Available online at www.sciencedirect.com Available online at www.sciencedirect.com
Procedia Engineering
Procedia Procedia Engineering 00 (2011) Engineering 15000–000 (2011) 28 – 32 www.elsevier.com/locate/procedia
Advanced in Control Engineering and Information Science
Implementing for Networked Motion Control System with Large-Capacity Data Acquisition Lei Wanga, Mu Guolib, Jing Wangb a* b
a School of Electronic & Information Engineering, Dalian University of Technology, Dalian, 116024,China State Key Laboratory of Coastal and Offshore Engineering, Dalian University of Technology, 116024,Dalian China
Abstract As one of the most popular real-time Industry Ethernet protocols, EherCAT has become widely used in distributed networked motion control systems. By the help of the unique features, the EtherCAT protocol is adopted to construct a networked servo motion control system with low delivery delay and high accurate clock synchronization in the paper. This system can achieve not only high-speed synchronous transmission of motion control commands and feedback information but also high-efficiency large-capacity data acquisition with low delay and jitter. A set of experiments, which evaluate the crucial performance for the implemented networked motion control system including messages access delay and jitter, are performed. Through analyzing the experimental results, the system performance is verified.
© 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of [CEIS 2011] Keywords: EtherCAT; motion control; data acquisition; clock synchronization
1. Introduction Distributed servo motion control systems can be finding applications in a broad range of area such as printing machines, robotics, surface mounting machines, assembly lines, and CNC machine tools, etc. To cope with the desire to reduce wiring, lower cost, and increase reliability, various Industry Ethernet
* Corresponding author. Tel.: +86-411-84708520-8207; fax: +86-411-84708018. E-mail address:
[email protected].
1877-7058 © 2011 Published by Elsevier Ltd. doi:10.1016/j.proeng.2011.08.007
2
Lei et al. Procedia Engineering Engineering 00 15 (2011) (2011) 000–000 28 – 32 LeiWang Wang,et al// Procedia
protocols have been introduced into those control systems to construct the networked motion control systems in recent years. These protocols include SERCOS III [1], PROFINET, POWERLINK, EPA, and EtherCAT, etc. Different from other networked control applications, most of networked motion control systems require specially the protocols with less 1 millisecond cycle time and high accurate time synchronization. Therefore, it is necessary for comparing and evaluating the main performance indices of various industry Ethernet protocols to selecting an appropriate protocol. Up to now, many researchers have evaluated the performance of several main stream Industry Ethernet protocols [2-5]. At the same time, a considerable number of Industry Ethernet protocols have been introduced into motion control systems to implement networked motion control [6-10]. However, few researchers investigate the hybrid networked motion control systems with multi-axes servo motors and a great deal of sensors. In view of the special application requirements for high real-time performance and large amounts of information transmission, we select EtherCAT protocol to design the class of networked motion control systems. This paper presents the design of the EtherCAT as communication network for interconnecting servo motors and a large amount of sensors of real-time networked motion control systems. In this paper, the performance of the control system is experimentally evaluated on the designed networked motion control system test platform. The rest of the paper is organized as follows: in Section 2 we introduce briefly the EtherCAT protocol; in Section 3 we describe the overall system architecture and the design of software and hardware. Experimental results are shown in Section 4 for testing and analyzing the system performance. The paper is concluded in Section 5. 