Temperature Control System For Microchip Electrophoresis

With the development of analytical chemistry and continuous increasing of medical demand, the microchip electrophoresis has great potential for development in accompanying with the micromanufacturing technology. Due to its advantage of small size, low cost and rapid analysis, the microchip electrophoresis has been more and more widely used. Microfluidic chip’s temperature is an important factor in electrophoresis process, and has a great influence on the column efficiency. High temperature will make the signal detected peak broadening even deformed. This thesis designs a high-precision temperature measurement circuit and a suitable temperature control system for microchip electrophoresis.A lot of temperature measurement methods of microchip electrophoresis have been studied and compared in this thesis. Characteristics of non-contact temperature and contact temperature measurement are mainly analyzed. During experimental exploration, infrared temperature sensor TMP006 was used in non-contact temperature measurement. TMP006 is flexible and easy to use. But it is susceptible to outside interference with poor stability. The temperature of microchip electrophoresis and the closure shell were measured to do comparison. Non-contact temperature measurement design can achieve functional requirements. But due to noise, the design needs to be optimized.Based on the research of contact temperature sensor, the thermal resistance is chosen as a temperature sensor in contact temperature measurement due to its high linearity and repeatability. According to the analysis of digital converter (ADC), △∑ modulator can significantly reduce quantization noise during A/D conversion process and it is suitable for high-precision measurement of low-frequency signals. A first-order RC low-pass filter for front-end filtering is designed considering the relationship among Sampling rate, anti-aliasing filter analysis and the digital filter.ADS 1148 is a powerfull chip, which integerates the constant flow source, PGA and △∑ADC and it is the perfect choice for the temperature detecting applications using Resistance Temperature Detector(RTD). In this system, three wires measuring method for the RTD with two constant flow sources is applied and proportional allocation method is used to provide the external reference voltage for ADC, which can help to eliminate the noise effect from the constant flow source. With the temperature measuring range to be 0℃-100℃, hardware compensation is utilized to expend the ADC’s working bandwith. This thesis has also done the low-noise PCB layout design and made the PCB board. Moreover, a MCU (MSP430F169) is used to communicate with ADS1148 through SPI interface.At last, the whole circult is simulated and verified in TINA software, which proves out to function well. The actual experiments reveal the mismatch from the constant flow source. Therefore we propose the method using two constant flow sources to conduct two transformation, to reduce the measuring deviations. After the comparison between the measurement results under different output rates, it is indicated that the noise restraining ability decreases when the output rate goes up.This thesis describes the overall structure of the control system. The temperature measurement circuit based on ADS1148 is used as a temperature detection module; solid state relays control the heating film as a heating output module, MSP430F169 microcontroller is the control module of the system, and each part of the configuration program of the microcontroller is described.According to the temperature rising characteristics of the microchip electrophoresis, this thesis creates an object model and design an appropriate temperature control program. First, a detailed understanding of PID control theory is proposed to analyze the influence of various parameters on the performance of the control system. And then compare the difference between the incremental equation of the digital PID algorithm and the position equation of the digital PID algorithm. Then, this thesis summarizes modeling methods and PID parameter tuning methods of the plant model at home and abroad. Subsequently, analyzes the modeling methods by SIMULINK and calculates the identification error. The smallest error method is choosen as the wanted modeling methods. Then, compares the simulation of the classic PID tuning methods with the IMC-PID design methods and analyzes the response results. Different from the conventional PID control methods, Lee’s internal model PID Tuning method has no overshoot, good robustness, and meets the temperature control requirements of the microchip electrophoresis.