Small Size and Low‐Cost TENG‐Based Self‐Powered Vibration Measuring and Alerting System

Vibration measurement systems containing sensors, signal conditioning, and data acquisition devices, are important for monitoring motors, gearboxes, turbines, etc. Microelectromechanical and piezoelectric sensors are predominantly used for vibration measurements. However, they are not cost‐effective, flexible in design, and incapable of self‐powering. Recently, triboelectric nano‐generator (TENG)‐based vibration sensors have been considered as a possible alternative to resolve this problem, and tremendous progress has been achieved. Previous work on TENG‐based sensors is limited to optimizing the sensor design, while the signal conditioning and data acquisition of TENG signal still need investigation for actual applications. This work develops a TENG‐based vibration measurement device and self‐powered alerting system that is integrated with the signal condition and data acquisition systems. The experimental results show that the proposed measurement system successfully measures signals within the range of 0–1800 Hz frequency. Meanwhile, the TENG generates a high output, up to 80 V and 0.55 µA from small size TENG area (3.6 cm2). The signal is adequate to harvest energy for self‐powering to drive alerting components (harvest 320 mJ in 36 h, which drives alarming for duration of 1.5 s). The proposed device is cost‐effective (30 $), small (105 cm3), and consumes less power (0.18 W) in comparison to commercial devices.


Introduction
Vibration measurements are essential for instrumentation and measurement systems in several scientific and engineering fields such as automotive engineering, biomedical engineering, mechanical engineering, aerospace engineering, mining engineering, civil engineering, etc. [1] In these fields, vibration measurements are important for collecting structural healthiness information, [2] machine condition monitoring information, [3] energy-scavenging potential. [20] The self-powering mechanism is highly desirable for sensing devices to improve flexibility and portability. [21] Some of the successful previous works are; an effective TENG-based self-powered sensor was invented for wireless tire pressure and speed monitoring, [22] detection of humidity and temperature, [23,22] and motion monitoring; [24 ] Moreover, a TENG-based self-powered sensor for vibration conditioning was developed, which was evaluated to monitor automobile engine vibration [25] and mechanical gear system. [26] Recently, we designed a spring-assisted TENG to measure vibration up to high frequency and utilized it for detecting bearing defects effectively as machine condition monitoring. [27] Even though several works are done toward developing TENG-based vibration sensors, further developments on the miniaturization and flexibility of TENG-based self-powered sensors is strongly required for developing exclusive TENG-based vibration sensors, that is applicable to measure vibration from various part of machine components. [28] In the measurement system, the analog signal detected by the transducer is amplified, excited, linear, filtered, and digitized by using signal conditioning and data acquisition methods before being interpreted. Several alternative methods are available for this task such as compact data loggers (recorded data over the period), Pc-based acquisition devices (display data via computer), and acquisition systems (integrated and synchronized sensor network system). [29][30][31] The main factors considered to select the right signal conditioning method are portability, power supply, precision, linearity, availability, features, and cost. [32] For signal conditioning of TENG-based transducer, researchers are utilizing conventional data acquisition devices, such as a Keithley electrometer, oscilloscope, and compact data acquisition system, which are not easily portable, expensive, and consume high power.
Here we present a TENG -sensor-based measurement device for vibration measurement in form of a logging measurement system that is easy to use, low cost, portable, and less power consuming. The sensor part is made based on our previous work. [27] The measurement device included Arduinonano and MATLAB for signal conditioning and data acquisition. The performance of the measurement device is compared with conventional measurement devices such as Electrometer (Keithley 6514) and an SKF accelerometer (PCB-080 M162) with microlog. We also demonstrate that the whole system can be powered by the energy harvested from the TENG sensor unit.

TENG-Based Sensor Structure
The TENG design is mainly based on our previous work. [27] Briefly saying, it is a spring-assisted TENG, where the outer cylinder is fixed relative to the proof mass, which vibrates due to the excitation of two springs in the cylinder. The TENG part is constructed on a cylinder and the proof mass: where a PTFE film attached with a back copper electrode is pasted on the cylinder wall as a negative triboelectric layer and a copper film is pasted on the proof mass as a positive triboelectric layer. This TENG is used to convert vibrational energy into elec-trical energy during in-plane sliding of the triboelectric layer (details working principles in supportive information). In this work, we divide the spring-assisted TENG into two parts, that is, the measuring part (M-TENG) and the energy harvesting part (P-TENG). As shown in Figure 1a, M-TENG, and P-TENG, the two separate triboelectric films were put on the opposite side of the proof mass for sensing and energy harvesting purposes. This allows optimization of energy harvesting efficiency by directly harvesting and storing the P-TENG signal without passing the signal through signal conditioning components to avoid energy leakage and drops on circuit loads.

