Capacitive sensing technology in human contact applications

Capacitive sensing isn’t just used in your smartphone; it also finds its way into products like medical devices that must come in contact with human skin. This article shows how to use capacitive sensing to determine the quality of contact between the device surface and the user’s skin.

Capacitive sensing technology continues to gain traction in traditional human-machine interface applications such as laptop trackpads, MP3 players, touchscreen displays , and proximity detectors. In addition to replacing mechanical buttons with capacitive sensors , a little imagination and basic principles of human-machine interface design will allow many other applications to take advantage of this technology. Figure 1 shows some examples of application concepts that can be enhanced by using human contact detection.

Figure 1. Device using capacitive sensor electrodes.

For devices such as those shown in Figure 1, it is often helpful to know information about the quality of contact between the device and the skin before activating the device or taking a measurement. These devices include medical probes that need to be in close contact with the skin, biopotential electrode sensors, or housings that hold catheter tubing. To determine contact conditions, several capacitive sensor electrodes (shown in green) can be embedded directly into the plastic housing of the device during the injection molding process. The host microcontroller reads status registers on the capacitive sensor controller IC that indicate how close the capacitive sensor is to the skin. A basic detection algorithm running on the host microcontroller then processes the status register information to determine if each sensor electrode is making adequate contact with the skin.

In traditional capacitive sensing human interface applications, people typically initiate contact with the sensor electrodes by touching them with a finger. The example in Figure 1 uses capacitive sensors in a non-traditional way, where the user places a device containing capacitive sensing electrodes on the body. Developing this type of application is straightforward, but there are some key guidelines that should be followed to build a robust and reliable system.

Capacitive Digital Controllers. Developing a high-performance contact detection application begins with selecting an appropriate capacitive digital controller ( CDC ). For the application shown in Figure 1, the contact between the device surface and the skin is measured directly through the subtle changes in energy distributed across the array of capacitive sensor electrodes that occur when the device makes contact with the skin. The accuracy of this measurement depends on the sensitivity of the CDC analog front end and the number of sensor electrodes. Capacitive sensors manufactured using traditional PCB processes are typically in the 50 fF to 20 pF range, so the high-precision measurement techniques of a 16-bit CDC are ideal.

When selecting a CDC, look for key features such as a high-resolution analog front end with a 16-bit ADC , programmable sensor sensitivity settings, programmable sensor offset control, on-chip environmental calibration, sufficient capacitive input channels to support the ideal number of sensor electrodes, and an integrated design that eliminates the need for external RC devices for sensor calibration. These features support reliable and flexible applications that provide the best user experience. For example, programmable sensitivity allows interface designers to preset the best sensor sensitivity for a specific application rather than using a fixed solution that may result in poor sensitivity. Programmable offset control is another important feature for interface designers because the offset value of each production batch of sensor boards may vary slightly. Rapid pre-characterization allows host firmware settings to be changed before putting a new sensor board into mass production. For applications where the ambient temperature or humidity is expected to change, on-chip environmental calibration allows for a more reliable solution. Note that electrode sensors are built using standard PCB copper traces; the properties of the substrate change with changes in temperature and humidity, which will change the baseline level of the sensor output. If the CDC supports on-chip calibration, this baseline drift can be dynamically compensated during product use.

Small electrodes require high sensitivity. The goal of the measurement is to determine how close the device is to the skin; the better the contact quality between the skin and the device, the more accurate the device reading. The accuracy of the measurement depends on the number (more electrodes, higher resolution) and size of electrode sensors distributed over the device contact surface area. For the application shown in Figure 1, the surface area of ​​the device is generally small, requiring designers to use small sensor electrodes when developing the application.

In order to reliably measure the small capacitance changes associated with small sensor electrodes (typically less than 50 pF), a highly sensitive analog front-end controller is required. Keep in mind that the type and thickness of the plastic cover material will further affect the small signal emitted by the sensor through the plastic. The controller's analog front-end measurement must be sensitive enough to measure this small signal while maintaining a good signal margin between the measured signal and the threshold level detection setting under all operating conditions (such as different supply voltages, temperature and humidity, and thickness and type of cover material). Lower signal margins increase the risk of false detection and sensor instability. To minimize the risk, when using a CDC with a 16-bit ADC, a margin of at least 1000 LSB should be maintained between the sensor baseline level (sensor is not in contact with the skin) and the contact threshold level.

[page]

The AD7147 and AD7148 CapTouch programmable controllers for single-electrode capacitive sensors feature 16-bit resolution, femtofarad-level measurements, and 16 programmable threshold detection level values ​​over the full-scale range. These controllers support small sensor electrodes as small as 3 mm × 3 mm under 1 mm of plastic cover material (dielectric constant 3.0) while still maintaining a full-scale signal margin of 1000 ADC LSB. Full-scale signal margin is the difference between the sensor output when there is no skin contact and when there is skin contact.

Maintain reliable performance. Capacitive sensor electrodes are made of standard copper or flexible materials on PCBs. The properties of this material change with changes in temperature and humidity. This change causes the baseline level to shift (all sensor threshold levels are referenced to the baseline level). Large baseline shifts increase the risk of the contact threshold level being too low or too high (too low or too high depends on the direction of the baseline shift), which can cause false contact errors or make the threshold level either too sensitive or not sensitive enough, resulting in unstable contact status. To maintain the original signal contact threshold detection level margin (sensitivity) of the sensor, the CDC needs to automatically track the magnitude of the baseline offset error and readjust the threshold setting accordingly. The example in Figure 2 shows how the threshold levels of the AD7147 and AD7148 automatically track and adjust for changes in baseline offset caused by changing environmental conditions.

Figure 2. AD7147/AD7148 on-chip environmental calibration.

Eliminate measurement errors. Adding capacitive sensors to the device The act of retrofitting the electrode array can create space constraints, forcing the designer to place the CDC far from the capacitive sensor. This increases the length of the parallel sensor traces and makes the routing dense, which is not conducive to capacitive sensing applications because the traces at different DC potentials form stray coupling paths as shown in Figure 3A. The ground plane of the PCB cannot prevent this because the traces and the ground plane are at different DC potentials and still form stray capacitance (Figure 3B).

Figure 3. Path of stray capacitance, showing the results for parallel traces without a copper pour (A), on a ground copper pour (B), and on a copper pour with the same DC potential as the trace (C).

One way to eliminate stray capacitance errors is to surround adjacent traces with a plane driven by the same dc level as the capacitive sensor electrodes and traces. The AD7147 and AD7148 devices eliminate stray capacitance by providing a dedicated ACSHIELD output with this capability, as shown in Figure 3C.

Consumer health care devices such as spa skin care products are moving from professional institutions to ordinary families, and users are no longer specially trained technicians who are familiar with the products and their applications. Therefore, many of these products require smarter user interfaces to enable untrained users to master the correct use of the products. Capacitive detection provides new options for user interface designers, allowing them to explore various innovative methods to meet new user interface requirements. Capacitive digital technology provides information about the contact between the capacitive sensor electrode and the skin, which can be used to maintain optimal product performance and safety.