传感器信号调理应用现场可编程模拟阵列.doc

SENSOR SIGNAL CONDITIONING USING FIELD PROGRAMMABLE ANALOGUE ARRAYS (FPAA) 传感器信号调理应用现场可编程模拟阵列(FPAA) ABSTRACT Field Programmable Analogue Array (FPAA) is a new analogue design system. This paper introduces the application of FPAA in the signal conditioning of the temperature sensors. The major signal conditioning tasks undertaken are linearization, amplification and offset removal. Signal conditioning is very important for developing a sensor with good performance. Although there are many techniques for the signal conditioning, using FPAA has some significant advantages compared with the traditional methods. In this paper, the design, simulation and measurement results of sensor conditioning circuit using the second-generation FPAA will be presented. The results obtained show that FPAA can provide an easy, convenient and reliable way for the signal conditioning of temperature sensor . 1 INTRODUCTION Sensors are widely used in process control, automation, data acquisition, test equipment, instrumentation and communication systems. All the sensors properties are influenced by temperature, dirt and other environmental parameters. Therefore some data processing circuits are usually integrated with the sensor units depending on the requirements. Sensor signal conditioning can be defined as the manipulation of the output signal of a sensor, probe or transducer including signal conversion, attenuating, amplifying, filtering and linearising. linearising 。 Signal conditioning of any analogue signal is required when the output signal is not up to the level or form in which it is required by the system. Previous techniques of sensor signal conditioning using discrete circuits have major disadvantages such as lack of precision due to component tolerances and mismatches; techniques developed did not support complex functions and signal conditioner developed was confined to the specific sensor and cannot be used for multiple sensors. The main challenge for the sensor signal conditioning is to develop a system that can easily modify the sensor output as required with good accuracy and linearity. Using FPAA can achieve this purpose. FPAA introduces a radical software-centric approach to analogue circuit design where easy-of-use is a major feature. The technology has the econfiguration flexibility,allowing complete analogue signal conditioning/processing systems to be integrated using a Graphical User Interface (GUI). Because FPAA is programmable and reconfigurable, just one device can provide multiple sensor conditioning circuits under the real-time control of a digital microprocessor. Sensor signal linearisation, offset compensation, calibration, and signal modulation circuits can now be implemented in minutes on a drift-free integrated silicon platform. 2 SYSTEM DESIGN The system design challenges include sourcing stable references and stimulus; multiple sensors with differing signal conditioning needs, methods of calibration and maintenance and manufacturing considerations. FPAA can provide the sensor signal conditioning with an alternative solution. FPAA ，Design of the system using FPAA involves: (1) Amplification, offset removal and linearization of the signals from sensors; (2) Simulation using FPAA software before using the circuits for real time system; (3) Interfacing the Sensor system with the FPAA; (4)Carrying out test in order to condition the signal obtained from different Sensors. The FPAA was the heart of the whole system which takes the input signal that was initially converted from the physical signal by the sensors. The basic purpose of the system was to manipulate the sensor signal into a signal that was appropriate for any control system. The FPAAs that are used is Anadigms AN220E04 which is also called as “Dynamically Reconfigurable FPAA”. As the signal passes through the FPAA, the offset is firstly removed, and then the signal is amplified and filtered. Thus by using the FPAA signal conditioning the sensor signal with higher precision can be achieved. Figure 1 System block diagram of FPAA sensor signal conditioning Figure 1 shows a system block diagram of FPAA sensor signal conditioning. The main features of the system are: (1) Performance: The system should accept the voltage signal from the temperature sensor interfaced to the FPAA and process it accordingly depending on the specific feature selected among the different signal conditioning tasks. The system should convert the sensor output signal into a good quality signal that has been amplified, linearised and can be directly fed to a control system. The FPAA should also provide a predictable and a stable stimulation to the sensors interfaced to it.。 (2) Power supply: +5 volts. (3) Sensors: The sensors used for signal conditioning are two different types of temperature sensors ,specifically thermistor based and thermocouple based temperature sensors. (4) Input: The input to the FPAA system is the signals obtained from the temperature sensors. The signals from the two different sensors are as follows: (a) Signal from thermocouple: The output signal from the thermocouple is a voltage signal that in turn corresponded to the change in temperature and is directly fed to the FPAA in addition with a couple of resistors to provide a common mode bias signal. (b) Signal from thermistor: The resistance obtained from the thermistor terminals corresponds to the change in temperature. In order to convert the resistance obtained into a voltage signal a wheatstone bridge is constructed so that the signal can be fed to the FPAA and serve as the input signal to the FPAA. (5) Process: The process basically involved the conversion of the sensor signal to a level that could be directly connected to a control system. The whole process of signal conditioning involves offset removal, amplification, filtering and linearization. The design procedure for the system is as follows: (1) Design the circuits with required functions in FPAA using FPAA software and various Configurable Analogue Modules (CAMs); (2) Simulate the designed circuits using FPAA software to view the performance of system; (3) Download the designed circuits to the FPAA development board, interface the sensor signal and test the real system. 3 SYSTEM IMPLEMENTATION The implementation of system involves the selection of the sensors and the design of the circuits in FPAA (AN220E04). 3.1 Selection of the sensors The first task involved in the implementation of the design is to find the suitable sensors that exhibit the characteristics required for interfacing it to the FPAA and easily allow the successful signal conditioning using the FPAA. The sensor parameter that is particularly selected for signal conditioning is temperature. The next step for selecting the sensors is to select them from the different available categories. Thermocouple and thermistors are selected as the output signals from them need amplification and linearised. As can be seen, the purpose of the design can be easily defined and proved by using these two sensors. (1)Selection of thermocouple Thermocouples provide an economic means of measuring temperature with many practical advantages, they are extremely robust, capable of measuring over a very wide temperature ranges and very easy to install. For convenience two thermocouples are selected: Bayonet Type J (RS 219-4747) and Type K (RS 290-5042). The main features and specifications of these two thermocouples can be found in their data sheets . (2)Selection of thermistor A thermistor is an electronic component that exhibits a large change in resistance with a change in its body temperature. The thermistor that was selected is the PT100 type (RS 376-1477). The sensor contains a Positive Temperature Coefficient (PTC) and is platinum temperature sensor. The resistance of the sensor can be taken directly to a controller. In order to measure the resistance or specifically the change in temperature two terminals are given which can be connected to a device measuring resistance or to a wheatstone bridge to convert the resistance to a voltage signal. 3.2 Design of the circuits using FPAA software This section describes the steps for designing sensor signal conditioning circuits by using FPAA software: AnadigmDesigner2?. The design is divided into two steps: (1) Gain requirements; (2) Linearization. (1)Design of the signal conditioning circuits for thermocouple (a) Gain requirement for thermocouple The gain required for a thermocouple is defined by the temperature range required and the sensitivity of the thermocouple. To achieve perfect results, the thermocouple voltages requires to be amplified before any further signal conditioning is done. In order to prevent the amplifier to go into saturation due to offset voltages a Chopper amplifier is used as the input cell.This technique can achieve gains up to 128 with very little offset at the output signal.The chopper clock frequency is set to 125 kHz and the clock frequencies for all other CAMs are set to 250kHz. The Chopper output is filtered by using the FilterBiquad CAM, and then connected to a GainHold CAM which can provide a high quality output signals with automatic offset compensation.Fig.2(a) shows the gain requirement circuit for the thermocouple. It is necessary to set the gain for the chopper amplifier to 64 or less than that for optimal performance of the circuit. (a) (b)Figure 2 Signal conditioning circuits for thermocouple (b) Thermocouple linearisation A typical thermocouple response for temperature versus voltage relationship is nonlinear, which means that the output voltage is not ideally proportional to the input temperature. In order to calculate the temperature the thermocouple voltage requires a Look-Up