2. EtherCAT Protocol In EtherCAT protocol the basic Ethernet frame structure is not changed. Thus, EtherCAT is able to be compatible with other Ethernet protocols. EtherCAT frame can be distinguished from other Ethernet frames through the EtherType field based on the special type identifiers 88A4. A standard Ethernet frame consists of five fields: preamble, start of frame delimiter, header, data, and a Frame Check Sequence (FCS). One EtherCAT frame is encapsulated in the data field of Ethernet frame. Every EtherCAT frame starts with a 2-bytes EtherCAT frame header. The value is used to ascertain the length of the following one or more data called EtherCAT telegrams. To communicate between a master station and multi-slave stations by uniform standard format, EtherCAT datagrams are defined as process data objects (PDOs) in the object dictionary. PDO is subdivided into PDO header followed by one or more PDO variables and Working Counter (WKC). Before the master sends a data frame, WKC has an expected value. When the frame passes through each slave station, the slave controller increases the WKC value in hardware. After the frame goes back to the master station, actual WKC value is compared to the expected value. By ensuring the working counter is equal to the expected value, each node can be ensured that it receives the entirety of the frame. Corresponding PDO variables are correlated with the Input/output control variables of distributed field controllers through the data object dictionary. 3. Implementing of networked motion control system 3.1. The Whole System Structure In this study, we construct the networked wave maker system including 10 slave stations. Fig 1 shows the whole structure of the system. The master station is a common PC plugged a standard Ethernet network card. In the following we will present detailedly the system from both master station and slave station
29
30
Wang et al. / Procedia Engineering 15000–000 (2011) 28 – 32 LeiLei Wang / Procedia Engineering 00 (2011)
Fig. 1. Structure of networked wave maker control modules
3.2. Circuit Principle of Slave Station We will present the circuit principle through the aforementioned four parts. 3.2.1. EtherCAT network interface circuit In the circuit, the ASIC named ET1100 is regarded as EtherCAT slave controller. It performs the function of Data Link Layer. The layer corresponds to Layer 2 in ISO model and provides real-time communication assurance among devices connected via EtherCAT network. Furthermore, the layer implements the function of the frame check and accomplishes data transmission by extracting data from and/or inserting data into the Ethernet frame. This is done depending on Data Link Layer parameters which are stored at predefined memory locations. However, the data frames transmission depends on network Physical Layer which is often termed PHY. The layer can receive data bits stream from Data Link Layer and encodes the bits into signals. Sequent, the signals are sent to the transmission medium and received by the next Physical Layer interface. And the signals are passed to the Data Link Layer of the next slave controller after they are decoded. To realize the function of Physical Layer, the Physical Layer interface chip is used and is linked to ET1100 via Medium Independent Interface (MII). At the same time, the chip is connected to medium port connector RJ45 via network transformer HR601680 which is used to improve the signal anti-interference capability. In addition to the above mentioned circuit and parts, ET1100 provides Serial Peripheral Interface (SPI) to link application layer controller. In the board, the affiliated circuits, such as clock circuit which generates clock signal by a 25 MHZ crystal oscillator, EEROM circuit used to store device configuration and device description, resetting circuit, etc., are also included. 3.2.2. Application controller circuit The application controller is the control core and it takes charge of running the EtherCAT protocol program and application data accessing program. Due to the advanced motor control peripheral features, fast and efficient CPU, and small cost-effective package sizes, we adopt dsPIC Digital Signal Controller (dsPIC33FJ256MC710) as the core controller. The chip is used to connect with the network interface controller and serial ADC, respectively. And the chip provides the control and data lines for all other
3
4
Lei et al. Procedia Engineering Engineering 00 15 (2011) (2011) 000–000 28 – 32 LeiWang Wang,et al// Procedia
circuits. It is a bridge between EtherCAT network interface circuits and field data acquisition and servo motor control circuits. 3.2.3. Sensors data acquisition circuit In the system, the data acquisition circuit for each slave station, with the capacity for 16 simple analog channels input, is designed. Sixteen sensors are connected with the channels input ports. Because the output signal from sensors is approximately 0.2V to 0.8 V, the collected signals must be amplified. To simplify the circuit design, the 16 sensors are connected to a multiplexer. Which channel is selected depends on the 4-bit binary address lines A0, A1, A2, and A3. These pins receive the address signals from the application microcontroller output pins RB3, RB2, RB1 and RB0. The values are decided by the channel numbers from the master station. According to the values the multiplexer selects one of the collected 16-channel signals and forwards the selected input signal into a single out line. The output signal feeds into the amplifier which is capable of implementing gains 5. And hence the amplifier out signal is converted into 1V to 4V. Finally, the analogue signals will be converted into digital signals via ADC. It provides a SPI to connect with microcontroller. 