Impedance Characterization of Wearable Capacitive Sensors

August 18, 2021 by Tim Ashworth

This blog post explains how the Zurich Instruments MFIA can be used to characterise wearable capacitive sensors, and introduces two recently published studies where innovative flexible wearable sensors have been developed which operate on the principle of capacitive sensing. These studies present different approaches to comfortable-to-wear sensors with sensing performance that exceeds existing rigid sensors. The characterization of these sensors was carried out by the papers' authors using the MFIA Impedance Analyzer.


Wearable sensors at the home-use level are growing in popularity and take myriad forms. They allow us to monitor various body parameters and can do so in real time. As their uptake and performance grows, this offers an opportunity to provide real-time, point-of-care monitoring in a decentralized system rather than the traditional clinic-based system. Such personal sensors should increase health awareness and promote healthier lifestyles, reducing the burden on health services worldwide.

Wearable capacitive sensors enable the measurement of many physical parameters critical to real-time heath monitoring. These range from walking and running to breathing and heart rate. Limb angle, blinking and even speaking can also be measured. The sensors are flexible and can be located on the body in a way that suits specific measurements.

Characterizing the requirements for wearable capacitive sensors

The requirements for a feasible sensor include different aspects, from high linearity to fast response times. Let's take each one in turn and look at how the MFIA can enable their characterization.

High linearity

Wearable capacitive sensors need to be able to offer the measurement of a wide range of forces and displacement. These forces can be high impact foot-fall measurements or small muscle movements from blinking, therefore the sensor needs to be able to operate linearly over broad ranges. This linearity allows for sensors that need minimal calibration, which is a clear advantage for wearables.


Figure 1: Reactance chart of the MFIA Impedance Analyzer showing its linearity in absolute impedance over a wide frequency range and impedance range.

High linearity can be demonstrated thanks to the wide capacitance measurement of the MFIA, which is linear over a wide range. When demonstrating linearity, it is important that the instrument making the measurement is more linear than the sensor under test. The reactance chart of the MFIA (see Figure 1) shows the wide range of capacitance over which accurate and linear measurements can be taken.

High sensitivity

Sensors need to be sensitive enough to distinguish small movement deviations within much larger movements. This can be demonstrated using the MFIA by first optimizing the operating frequency by sweeping this in the LabOne® Sweeper Module (as shown in Figure 2). The same can be done for the operating amplitude: sweep the amplitude and see how the capacitance response changes. It is then an easy step to select the optimum operating conditions.


Figure 2: Screenshot of the LabOne Sweeper module displaying frequency-domain data in both bode and nyquist plots. Multiple signals can be added to the Sweeper module to ensure all the relevant impedance parameters can be viewed in one place during the testing of the wearable sensor. The standard deviation and other math functions can be displayed for a given trace.

Once optimized, the capacitance can be measured in real time using the LabOne Plotter module while a calibrating load is applied. The standard deviation of the capacitance signal can be read in real time in the Plotter, allowing the sensitivity to be established.

High repeatability and reliability

The sensors need to return the same signal for a given stimulus over the lifetime of the sensor. This is important to be able to compare limb movements over a longer period of time. The repeatability of a sensor can be confirmed by repeatedly stressing the sensor while measuring the capacitance. The LabOne Plotter module of the MFIA (shown in Figure 3)  allows for long-term measurements of the shape of the capacitance change as the load is repeatedly applied over many hours or even days.


Figure 3: Screenshot of the LabOne Plotter module displaying time-domain data in realtime. Multiple signals can be added to the Plotter to ensure all the relevant impedance parameters can be viewed in one place during the testing of the wearable sensor. The standard deviation and other math functions can be displayed for a given trace.

Further, the LabOne DAQ module (shown in Figure 4) enables the triggered acquisition of each capacitance cycle during the load test, allowing for a comparison of each signature with high temporal resolution and with well-defined synchronization.

Robustness and stability

The sensors should perform as expected even after repeated high loads and over the lifetime of the sensor. This can be tested by measuring the capacitance for given loads from an external source and observing the response to ensure expected values and low overshoot. These measurements can be performed with the LabOne Plotter or the LabOne DAQ module of the MFIA.

Low mechanical and electrical hysteresis

The sensors should not suffer from hysteresis, which would lead to inaccurate measured signals that depend on the movement or electric history of the stimulus. This measurement is similar to the robustness test described above. A load is applied by an external source and the response of the sensor is measured with the MFIA. The measurement needs to be fast enough to observe any hysteresis. The MFIA offers short measurement time scales for such measurements.