Table, or analogue linearization technique.FPAA software provides a built in Look-Up Table with 256 values. The Transfer Function CAM in FPAA chip has a 256 value Look-Up Table, therefore it is the important block in building the linearisation circuit for the thermocouple. The Transfer Function CAM is used to generate an analogue voltage in accordance with the change in temperature being sensed.The CAM uses an internal ADC to digitise the input signal, which uses the 8-bit digital word as the address for a Look-Up Table.The 8-bit data word from the Look-Up Table is converted back to an analogue output voltage signal. The content of the Look-Up Table is user defined. The Transfer Function CAM can be used to take an input value and return back a linearised output value. In order to achieve the linearisation technique it is necessary to use a SumDiff CAM with the Transfer Function CAM. A “straight line” transfer function is provided by the lower input of the SumDiff CAM, and the Transfer Function CAM provides the required perturbations from the “straight line”. The gain circuit designed in the previous stage is included in the linearisation circuit with some added CAMs with programmable DC shift. In this case a voltage CAM is added to generate +3V constant voltage and is scaled using the lower input of the SumDiff CAM added to the filtered copper signal. A complete thermocouple linearisation circuit, shown in Fig.2(b), is realised by above designed CAMs such as Transfer Function and gain as well as the circuit to maximise the dynamic range. (2)Design of the signal conditioning circuits for thermistor The design procedure of the signal conditioning circuits for a thermistor is similar to that for the thermocouple discussed in 3.2.1. The realised circuits are shown in Fig.3, where (a) is the gain requirement circuit, and (b) the linearisation circuit. (a) (bFigure 3 Signal conditioning circuits for thermistor 4. SIMULATION AND MEASUREMENT RESULTS The circuits shown in Fig.2 and 3 can be simulated by using FPAA software, and downloaded to the FPAA development board for practical measurement. The simulation and practical measurement have been carried out for the linearisation circuits for the type J and K thermocouples because the linearisation circuit provides the gain requirement result. The simulation and practical measurement results for the type J thermocouple are shown in Fig.4,(a) shows the simulation result, and (b) the measurement result. In both of the results, the upper line indicates the circuit output voltage, and the lower line indicates the thermocouple output voltage. (a) (b) Figure 4 The simulation and measurement results for the type J thermocouple 5 CONCLUSIONS Using the FPAA software to design the circuits for different applications of the sensors can provide a simple implementation of complex functions and help in rapid prototyping and testing. The signal conditioning of the sensor is under the real time control and with high precision. Clearly, using the FPAA can obtain an easy and convenient way for signal conditioning of the temperature sensors. The circuits developed in AN220E04 provide high gain, high linearity and high accuracy for temperature Sensors. 附录B 外文翻译译文部分 传感器信号调理应用现场可编程模拟阵列(FPAA) 摘要 现场可编程模拟阵列（ FPAA ）是一种新的模拟设计系统。本文介绍了FPAA的应用中温度传感器的信号调节。主要信号调理任务是进行线性化，扩增和抵消清除。信号调理对发展具有良好的性能的传感器是非常重要的。虽然有许多方式的信号调理，与传统的方法比较使用FPAA有一些重要的优势。在本文件中，使用第二代FPAA设计，模拟和测量结果的传感器调理电路将实现。结果表明， FPAA可以对温度传感器的信号调理提供一种简单，方便，可靠的方式。 1 导言 传感器被广泛应用于过程控制，自动化，数据采集，测试设备，仪器和通讯系统。 所有的传感器性能都会受到温度、灰尘和其他环境参数的影响。因此，有些数据处理电路通常是根据规定的要求结合传感器单位。传感器信号调理可以被界定为操纵的输出信号的传感器，探测器或传感器包括信号转换，衰减，放大，滤波和整形 。信号调理任何模拟信号时，需要的输出信号是没有达到的水平或形式，它是所要求的系统。前技术的传感器信号调理电路采用分立的主要缺点，如缺乏准确性由于部分公差和错位；技术开发不支持复杂的功能和信号调节良好性能只限于特定的感应器，并不能用于多个传感器。 传感器的信号调理面临的主要挑战是建立一个系统，可以很容易地修改传感器的输出，要求具有良好的精度和线性度，使用FPAA可达到这个目的。FPAA介绍了一种先进的软件为中心的方法来模拟电路的设计，易于使用的是一个主要特点。这项技术的重新配置的灵活性，允许完成模拟信号空调/处理系统，以综合使用的图形用户界面(GUI)。因为FPAA可编程和重构，只需一台设备可以提供多个传感器调节电路的实时控制的数字微处理器。传感器信号线性，抵消补偿，校准和信号调制电路，现在可以实施分钟漂免费集成芯片平台。 2系统设计 传感器信号调理系统的设计挑战包括采购稳定参考信号和响应；，多个传感器信号调理的需求不同方法的校准和维修制造的考虑。FPAA 能提供传感器信号调节的替代解决方案。系统设计使用FPAA包括：(1) 传感器信号的放大，抵消漂移和线性；(2)使用FPAA仿真软件，才能使用的电路实时系统；(3) 传感器系统与FPAA的接口；(4)进行测试，以条件的信号不同的传感器。 该FPAA的核心是整个系统考虑传感器的输入信号从最初的物理信号的转换。其基本目的是为了操纵系统的传感器信号转换成一个信号，也是适当的任何控制系统。该FPAAs所使用的是Anadigm的AN220E04，这也是被称为“动态重构FPAA”。当信号经过FPAA时，首先删除漂移，然后筛选和放大信号。因此使用信号具有较高精度的FPAA信号调理传感器是可以实现的。 图1 FPAA传感器信号调理系统方框图 图1显示FPAA传感器信号调理系统方框图。系统的主要特点包括： (1) 性能：该系统应接受电压信号的温度传感器连接到FPAA和进程取决于相应的具体特点选择不同的信号调理任务。该系统应转换传感器输出信号转换成高质量的信号，表明已放大，线性化 ，可以直接连接到控制系统。该FPAA还应提供一个可预见的和稳定的信号传感器连接到它。 (2)电源：5伏特。 (3)传感器：传感器用于两种不同类型的温度传感器的信号调节。特别是热敏电阻和热电偶的温度传感器的基础。 (4)输入：输入的FPAA系统是温度传感器的信号。信号从两个不同的传感器如下： (a) 信号热电偶：从热电偶的输出信号是一个电压信号，从而符合温度的变化，并直接向FPAA与一对电阻器来提供了一个共模偏差信号。 (b)信号热敏电阻：电阻从热敏终端对应温度的变化。为了转换成一个阻力获得电压信号惠斯登电桥的建造，使信号可以连接到FPAA提供服务，并作为输入信号FPAA 。 (5)过程：这个过程基本上涉及转换的传感器信号的水平，可直接连接到一个控制系统。全过程的信号调理涉及抵消删除，放大，滤波和线性。 设计程序的制度如下： (1)设计的电路功能的FPAA需要使用的软件和各种FP