The dynamic reference is called the baseline and tracks the RawCounts when a finger is absent. This helps in ensuring that gradual changes (environmental changes) like temperature, humidity etc. are rejected. There may be threshold settings related to the update of this dynamic reference as well which should also be carefully tuned such that the baseline tracks gradual changes properly but does not follow the finger signal. It should act as a reference for the finger response.
Figure 5: Tuning algorithm
The tuning algorithm shown in Figure 5 highlights the major steps needed to determine whether a system has been properly tuned or not. Setting of the calibration parameters can be done either by using the numerical approach or through manual trials. The parameters, which need to be changed, depend upon the algorithm selected for capacitive sensing.
Tuning for production variations:
For mass production, the PCB or components may be sourced from multiple fabrication houses and vendors. There can be variations such as:
* Parasitic capacitance (CP) of the sensors and the dielectric of the PCB material may vary from board to board based on the quality and process.
* The tolerance of external components/internal parameters can result in a variation in counts/capacitance gain.
* The tolerance of overlay thickness and variations in its dielectric constant may cause different finger capacitance for different sensors.
These variations should be considered while tuning designs in order to avoid failure of boards due to the mentioned variations. Some basic statistical methods can be applied to calculate the expected variation in finger capacitance and gain. Thus, the signal and various calibration and threshold parameters should be based on the average and the variance observed.
Tuning time saved – Auto-tuning
As we have seen, tuning could become a tedious task. For applications where time-to-market requirements are stringent or where users do not want to spend considerable time tuning, there are controllers available that implement auto-tuning algorithms. Auto tuning calculates sensor parasitic capacitance at startup and configures all the parameter for the required sensitivity. It also monitors noise during run-time and adapts threshold parameters accordingly.
About the Authors
Priyadeep Kaur has completed her BE in Electronics and Electrical Communication Engineering from PEC University of Technology, Chandigarh and is currently working with Cypress Semiconductor India Pvt. Ltd. as an Application Engineer. Her interests are embedded systems, analog circuits, and DSP. She can be reached at email@example.com.
Pushek Madaan is currently working with Cypress Semiconductor India Pvt. Ltd. as a Senior Application Engineer. His interests lie in designing Embedded system applications in C and assembly languages, working with analog and digital circuits, developing GUIs in C# and, above all, enjoying adventure sports. Pushek can be reached at firstname.lastname@example.org.