Wireless sensor developments for physical prototype

Transcription

1 Wireless sensor developments for physical prototype testing SAS 2008, Atlanta, Georgia, USA, 12 February 14 February 2008 Edgar Moya, Tom Torfs, Bart Peeters, Antonio Vecchio, Herman Van der Auweraer, Walter De Raedt LMS International - IMEC

2 Outline Scenarios and objectives Sensors MEMS 3D stacking technology Interface Wireless link Compromises Radio Link Wireless receiver Interface with the acquisition system Data acquisition system requirements Block diagram Interface with the system Physical implementation Results Future research and conclusions 2 copyright LMS International

3 Scenarios Measurement system Sensors (Accelerometers) PC & data-acquisition front-end ( channels) Data acquisition Signature testing Experimental E i t l Modal Analysis Vibration analysis Excitation shakers or hammer force cell 3 copyright LMS International

4 Scenarios Civil Engineering Automotive Aerospace 4 copyright LMS International

5 Scenarios 5 copyright LMS International

6 Objective: short term Concept Integration of the MEMS sensors and radio transmitters Wired analogue sensors combined with wireless digital sensors Traditional sensors combined with MEMS sensors Data Acquisition system RF Module Sensor 3 Sensor 4 Sensor 5 6 copyright LMS International

7 Objective: long term Complete Wireless MEMS sensor network Integration of the radio receiver in the acquisition system (New Generation Acquisition systems) s) Data Acquisition system 7 copyright LMS International

8 Challenges Multichannel data acquisition Multi-axes sensors Vibration analysis: DC 20kHz (High sampling frequency) Synchronized data needed Noise level e limited and high dynamic range (contrary to some biomedic and car sensors) Size: 1 cm 3 Covering large measuring area Range: 300 m Measuring analog Time data Analog to digital conversion No high cost wiring MEMS sensors Reduce installation costs Size Radio (Bw Range) Synchronization Autonomy 8 copyright LMS International

9 Outline Scenarios and objectives Sensors MEMS 3D stacking technology Interface Wireless link Compromises Radio Link Wireless receiver Interface with the acquisition system Data acquisition system requirements Block diagram Interface with the system Physical implementation Results Future research and conclusions 9 copyright LMS International

10 MEMS Sensors Micro Electro Mechanical System Plus: Small size Low cost (25 $) Easy integration with electronics ect cs Low power supply (1v 3v) Multi-axes sensor Minus: Low sensitivity (power supply / measurement range) Higher noise level Normally not designed for modal analysis Not massive market Slow development 10 copyright LMS International

11 3D Stacking It is a layered, modular system containing: Bottom layer Power management (voltage regulation and switch on/off) Second layer Sensor functionalities Third layer low power microcontroller Top layer low-power radio with integrated antenna Li-ion ion battery (25 mm x 20 mm x 4 mm, 3.7 g) The prototype system uses connectors for the vertical interconnect A more compact implementation can be made using solder ball interconnect technology Antenna Radio Chip Recharger Microcontroller MEMS sensor Power Switch Battery 11 copyright LMS International

12 MEMS Sensor Interfaces Automotive Aerospace applications: 2D accelerometer was selected. For 3D An additional vertical board in the system Analog output Internal 12-bit ADC on the microcontroller layer FREESCALE MMA6233Q (High measurement range) Measurement 10 g Sensitivity 120 mv/g Frequency range 0.9 KHz Voltage supplier 3 v Noise 30 ug/ Hz 2 axis 12 copyright LMS International

13 MEMS Sensor Interfaces Environment Civil applications: 3D accelerometer which also includes the analog-to-digital converter (ADC) ADC included in the sensor: 12 bits KIONIX KXP (Low measurement range) Measurement Sensitivity Frequency range Voltage supplier Current supplier Noise 3 axis 2 g 819 counts/g, 12 bits 1.7 KHz 3 v 1 ma 175 ug/ Hz 13 copyright LMS International

