Thermoelectric Power Sensor Research Articles

Accuracy improvement of micro‐electro‐mechanical system microwave power sensor based on self‐reference algorithm

AbstractThis article presents a self‐reference algorithm to improve the accuracy of micro‐electro‐mechanical system thermoelectric microwave power sensor (5‐15 GHz). The output thermovoltage of the sensor is not only a function of the input signal power, but also relates to the signal frequency, resulting in low accuracy of the sensor. In this article, a mathematical model is established for the sensor to analyze the reasons for frequency dependence. Moreover, an algorithm is derived from analyzing the characteristics of the model and the experimental results. The algorithm relieves the dependence on the thermovoltage and frequency. Experiments show that the relative error of the sensor based on the proposed self‐reference algorithm decreases from 22% to 1.6%, and the SD is smaller than 0.4 mW. This self‐reference algorithm plays critical role in the practical application of microwave power sensors.

RF/MW Power Standard Realization Without Unitary Efficiency Assumption at DC/LF

Thermoelectric power sensors are widely used in commercial power meters at RF and microwaves, due to their superior robustness, stability, and accuracy if compared with other types of power sensors. Furthermore, their electrical architecture and related performance turned out to be very useful in the realization of the broadband primary power standards as alternative to resistive power sensors, i.e. bolometers. Hereby we present a comparison in term of effective efficiency of a thermoelectric sensor calibrated by applying two different methods of power substitution when used as thermal load in a coaxial microcalorimeter at RF and microwaves. The aim is to test a technology that could enable the realization of a primary RF/MW power standard independently of the assumption of unitary efficiency at the dc (or LF) reference power.

DC-25 GHz and Low-Loss MEMS Thermoelectric Power Sensors with Floating Thermal Slug and Reliable Back Cavity Based on GaAs MMIC Technology.

Wideband and low-loss microwave power measurements are becoming increasingly important for microwave communication and radar systems. To achieve such a power measurement, this paper presents the design and measurement of wideband DC-25 GHz and low-loss MEMS thermoelectric power sensors with a floating thermal slug and a reliable back cavity. In the sensors, the microwave power is converted to thermovoltages via heat. The collaborative design of the thermal slug and the back cavity, i.e., two thermal flow paths, is utilized to improve the efficiency of heat transfer and to ensure reliable applications. These sensors are required to operate up to 25 GHz. In order to achieve low microwave losses at the bandwidth, the floating thermal slug is designed instead of the grounded one. The effects of the floating slug on the reflection losses are analyzed by the simulation. The fabrication of these sensors is completed by GaAs monolithic microwave integrated circuits (MMIC) and micro-electro-mechanical systems (MEMS) technology. Measured reflection losses are less than −25.6 dB up to 12 GHz and −18.6 dB up to 25 GHz. The design of the floating thermal slug reduces the losses, which is equivalent to improving the sensitivity. At 10 and 25 GHz, experiments exhibit that the sensors result in sensitivities of about 51.13 and 35.28 μV/mW for the floating slug and 81.68 and 55.20 μV/mW for the floating slug and the cavity.

Modeling of an 8–12 GHz receiver front-end based on an in-line MEMS frequency discriminator

This paper focuses on the modeling of an 8–12 GHz RF (radio frequency) receiver front-end based on an in-line MEMS (microelectromechanical systems) frequency discriminator. Actually, the frequency detection is realized by measuring the output dc thermal voltage generated by the MEMS thermoelectric power sensor. Based on this thermal voltage, it has a great potential to tune the resonant frequency of the VCO (voltage controlled oscillator) in the RF receiver front-end application. The equivalent circuit model of the in-line frequency discriminator is established and the measurement verification is also implemented. Measurement and simulation results show that the output dc thermal voltage has a nearly linear relation with frequency. A new construction of RF receiver front-end is then obtained by connecting the in-line frequency discriminator with the voltage controlling port of VCO. Lastly, a systemic simulation is processed by computer-aided software and the real-time simulation waveform at each key point is observed clearly.

