Solution · Grinding Process Monitoring
Grinding, understood pass by pass.
Grinding is where fast data pays off: contact is short, burn happens quickly, and the wheel changes with every pass. Toleron is developed on grinding machines, and its first production model assesses the condition of the grinding wheel.
The problem
What you are up against.
- 01
When to dress is decided by habit, by a stylus measurement or by a burnt part — rarely by the process itself.
- 02
Grinding strategy parameters live in an expert's head or a spreadsheet.
- 03
The most telling signals only exist while the wheel is in contact — and they are too fast for ordinary data collection.
The approach
Wheel condition, pass by pass
On the AI box, Toleron calculates a sharpness value for every pass in which the wheel is actually in contact. A decision step combines it with the operator's own measurements and recommends when to dress.
Strategies from grinding expertise
Grind Master integrates the KSF GrindingExpert service. You enter material group, hardness, wheel specification, roughness and form accuracy; you get dressing parameters and a step-by-step strategy for roughing, finishing, fine finishing and spark-out — with feed ranges, infeeds, stroke counts, removal rate with warning levels, contact time and burn risk. A separate dressing module answers the dressing question on its own, with only the inputs it needs. Every recommendation is saved to a history.
A recording of every pass
Every pass can be recorded in full detail — sensors, current and controller data — started automatically at the machine, so the archive holds the contact and nothing else. The result is a record per pass that grows with production and becomes the raw material for the next model.
What we use
- Additional sensors at the grinding machine
- Spindle current and load
- The machine's own signal for each pass
- Dressing events
What you get
- A sharpness value per pass and a dressing state, visible as curves
- Grinding and dressing strategies with warning levels, saved to a history
- Full recordings of every pass for process engineering
- A growing, well-aligned data set for the next model
Tested on
- LabA.Haas Multigrind CB — tool grinding, Sinumerik 840D sl
- LabZiersch surface grinder — no controller connection, additional sensors only
- LabKSF GrindingExpert — surface and pendulum grinding strategies
Tested on the lab machines in Tuttlingen — proof that it works, not customer references.
Go deeper
Related
AI Models & Assistant
AI models that run at the machine and are tested before they go live — plus an assistant that answers from your own documents.
Learn moreCut Detection
Evaluates only the time the tool is actually cutting — not the minutes in between.
Learn moreRetrofit Sensors
Retrofitted sensors capture what the controller does not report — in full detail, on old and new machines.
Learn moreNext step
Bring grinding process monitoring to your machines.
We start with one machine and the signals it already has. Tell us the controller and the question, and we will tell you what the first weeks look like.