14 Outline Scenarios and objectives Sensors MEMS 3D stacking technology Interface Wireless link Compromises Radio Link Wireless receiver Interface with the acquisition system Data acquisition system requirements Block diagram Interface with the system Physical implementation Results Future research and conclusions 14 copyright LMS International

15 Wireless Compromises 1.- Size Vs Autonomy Longer autonomy on time Bigger batteries. Bigger size More distortion in the measure 2.- Power Vs Sensitivity Sensitivity Power / Range Lower power Lower sensitivity 3.- Distance Vs Power Longer distance Higher consumption Higher consumption Shorter autonomy 4.- Sampling Frequency Vs Number of Sensors Higher Sampling Freq per sensor Broader Bandwidth per sensor Broader Bw per sensor less # sensors 5.- # Axes Vs # Sensors Higher # Axes Broader Bw less # sensors 15 copyright LMS International

16 Radio link Current design: The Nordic Semiconductor nrf2401a ultralow power 2.4GHz transceiver Sampling Frequency: 2.5 KHz X 3 axis = 7.5 KHz x 12 bits = 90 Kbps (DATA) Data info + Control info + Error info + Radio info = 250 Kbps (Max. good quality) Distance = few meters ( < 8 m) Battery = around 8 hours Bandwidth is the Bottleneck for online measurements 2.5 KHz max sampling freq. Less Sampling Freq per sensor Reduced number of sensors Difficulties for online measurements 16 copyright LMS International

17 Wireless receiver USB stick or Wire connection UART output data Synchronization: Periodic beacons every 25.6 ms Sensor network get synchronized (µs) Between beacons, sensors transmit Sampling moments Radio communication: time time Beacon: receiver transmits; sensor nodes receive Data frames: sensor nodes transmit; receiver receives 17 copyright LMS International

18 Outline Scenarios and objectives Sensors MEMS 3D stacking technology Interface Wireless link Compromises Radio Link Wireless receiver Interface with the acquisition system Data acquisition system requirements Block diagram Interface with the system Physical implementation Results Future research and conclusions 18 copyright LMS International

19 Data acquisition system requirements Communication via the audio QDA port: Audio format: SPDIF Minimum Sampling Frequency: 6.4 KHz No synchronization Interface between Receiver and data acquisition system: Format conversion: UART SPDIF Upsampling: 2.5KHz 6.4 KHz Radio Receiver INTERFACE UART SPDIF 2.5KHz 6.4 KHz SCADAS III (QDA) 19 copyright LMS International

20 Block diagram Fixed Input Sampling Frequencies. Minimum 6.5 KHz IMEC LMS LMSInst FPGA Resampler Receiver UART SPARTAN3 SPDIF SPDIF SCADAS AD1896 XC3S1500 Fs=2.5 KHz Fs=6.5 KHz 20 copyright LMS International

21 Interface with the system Concept validation Validation of the receiver and the interface to the SCADAS Verification with shaker test 21 copyright LMS International

22 Physical implementation Power Source Test.Lab FPGA SCADAS 22 copyright LMS International

23 Physical implementation FPGA 23 copyright LMS International

24 Results: Time domain F B Time Point1 Time Point5 g Real Real g s 0.27 g Real F B Time Point5 Time Point s copyright LMS International

25 Results: Frequency domain ( g /Hz) db db ( g 2 /Hz) F F B PSD Point1 PSD Point2 PSD Point Hz ( g 2 /Hz) db db ( g 2 /Hz) F F B PSD Point1 PSD Point2 PSD Point Hz copyright LMS International

26 Outline Scenarios and objectives Sensors MEMS 3D stacking technology Interface Wireless link Compromises Radio Link Wireless receiver Interface with the acquisition system Data acquisition system requirements Block diagram Interface with the system Physical implementation Results Future research and conclusions 26 copyright LMS International

27 Future Research Study the degradation of the signal at high frequencies Multiple wireless sensor nodes operating in a network New architectures Synchronization and networking wired and wireless sensors Use of repeaters must be studied New radio technologies broader bandwidth MEMS reduce noise level Focus the platform to the application 27 copyright LMS International