System Framework
The proposed measurement system has two separate parts. The first part is the logging measurement part, which consists of a transducer (TENG) and a signal conditioning module (Arduino-nano, voltage divider, and computer) (Figure 1a,b). The Arduino-nano is used to detect the analog signal of the M-TENG, then the conditioned signal will be displayed via a computer using a MATLAB interface. However, the expected signal range to be generated by the TENG is 0-80 V, while Arduino only detects an analog signal in the range of 0-5 V. Therefore, a voltage divider with 0.6 and 10 MΩ external resistance is installed to tune the input voltage to 0-5 V according to Equation (1) before delivering the signal to the Arduino-nano. The second part is the self-powering and alerting system which consists of a power management module (the rectifying bridge, capacitor) and alerting component (LED and Buzzer) as shown in Figure 1b. The rectifying bridge converts an AC to a DC. The capacitor will store the rectified signal and monitor based on the capacitor voltage level. The stored energy and regulated voltage will be used to power the Arduino nano to drive the Arduino nano-based alerting system as shown in Figure 1b. The alerting components will be driven by the Arduino nano to deliver light and sound notifications at a certain level of vibration based on a pre-defined value of vibration level which was loaded via the Integrated development environment of the Arduino (IDE) program to the Arduino nano.

Assembling of the Measuring and Alerting System
As shown in Figure 2a, The TENG is constructed from, a thin hollow cylinder, lids, compression spring, proof mass, and polytetrafluoroethylene (PTFE)/copper (Cu) triboelectric film. The measuring and alerting components were assembled on a breadboard plate with a size of 100 × 35 × 10 mm as shown in Figure 2b. A voltage divider, Arduino-nano, rectifier, LED, and buzzer are connected to the breadboard via breadboard grids.
The signal from P-TENG and M-TENG is connected to the breadboard by using cables soldered to the respective TENG. Then the analog signal on the reduced part of the voltage divider is tied to the A5 port of Arduino nano via the analog input port by using a jumper wire. Similarly, the P-TENG signal connected to the breadboard was transferred to the rectifier www.advelectronicmat.de and capacitor via a series connection on the breadboard. Then the stored voltage on the capacitor passed to Arduino nano via power input and ground pin by using a jumper wire. The LED and buzzer are connected to the Arduino nano by using jumper wires via power output and ground port to receive driving power. The Arduino nano is connected to a computer via a USB 2 cable and a MATLAB program written to read the analog signal on the Arduino nano port (the detailed code is shown in  www.advelectronicmat.de the supporting information). Then Arduino-nano-based measurement system and Electrometer via LABVIEW were used for the characterization of proposed TENG properties and compared for their performance. The overall dimension of the Arduino-nano-based measurement system (i.e., the size of the TENG sensor, voltage divider, and Arduino) is 100 × 35 × 30 mm which is 1.35% of the Keithley electrometer (350 × 220 × 100 mm) as shown in Figure 2b (1.08-36.5% of conventional vibration measurement device size in the range of 288-9657 cm 3 as data presented in Table S1, Supporting Information). Further, a practical demonstration of the Arduino-nano-based measurement system compared with the SKF commercial accelerometer (PCB-080 M162) for measuring the vibration acceleration of the tribometer device was performed. Here, the electrical signal output of the measurement system is calibrated for an equivalent acceleration value measured by an accelerometer by using a shaker as a uniform vibration source. Then the MATLAB program was written to convert the output signal of TENG to an equivalent acceleration value for the proposed measurement system to develop the relationship between voltage and acceleration. Then sample measurements were collected from the tribometer reciprocating motion for comparison.