Fast response time

The sensor should operate fast to measure the movement with enough temporal resolution to be able to distinguish similar movements. The response time can be measured with the MFIA using the LabOne Plotter and DAQ modules. These allow for time-domain measurements on a user-definable time scale ranging from 10 us (at 1 MHz) down to 10 s for slow, long-term movements.


Figure 5: Screenshot of the LabOne DAQ module displaying time-domain data. Multiple signals can be added to the chart to ensure all the relevant impedance parameters can be viewed in one place during the testing of the wearable sensor. The DAQ module allows for triggered acquisition, as shown in this example measurement of a piezo sensor. Read more about this measurement in this blog post.

The DAQ module enables triggered acquisition of transients, which means the step response of the sensor can be measured quickly, precisely and repeatably. It can be configured for fast acquisition of parameters, and for averaging of several transients where averaging is necessary due to high-noise environments.

Characterization of Wearable Capacitive Sensors – Real-World Examples

Let's take a look at two recently published papers where the impedance charcterization of the wearable capacitive sensors in the time- and frequency-domain was carried out by their authors using the MFIA.

Textile-based wearable capacitive sensors for improved physiological monitoring

Sensors that will enable health monitoring need to be flexible for the wearers to feel comfortable, yet robust enough that they can be worn for extended periods of time. One such wearable sensor has been developed by the group of Prof. Wuliang Lin at the University of Manchester, UK. The researchers developed a fabric-based sensor that is both flexible and robust, and yields important information on human breathing, speaking, blinking and joint motions. These sensors comprise a dielectric between a cotton substrate along with a conductive layer. The fiber surface was modified to ensure robust adhesion of the layers. The resulting sensor is not only comfortable and robust, but offers high sensitivity and a fast response time.

The authors used the MFIA to track the capacitance change as the sensor is loaded, and as the distance and angle between electrodes changes. The measurements confirm the high sensitivity of the sensor, which can measure joint motions and even small movements such as talking, breathing and blinking. The results also show that the time-domain capacitance signal can measure movement due to breathing and distinguish between differ breathing rates. Interestingly, the time response of the sensor was fast enough to be able to distinguish between different spoken sentences.

Here is the full paper: Chen, L. et al. Textile-Based Capacitive Sensor for Physical Rehabilitation via Surface Topological Modification. ACS Nano 14, 8191–8201 (2020).

Microcellular porous wearable capacitive sensors

Another example is given by the wearable sensors developed in the group of Prof. Gursel Alici at the University of Wollongong, Australia. Here, the researchers worked on a porous elastomeric capacitive sensor improving on the typical parameters required for real-time health monitoring. These sensors are not just flexible and comfortable to wear, they are also much more sensitive and linear than typical existing wearable sensors based on piezoelectric or polymers. They are robust, reliable and repeatable. These capacitive sensors have the additional advantage of providing a reliable reading even after the movement has stopped. In contrast to existing piezoelectric sensors based on electroactive polymers, these microcellular sensors can therefore capture both dynamic movement and static positions important for heath monitoring.

The authors used the MFIA to go through the following sensor characterization steps:

  • Measured the dielectric of the sensor in the time domain as the pressure was varied by an external jig, and measured the dielectric of the sensors for various porosities.
  • Measured the capacitance of the sensor for a repeating force to quantify its repeatability and reliability.
  • Measured the drift of the capacitance of the sensor for a given load to confirm the low drift.
  • Probed stability and reversibility through impedance measurements of 10000 load cycles provided by an external source over 20 hours. The high temporal resolution of the MFIA allows for the shape of early cycles to be compared with that occurring after 10000 cycles.
  • Established the sensor's lower detection limit by measuring the capacitance while applying ever decreasing loads.

Here is the full paper (along with videos showing capacitance measurements for limb movement in the Supporting Information): Sencadas, V., Tawk, C., Searle, T. & Alici, G. Low-Hysteresis and Ultrasensitive Microcellular Structures for Wearable Electronic Applications. ACS Appl. Mater. Interfaces 13, 1632-1643 (2021).


Acknowledgments: I'd like to thank the authors of the above papers for their fascinating studies – we look forward to reading more. I'd like to thank Dr. Tom Searle further for helpful discussions and for the inspiring videos.