A Four-Port Microwave Phase Detector at $X$ -Band Based on MEMS Power Sensors

A four-port microwave phase detector based on MEMS power sensors at $X$ -band (8–12 GHz) is proposed in this paper. It consists of two power dividers, two power combiners, two phase shifters, and two MEMS power sensors. Four power dividers/combiners and two phase shifters compose a four-port junction to expand the phase detection range. A thermoelectric power sensor and a capacitive power sensor compose the MEMS power sensor for low and high power measurement. Two types of power sensors can improve the measurement range and enhance the overload capacity. This microwave phase detector is fabricated by GaAs monolithic microwave integrated circuit technology. Experiments show that its return loss is better than −10 dB, and the isolation is better than −13.2 dB over $X$ -band. Measured results of the phase shift fit well with the calculated results in an entire cycle of −180°~+180°. The phase detection sensitivities are $16.62~\mu \text{V}$ /° and 23.94 aF/°, and the response times are 0.408 ms and less than 20 ms for the two types of sensors, respectively. The dynamic range of phase detector can be expanded up to 4 W, which is almost quintuple our previous works.

An analytical model for self-assembling microwave power sensor with thermopile and curled cantilever beam

This paper builds a complete analytical model for self-assembling microwave power sensor with thermopile and curled cantilever beam. The power sensor consists of a thermoelectric power sensor for low power detection from 0.1mW to 100mW and a capacitive power sensor for high power detection from 100mW to 800mW. The curled cantilever beam is designed to improve the dynamic range and microwave performance of the power sensor. The analytical model includes an electromagnetic model, a static mechanical model, and a stationary heat transfer model. The electromagnetic model takes the S-parameter of indirect-heating thermoelectric power sensor into consideration. The static mechanical model considers the fringing fields between the curled cantilever beam and the measuring electrode as well as the initial curling induced by the stress gradient. The experimental results show a good agreement with the theory and validate the effectiveness of the presented models. Furthermore, good performance of the power sensor has found. The measured return loss is lower than −25.5dB at 8–12GHz and the output of the power sensor has good linearity with the incident radio frequency power up to 600mW.

A Sandwich-Type Thermoelectric Microwave Power Sensor for GaAs MMIC-Compatible Applications

This letter presents a sandwich-type non-microelectromechanical system (MEMS) thermoelectric microwave power sensor, which is based on the conversion principle of microwave power-heat-electricity. To reduce heat loss, a thermopile is designed to lie beneath two matching load resistors, electrically isolated by a dielectric layer. The design enables this sensor to eliminate the need for MEMS technology. To verify the design validity, the sandwich-type non-MEMS microwave sensor is fabricated together with a MEMS sensor. This improved sandwich-type sensor offers compatibility with the GaAs monolithic microwave integrated circuit technology. Experiments show that these sensors produce reflection losses of less than −20 dB up to 20 GHz. At 1, 10, and 20 GHz, the measured average sensitivities are ~101, 71.7, and $54.1 \mu \text{V}^{-1}\cdot \text{mW}$ for the traditional MEMS sensor and approximately 130.2, 92.3, and $66.8 \mu \text{V}^{-1}\cdot \text{mW}$ for the improved non-MEMS sensor, respectively, with good linearity of the output responses.