Small Size and Low-Cost Measurement System Characterization
The electrical output performance of the developed TENG was evaluated for an input vibration of 0-1800 Hz and 0.5-3 mm amplitude by using a conventional Keithley Electrometer (as shown in Figure 3a,b. The results indicate that the output is directly proportional to the input vibration frequency over a broad range. The TENG which consists of a copper electrode generates higher output compared to the aluminum electrode TENG. This is due to the larger electronegativity difference between PTFE and copper electrodes since copper electronegativity is 1.9 and aluminum electronegativity is 1.61. [33,34] The output of P-TENG is relatively higher than M-TENG as the contact surface area of P-TENG is larger than M-TENG. Because the contact surface area is one of the factors that influence the output of TENG. [35] As a result, the power output of the device increased up to 80 V and 0.55 µA, which is highly desired for a self-powered sensor. The device generates a high electrical signal at a reduced size and weight when compared to previous work on TENG or the device based on an electromagnetic mechanism. [36,37] Alike other energy harvesting devices the electrical output of a TENG is a function of load resistance. [16] As shown in Figure 3c, the output power density fluctuated as the external load resistance changed for various input frequencies. Initially, the output power density increased as the external load increased, and the maximum instantaneous power output density was obtained at 10 MΩ. When the external load increases from 10 to 40 MΩ the output power drops significantly (at ≈2 mW cm −2 ). Accordingly, the ideal external load resistance is ≈10 MΩ, which is one of the preliminary indications to propose a voltage divider load for an Arduino nano-based measurement device.
To reduce the data collection device size, an Arduino-nanobased device was developed (as described in characterization techniques). The output performance of the developed TENG from copper electrodes and PTFE was evaluated by using the developed Arduino-nano-based device (Video S1, Supporting Information). As shown in Figure 3d, the open-circuit voltage is proportional to the input frequency. The curve of the output voltage is very similar to the one measured by a conventional electrometer with a difference in magnitude of the output value, which may be due to performance and external load resistance difference between voltage divider/Arduino-nano, and the electrometer (10.2 MΩ internal load). However, as shown in Figure 3d, the output voltage value difference between Arduino nano and electrometer can be easily fitted by adding the gap difference of 13 V. In addition, Figure 3e,f shows an excellent similarity of signal features collected by Arduino-nano-based device and conventional electrometer (additional signal feature comparison is available in Figure S3, Supporting Information). Figure 4a demonstrates the effect of input acceleration of vibrations on the electrical output performance of the developed TENG. The graph shows that the input acceleration influenced the output of the TENG to increase in a proportional relationship. The shaker used for the experiment has an interdependent frequency and amplitude setup, therefore a line graph is used to show the relationships.
The practical application of the TENG-Arduino nanobased measurement system is investigated by measuring reciprocating tribometer acceleration. The pin holder of the

www.advelectronicmat.de
reciprocating tribometer device moves linearly in a reciprocating motion to create sliding of a pin over the disc for friction measurement analysis. Here, the acceleration of the pinholder is measured for the comparison of the two-measurement systems. Initially, the relationship of output voltage and acceleration developed in the form of an equation by using a vibration shaker for an Arduino-nano-based measurement system as shown in Figure 4b. The relationship equation was derived from acceleration data collected by changing the vibrational displacement (amplitude in the range of 0.5-3 mm) and frequency (0-1800 Hz) concurrently as an input factor of vibrational acceleration. This enables converting equations to be approximate and adapted for various situations of measurement. The acceleration-voltage equation is included in the MATLAB program for measurement conversion, that is voltage to acceleration. Accordingly, the acceleration of a reciprocating tribometer was measured by using the developed Arduinonano-based device and an SKF commercial sensor for various input frequencies.
As the result shown in Figure 4d, the measured accelerations are almost the same with an error of less than 0.01%. In addition, the collected signal is very stable, and uniform as shown in Figure 4e,f. The magnitude and peak number of the signal increase with input frequency in a similar way for the TENGbased sensor and the commercial sensor. On the other side, the Arduino-nano-based measurement system can be powered by a 6 V battery, which is 2-12% (1.5-80 W or 90-125 V) of conventional device power consumption (Table S2, Supporting Information). As a result, it can be said that the proposed small size, and low power-consuming TENG-Arduino-nano-based logging measurement device performs well for mechanical measurement applications, particularly for monitoring the vibration of mechanical equipment. Further, the two-sided balanced spring, overall weight, and symmetric cylindrical structural arrangement allow the reduction of intimate sliding friction to increase the durability and stability of the sensor ( Figure S4, Supporting Information).

