulations require manufacturers to clearly display energy ratings on home appliances, and BSH has its own commitments for sustainability, which every product must adhere to.

To meet these challenges, BSH has added simulation to its development processes. To model these complex systems and streamline optimization, BSH engineers use Simcenter™ STAR-CCM+™ software and Simcenter Amesim™ software, which are part of the Siemens Xcelerator business platform of software, hardware and services.

Achieving a deeper understanding of physics

A coffee machine processes hot water, coffee and often milk for various milk-based beverages, each of which presents physical challenges and requires a preparation process that must remain accurate and stable.

BSH Development Engineer, Christian Nerbl, says the company introduced simulation to get the necessary insights into the physics of coffee machines. “With normal testing you can calculate some aspects of heat transfer but not transient heat transfer or time-dependent physical processes,” says Nerbl. “Simulation gives us a much deeper understanding of the physics involved and how the machine behaves.”

Nerbl explains that simulation helps BSH put the customer at the heart of development. “We put a lot of effort into gathering and acting on feedback,” says Nerbl. “Customers don’t just want a quick coffee, they want a good, consistent coffee. By modeling and simulating the machine, we can test different parameters such as heating power, pressure levels, temperature profiles and how both the control strategy and the parameters are set and timed. This enables us to optimize the system to save energy and cost and deliver the same flavor and quality in every cup of coffee.”

This optimization helps improve overall sustainability as it reduces the amount of coffee each cup requires without compromising on taste. Even just a few grams per cup amounts to a significant impact when multiplied across BSH’s large annual production volumes.

Optimizing guided by simulation

Each process of a coffee machine involves a large number of parameters. Simulation allows engineers to calculate the results of adjusting these parameters much faster than testing on a physical machine. Crucially, it helps them decide if a parameter should be adjusted one way or the other, which saves a lot of time before moving on to physical testing.

BSH uses a combination of simulation solutions and methods, like computational fluid dynamics (CFD), to achieve its desired results.

“We begin by building a CFD model in Simcenter STAR-CCM+,” says Nerbl. “This allows us to analyze key fluid and thermal behaviors in different subcomponents of the machine, such as the brewing unit. We incorporate this knowledge into a Simcenter Amesim model and run design explorations to analyze different strategies, optimize each process and tune the parameters of the control module.”

Simulation is also a vital tool later in the development cycle when the design needs modifications. “Sometimes for production reasons we might need to change a component such as the length of a hose or the material used,” says Nerbl. “Even the smallest changes can have an impact on performance, so we need to understand how this will affect heat transfer or thermal losses. With simulation, we can quickly calculate what the effect will be before the changes are approved.”

Nerbl notes that the flexibility and open-access design of Simcenter was essential for success. “We had to connect the physics of Simcenter to the control world in our department,” says Nerbl. “We weren’t experienced with the solver or with co-simulation technology, but the fact that Simcenter is designed to work with other products made it easy to integrate with our existing tool chain. We also had great support from Siemens to get everything working just as we needed.”

By using Simcenter solutions, BSH was able to reduce physical testing by making optimized design and material choices earlier via simulation results and decrease the time it takes to the convert the power supply from European Union (EU) to United States (U.S.) regulations, from weeks to days.

Answering complex questions quickly

Leonhardt says that their engineering team still needs testing to validate designs, but now they can reach that point faster. He explains that, in this instance, the parameter case study was carried out as a change proposal. Crucially, they could skip some parameter combinations after simulation delivered deeper insights.

“Simulation helps us optimize designs sooner and reduce the number of physical tests we need for validation,” says Leonhardt. “This saves us a significant amount of time and money. These savings will increase for future projects as simulation becomes further established in our development process.”

“With Simcenter Amesim, we can answer some questions almost immediately,” says Nerbl. “We use a combination of standard Simcenter Amesim libraries together with our own custom sub-models. We configure the model, run a design of experiments or parameter study and then postprocess the results. For more complex analyses we may need to combine or adapt different models, which takes more time, but is still much faster than physical testing.

“We had to adjust several parameters for the pump and heater, but we calculated the ideal configuration with Simcenter Amesim, and it worked on the first iteration of physical testing. This meant that the work was completed in just a few days rather than several weeks.”

Embracing the power of simulation

Nerbl has no doubt about the importance of simulation now and in the future. “If you want to stay competitive and keep developing your skills, you have to embrace simulation,” says Nerbl. “Leveraging the simulation capabilities within Simcenter has made my work more efficient and meant that I don’t have to wait for a physical machine to be modified – I can just modify the virtual model myself. I can do all the preparation while waiting for the hardware changes and then directly load the software and parameters to run the final tests.”

Leonhardt agrees and believes optimization and parameter studies will become increasingly important. “It’s about building stable and robust workflows for optimization,” says Leonhardt. “Especially coupling, robust optimizers and multi-objective optimization. All these elements need to work together reliably and it’s an area where I see a lot of progress happening in the future.”