Application of the thermoelectric MEMS microwave power sensor in a power radiation monitoring system**Project supported by the National Natural Science Foundation of China (No. 11304158), the Province Natural Science Foundation of Jiangsu (No. BK20140890), the Open Research Fund of the Key Laboratory of MEMS of Ministry of Education, Southeast University (No. 3206005302), and the Scientific Research Foundation of Nanjing University

A power radiation monitoring system based on thermoelectric MEMS microwave power sensors is studied. This monitoring system consists of three modules: a data acquisition module, a data processing and display module, and a data sharing module. It can detect the power radiation in the environment and the date information can be processed and shared. The measured results show that the thermoelectric MEMS microwave power sensor and the power radiation monitoring system both have a relatively good linearity. The sensitivity of the thermoelectric MEMS microwave power sensor is about 0.101 mV/mW, and the sensitivity of the monitoring system is about 0.038 V/mW. The voltage gain of the monitoring system is about 380 times, which is relatively consistent with the theoretical value. In addition, the low-frequency and low-power module in the monitoring system is adopted in order to reduce the electromagnetic pollution and the power consumption, and this work will extend the application of the thermoelectric MEMS microwave power sensor in more areas.

Analytical and electrical modeling of a MEMS thermoelectric microwave power sensor

A new analytical method which can support full computer-aided circuit design for a microelectromechanical (MEMS) thermoelectric microwave power sensor operable at X-band is presented in this paper. A reduced dimensional form of a heat transfer equation is built in the radial direction based on point heat source approximation and precise volume mesh of the sensor. Frequency dependent losses are discussed to explain the sensitivity degradation with frequency. Based on this analysis, an electrical model for the power sensor is established. This approach facilitates the modeling of the coupled energy domains in the power sensor straightforwardly by means of a computer-aided simulation. Better estimation property is obtained compared to our previous works. Verifications are both conducted through ANSYS simulation and measurements with close agreements are observed. Relative errors between them are smaller than 13%. The deviations are discussed in detail. Fabrication of the thermoelectric power sensor is based on MEMS technology and the GaAs MMIC process.

Research on the Response Time of Indirect-Heating Microwave Power Sensor

This paper mainly presents the response time with different parameters to study the dynamic response characteristic of indirect-heating thermoelectric microwave power sensor in order to provide reference for numerous practical applications. The fabrication of thermoelectric power sensor is compatible with the microelectromechanical systems technology and GaAs microwave monolithic integrated circuit process. An equivalent model of a typical thermoelectric power sensor is proposed to briefly analyze the thermal time constant in cooling process and theoretically predicted trends are nearly in accordance with the measured fall time when the membrane has nearly the same length of the thermopile. Measurements indicate that the response time does not change obviously with the incident power and ambient temperature, and increases with the length of thermopile and distance between resistors and thermopile. When the length of thermopile is 100 μm and distance between resistors and thermopile is 10 μm, the rise time and fall time of power sensor under normal ambient temperature 300 K are 1.06 and 0.37 ms, respectively. Furthermore, the return loss and the output voltage are measured with a large incident power to characterize the performance and study the failure phenomenon of the power sensor. The obtained burn-out power of the power sensor is ~1.1 W.

A 3D Model of the Thermoelectric Microwave Power Sensor by MEMS Technology.

In this paper, a novel 3D model is proposed to describe the temperature distribution of the thermoelectric microwave power sensor. In this 3D model, the heat flux density decreases from the upper surface to the lower surface of the GaAs substrate while it was supposed to be a constant in the 2D model. The power sensor is fabricated by a GaAs monolithic microwave integrated circuit (MMIC) process and micro-electro-mechanical system (MEMS) technology. The microwave performance experiment shows that the S11 is less than −26 dB over the frequency band of 1–10 GHz. The power response experiment demonstrates that the output voltage increases from 0 mV to 27 mV, while the incident power varies from 1 mW to 100 mW. The measured sensitivity is about 0.27 mV/mW, and the calculated result from the 3D model is 0.28 mV/mW. The relative error has been reduced from 7.5% of the 2D model to 3.7% of the 3D model.