Self-Powering and Alerting System Characterization
The image of the self-powering and alerting system is shown in Figure 5a. The signal of the M-TENG is collected via a voltage divider and directly transmitted to the Arduino-nano for detecting vibration signal level. The Arduino-nano is programmed to show practical demos, for instance, the Arduinonano is programmed to deliver light when acceleration reaches 12 m s −2 or an equivalent voltage of 15 V. The Arduino-nano is also programmed to deliver sound when the vibration level reaches 23m s −2 or an equivalent voltage of 25 V (the detailed program can be found in the appendix and a related Videos S2 and S3 (Supporting Information) is provided as supporting information). The P-TENG energy is harvested to drive the alarming system based on the Arduino-Nano-based programmed setup. The rectifier converts AC to DC. Two various capacitors are used for energy storage to compare self-powering performance, a 0.22 mF and a 10 mF capacitor are being used to store the harvested energy.
The minimum voltage required to drive the alerting system is evaluated by using a conventional power supply (EA-PSI 5080-20A). The device supplies voltage in a range of 0-80 V. It is found that the circuit requires at least 3.60 V and 0.05 A (180 mW) to start delivering the alarm notification. Then the expected voltage is harvested by using P-TENG at a frequency of 200 Hz and 1 mm amplitude. The charging and discharging performance of the capacitor during energy harvesting is measured by a Keithley electrometer to evaluate energy harvesting efficiency and consumption duration (as shown in Figure 5c-f). It is found that the 0.22 mF and the 10 mF capacitor are charged up to 8 V (7.04 mJ for 0.22 mF and 320 mJ for 10 mF) in 8 and 36 h, respectively. The result shows that the 10mF capacitor takes a long time to charge up as this capacitor stores high energy at a similar voltage. [38] The energy harvesting time can be improved by increasing the sensor size or improving the material properties of the triboelectric film. The discharging duration of the capacitor is only 1.1 and 6 s for 0.22 and 10 mF capacitors respectively when connected to an equivalent load to the alarming system.
The actual alarm-delivering duration of the alerting system is measured when stored energy power on the alerting system lights the LED bulb and forces the buzzer to give an alarming sound when the vibration reaches a pre-defined high value. The result shows that the alarming notification starts to disappear after 1.57 s for 10 mF capacitors (Videos S2 and SS3, Supporting Information). This is due to the alerting system requiring at least 0.18 W to run as already identified by using the conventional power supply. As well based on harvested energy and the required power supply, the theoretically calculated power consumption time is 1.7 s which is similar. The result indicates that we can develop a self-powering and alerting system from TENG.
Furthermore, the developed self-powered measurement system is cost-effective. As shown in the graph of Figure 6, based on a sample survey of a different commercial device, the overall cost of this device is compared to the other commercial measurement devices. The result indicates that the cost of the developed measurement device is very low, which is ≈30 $ for the whole components (details of cost are in supporting materials) and 0.6% for a high-quality conventional device (Keithley electrometer), and 10% for a medium conventional device (oscilloscope). In general, the data shows the device only costs 0.6-9.7% of commercial vibration measurement devices (Table S1, Supporting Information). Therefore, the developed device has advantages when compared to high-quality commercial devices in terms of cost, power consumption, and size at a similar performance, with a possibility to optimize functional diversity like another commercial device.

Conclusion
The developed TENG-based vibration sensor has the potential to generate a voltage of 80 V and a current of 0.5 µA. We successfully developed a TENG -sensor-based vibration measurement system with low cost and small size that has good performance, linearity, and stability. This was realized through structural design optimization, material selection, power management, and interface programming. The designed device can measure a range of vibration frequencies of 0-1800 Hz. The developed device demonstrates a promising potential for self-powering applications during measurement and alarming activities. The developed Measurement system is portable due to it is small size 105 cm 3 (1.08-36.5% of conventional vibration measurement device size), consumes less power (0.18 W, 2-12% of conventional vibration measurement devices), and is inex-pensive (costs ≈30 $, 0.6-9.7% of conventional vibration measurement device). Thus, the proposed measurement device has a high potential for applications as a linear motion and vibration measuring device for a wide range of acceleration and frequencies with excellent performance at a very low cost.   Table S2. Supporting Information). www.advelectronicmat.de

Experimental Section
Fabrication of TENG: As it is shown in Figure 2a, the 18 mm diameter, 1 mm thick, and 18 mm high hollow cylinder, and the 16 mm diameter and 8 mm thick poof mass were prepared from an aluminum bar by machining. Then, the PTFE with back copper or aluminum film with a dimension of 8 × 45 mm was stuck to the proof mass manually by using two-sided plastic tape. Similarly, two separate copper films of 12 × 30 mm for P-TENG and 12 × 15 mm for the M-TENG part were manually fastened on the internal wall of the cylinder symmetrically at the middle of the cylinder. The carbon (EN 10 270 Pt1 patented) spring which had a 5 mm free length and a 14 mm outer diameter from wizard springs company was utilized. The spring was soldered to the proof mass and inserted in the cylinder, which was then soldered to the top and bottom lids on the opposite end side. A 0.65 mm diameter wire cable soldered to the upper edge of the copper electrode/film was used to transfer signals.
Characterization Techniques: For the characterization of the performance of the prepared TENG, a shaker (JZK-2 shaker with signal amplifier) from Tira GMBH company was employed as a vibrator to generate uniform vibrations. The vibrator frequency was controlled by a signal generator to generate uniform periodic vibrations for various inputs such as amplitude in the range of 0.5-3 mm, frequencies in the range of 0-1800 Hz, and accelerations in the range of 0-60 m s −1 . The corresponding acceleration of the vibrator was measured by a commercial SKF accelerometer (PCB-080 M162). The output performance of the TENG, which was the open-circuit voltage and shortcircuit current was measured by an electrometer (Keithley 6514 from Keithley Tektronix company) displayed via LabVIEW and an Arduinonano displayed via MATLAB.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.