Contact Us | Working at ATS | Investor Relations
All but very simple automated processes use some form of feedback. Familiar forms of feedback include motor encoders, thermocouples, load cells, LVDTs (linear variable differential transformer). For many processes, a single form of feedback is sufficient to maintain good control. In others, several independent variables may be monitored to a target range to ensure a successful process.
In more complex processes, the parameters are not independent—each parameter must vary precisely with respect to the other, often maintaining a complex relationship over the time period of the process. One example of this is metal forming.
When sheet material is formed in a die, the displacement and force vary as the material exhibits elastic and plastic deformation, particular geometry collapses, and the die bottoms out. By measuring the force and displacement over a series of parts, and then analyzing that data for both good and bad parts, a signature for the process emerges.
Once the signature limits are established, the equipment can monitor the variables of interest and use that signature to determine whether the process was successful and either accept or reject the part accordingly. The equipment may also use the signature in real time, to alter one or more of the variables, to compensate for material or tolerance variations and increase the overall yield.
We have implemented signature analysis on many successful projects, integrating world-class data collection and analysis units into the automation.