A novel in-line frequency sensor based on coupling capacitance for X-band application

This paper presents a novel in-line frequency sensor, based on coupling capacitance, for X-band applications. The novel frequency sensor can achieve absolute frequency measurement with a simple structure and no DC power consumption. Fabrication of the frequency sensor is completely compatible with the GaAs monolithic microwave integrated circuit process. A well-designed metal–insulator–metal capacitor is employed to couple a certain percentage of incident power and a thermoelectric power sensor is used to measure the coupled power. The sensor design is guided by HFSS simulation and a lumped circuit model. The results validate the effectiveness of the simulation and model, and show relatively good performance of the frequency sensor with simple and reliable components. The net sensitivity of the frequency sensor is about 1.43 mV (W•GHz)−1, and the measured S11 and S21 are better than −14.8 dB and −1.39 dB at X-band.

Equivalent lumped circuit model and S-parameter of indirect-heating thermoelectric power sensor

In this paper, an equivalent lumped circuit model used to obtain the S-parameter of indirect-heating thermoelectric power sensor is proposed. The working principle of indirect-heating thermoelectric power sensor includes electric-thermal conversion by terminal resistors based on Joule effect and thermal-electric conversion by thermopile based on Seebeck effect. The lumped circuit model takes into account the parasitic parameters of terminal resistors and the electromagnetic coupling between resistors and thermopile. In order to validate the model, S11 of the thermoelectric power sensors with different lengths of thermopile and distances between resistors and thermopile are measured from 1GHz to 20GHz. The calculated S-parameters, based on the lumped circuit model, agree reasonably well with the measured results. The results of equivalent lumped circuit model indicate that the S11 is influenced by the distance between resistors and thermopile, and does not change obviously with the length of thermopile and incident power. The present lumped circuit model is useful to reasonably determine the S-parameter of indirect-heating thermoelectric power sensor.

A GaAs MMIC-based X-band Dual Channel Microwave Phase Detector based on MEMS Microwave Power Sensors

By sensing the output powers of two channels with MEMS microwave power sensors, a wideband 8-12GHz dual channel microwave phase detector is presented. In order to detect phase difference between reference signal and testing signal in entire 0-360° at X-band, artificial 45° phase lead and 45° phase lag are introduced in two channels, additionally. The proposed phase detector is composed of power dividers, power combiners, 45° (at 10GHz) coplanar waveguide (CPW) transmission delay lines, MEMS capacitive power sensors and thermoelectric power sensors. The two CPW transmission lines result in 45° phase lead and 45° phase lag at 10GHz, and the two different kinds of power sensors are used to detect the combined powers. The fabrication of the phase detector is compatible with GaAs Microwave Monolithic Integrated Circuit (MMIC) technology. The phase detection measurement is accomplished at 10GHz with 0-180° phase shift between testing signals generated by a tunable analog phase shifter. Phase detection results in two channels show that the normalization of the measured results fit the calculated results well, though the results of phase lead have a little of deviation.

Millimeter-Wave Thermoelectric Power Transfer Standard

This paper describes a thermoelectric waveguide power sensor for the frequency range 75–110 GHz, which has been calibrated in a waveguide twin-type microcalorimeter. The sensor is suitable as a transfer standard to provide traceablity for high-frequency power calibrations. The design and calibration of the sensor are described. The calibration is based upon a novel measuring quantity, referred to as generalized efficiency. The results of mirocalorimeter measurements and sensor performance are given. The new transfer sensor is tested against conventional thermistor waveguide sensors in a direct comparison setup, showing good agreement within the estimated uncertainties.

Optimization of Thermoelectric Microwave Power Sensors Based on Thin‐Membrane Structure

In this work, the thermal conduction property of thermoelectric microwave power sensors is researched. The fabrication of the thermoelectric microwave power sensor consists of a front side and a back side processing using GaAs Monolithic microwave integrated circuit (MMIC) process and MEMS technology. An isolation structure on the front side is designed to prevent the thermal conduction from the resistor to the Coplanar waveguide (CPW). A thin-membrane on the back side is designed to prevent the thermal conduction from the resistor to the substrate. For the microwave power sensor without an isolation structure, the sensitivity is about 0.138, 0.136, 0.132, 0.115 and 0.111mV/mW at 8, 9, 10, 11 and 12GHz, respectively. For the microwave power sensor with an isolation structure, the sensitivity is about 0.142, 0.139, 0.135, 0.117 and 0.115mV/mW at 8, 9, 10, 11 and 12GHz, respectively. As a result, the higher thermal conduction efficiency and the higher sensitivity are obtained for the optimized thermoelectric microwave power sensors.