3.2.4. Servo motor interface and control circuit The part of circuit includes mainly pulse sending/receiving interface, servo motor encoder feedback interface, servo driver and servo motor. The microcontroller receives the control commands from the master station and generates direction and the number of pulses required to rotate for a given angle. Pulses are sent to servo driver via interface circuit as a square wave, the number of pulses determines the angle of rotation and frequency of square wave determines the speed of rotation. And then the driver controls the motor to move to destination position and it takes both speed and position feedback from the encoder of servo motor. The feedback values can be detected and calculated by the microcontroller. 4. Experiment results test
Fig. 2. Irregular wave high signals from wave high sensors
Fig. 3. Averaged cycle time for 10 groups of sample points
31
32
Wang et al. / Procedia Engineering 15000–000 (2011) 28 – 32 LeiLei Wang / Procedia Engineering 00 (2011)
In order to illustrate clearly the advantage of EtherCAT, the system is tested in ocean experiment pool. The tested irregular wave high signals are shown in Fig 2. The results show that the system can cope with the ocean experiment requirements. But also we test the system cycle times. And then we have analyzed 10 groups of cycle time values and calculated the averaged values. These values are depicted in Fig 3. The figure indicates that the maximum delay jitter is only 3.2 microseconds. 5. Conclusions The paper has introduced the design of a class of typical networked motion control system based on EtherCAT protocol. This kind of networked motion control system with large amount of measurement sensors asks for not only achieving motion control but also numerous sensor signals transferring by less than 1 millisecond communication cycle time. Through testing the designed control system platform, the experimental results show that EtherCAT protocol is the most suitable communication protocol for the type of mentioned networked motion control system. Acknowledgements The authors would like to acknowledge the support of funding (Grant No. 50879098) from Natural Science Foundation of China, and the support of the Education Department University Research Project of Liaoning Province (Grant No. LS2010032). And this work was supported by a grant from the Fundamental Research Funds for the Central Universities (Grant No. DUT10JR14). References [1] K. Chang Lee, S. Lee, and H. Hee Lee. Implementation and PID tuning of network-based control systems via Profibus polling network. Computer Standards & Interfaces 2004; 26: 229-240. [2] P. Ferrari, A. Flammini, and S. Vitturi. Performance analysis of PROFINET networks. Computer Standards & Interfaces 2006; 28: 369-385. [3] G. Cena, L. Seno, A. Valenzano, S. Vitturi. Performance analysis of Ethernet Powerlink networks for distributed control and automation systems. Computer Standards & Interfaces 2009; 31: 566-572. [4] L. Seno, and S. Vitturi. A simulation study of ethernet powerlink networks. in: Emerging Technologies and Factory Automation, Conference 2007; 740-743. [5] H. Makete. Real-time requirements for discrete time applications in automation systems. Elektron 2005; 22: 36-40. [6] I. Erturk. A new method for transferring CAN messages using wireless ATM. Journal of Network and Computer Applications 2005; 28: 45-56. [7] A. Flammini, P. Ferrari, E. Sisinni, D. Marioli, A. Taroni. Sensor interfaces: from field-bus to Ethernet and Internet. Sensors and Actuators A: Physical 2002; 101: 194-202. [8] A. Abarca, M. de la Fuente, J. M. Abril, A. García, F. Pérez-Ocón. Intelligent sensor for tracking and monitoring of blood temperature and hemoderivatives used for transfusions. Sensors and Actuators A: Physica 2009; 152: 241-247. [9] A. Flammini, P. Ferrari, D. Marioli, E. Sisinni, A. Taroni. Wired and wireless sensor networks for industrial applications. Microelectronics Journal 2009; 40: 1322-1336. [10] L. Wang, O. Peter, C. Andrew, L. Sherman. Remote real-time CNC machining for web-based manufacturing. Robotics and Computer-Integrated Manufacturing 2004; 20: 563-571.
5