High dynamic range microwave power sensor with thermopile and curled cantilever beam

A high dynamic range microwave power sensor compatible with the gallium arsenide monolithic microwave integrated circuit process is presented. The power sensor consists of a thermoelectric power sensor and a capacitive power sensor for low and high power detection, respectively. To improve the dynamic range and optimise the impedance matching characteristic, the curled cantilever beam is utilised and the slot width of the coplanar waveguide transmission line is modified. The measured return loss is lower than –25.5 dB at 8–12 GHz. The output of the power sensor shows good linearity with the incident radio frequency power. For the incident power from 0.1 to 100 mW, the obtained sensitivities by a thermoelectric power sensor are about 0.0842, 0.0752 and 0.0701 mV/mW at 8, 10 and 12 GHz, respectively. For the incident power from 100 to 400 mW, the measured sensitivities by the capacitive power senor are about 0.0400, 0.0301 and 0.199 fF/mW at 8, 10 and 12 GHz, respectively.

Research on Thermocouple Distribution for Microwave Power Sensors Based on GaAs MMIC Process

This letter proposes the research on the thermocouple distribution for indirect-heating thermoelectric microwave power sensors, in order to achieve good sensitivity and signal-to-noise ratio (SNR). These power sensors are designed to operate at 0.01-30 GHz, and the heat distribution of the sensors is simulated to optimize thermocouple distribution numbers. Three microwave power sensors with 2, 6, and 14 pairs of thermocouple distribution are fabricated based on GaAs monolithic microwave integrated circuits process. Measured reflection losses of the three power sensors are lower than -21.5 dB up to 30 GHz. These sensors result in good linearity between measured thermovoltages and the microwave power. At 30 GHz, the obtained average sensitivity is 36.4, 107.1, and 240.4 μV/mW and the corresponding SNR is 1.6 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sup> , 2.8 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sup> and 4.2 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sup> /W for the 2, 6, and 14 pairs of thermocouple distribution, respectively. Furthermore, experiments show that the sensitivity of the sensors decreases as the microwave frequency increases, with relative measurement errors of less than ~5%.

The research of indirectly-heated type microwave power sensors based on GaAs MMIC technology

In this work, the effect of the temperature and the humidity on the indirectly-heated type microwave power sensor is researched in order to extend its application field, and their mechanism is revealed. The fabrication of this microwave power sensor is divided into a front side and a back side processing using GaAs MMIC process and MEMS technology. One advantage of this microwave power sensor consists in compatibility with MMIC devices and other planar connecting circuits structures with zero dc power consumption. The measurement results show that the environment temperature has a significant effect on the performance of the indirectly-heated type microwave power sensor, and the humidity has a little effect on its performance. The temperature dependent coefficient is about 0.6 mV/(W K), which has an important reference value to other thermoelectric microwave power sensor. The reason is that the accuracy microwave power measurement will be realized as long as the environment temperature is tracked in the testing process.

Improvements on INRIM Coaxial Microcalorimeter and Outcome of a Model Comparison

This paper describes hardware and software improvements of the INRIM coaxial microcalorimeter together with their outcome on the primary power standard realization in the frequency band 0.05 - 40 GHz. A better temperature and power stabilization turned out to provide an improved signal/noise ratio and a drift reduction in every working condition of the microcalorimeter. The INRIM correction model is also compared to a traditional, but faster, one in terms of measurement uncertainty. Outcomes are presented in form of a 2.92 mm thermoelectric power sensor calibration together with results that show the improved stability and repeatability of the measurement system.