28 Conclusions Wireless sensor network combines measurement precision, low power consumption, wireless communication and low cost equipment. First approach has been developed for physical prototyping testing. The performance of the WSN was compared with the classical wired monitoring system Accurate modal testing can be carried out with standard wireless technology and MEMS sensors. 28 copyright LMS International

29 Acknowledgement This work was carried out in the frame of the MEDEA+ project 2A204 SWANS Silicon platforms for Wireless Advanced Networks of Sensors. The financial support of the Institute for the Promotion of Innovation by Science and technology in Flanders (IWT) is gratefully acknowledged. 29 copyright LMS International

30 Thank you for your attention SAS 2008, Atlanta, Georgia, USA, 12 February 14 February 2008 Edgar Moya, Tom Torfs, Bart Peeters, Antonio Vecchio, Herman Van der Auweraer, Walter De Raedt LMS International - IMEC

31 31 copyright LMS International

32 Outline SWANS Scenarios and Demonstrator Sensors Piezo-electric sensors MEMS Interface Wireless link Compromises Radio Link Interface with SCADAS SCADAS Limitations Block diagram Components Physical implementation Results Dissemination Future activities 32 copyright LMS International

33 Dissemination 1. - Papers accepted at conferences. Wireless sensor network for bridge vibration monitoring design and results, T. Uhl, A. Hanc, K. Mendrok & P. Kurowski, B. Peeters, E. Moya & H. Van der Auweraer. EVACES07. Porto, Portugal. October 2007 Bridge monitoring system using wireless sensor network hardware solution and preliminary tests. T. Uhl, A. Hanc, B. Peeters, E. Moya, H. Van der Auweraer. International Workshop on Structural Health Monitoring, Stanford University, Stanford (CA), USA. September 2007 Wireless sensor developments for physical prototype testing. E. Moya, T. Torfs, B. Peeters, A. Vecchio, H. Van der Auweraer, W. De Raedt. IEEE Sensor Application Symposium, Atlanta, Georgia, USA. February Demonstration during MEDEA+ forum 3.- University lecture + demo: Bart Peeters, "Wireless sensing for civil engineering Structural Health Monitoring", Dept. Structural Engineering, Università di Pisa, 20 November copyright LMS International

34 Cooperation with SWANS partners Block Diagram S E N S O R Signal Conditioning LPF LNA ADC RF Transmitter Battery 3D-Stacking Network RF Receiver SCADAS Interface SPI - SPDIF Resampler S C A D A S SENSOR LEVEL RECEIVER LEVEL LMS Automotive ot demonstrator 34 copyright LMS International

35 Cooperation with SWANS partners Block Diagram IMEC Nordic nrf2401a LMS IMEC FREESCALE Kyonix Nordic nrf2401a S E N S O R Signal Conditioning LPF LNA ADC RF Transmitter Battery 3D-Stacking Network RF Receiver SCADAS Interface SPI - SPDIF Resampler S C A D A S VARTA LPP CE SENSOR LEVEL LMS Automotive ot demonstrator RECEIVER LEVEL 35 copyright LMS International IMEC FPGA XILINX EVL1500 Analog devices AD1896EB

36 Cooperation partners IMEC : Automotive and Environmental Scenario 3D Stacking Interfacing with the sensor Radio managing Microcontroller managing Energocontrol: Environmental Scenario (WP5) Possibility of integrating their own wireless system with the LMS measurement system SCADAS Verhaert: Common platform for the environmental sensor AnSem: ADC designed with according to the LMS requirements 36 copyright LMS International

37 Outline SWANS Scenarios and Demonstrator Sensors Piezo-electric sensors MEMS Interface Wireless link Compromises Radio Link Interface with SCADAS SCADAS Limitations Block diagram Components Physical implementation Results Dissemination Future activities 37 copyright LMS International

38 Future activities - civil engineering Wireless Sensor Network will be tested in a modeled bridge (5 meters long). Civil scenario Comparison between 4 wireless sensors and 4 traditional sensors Data acquisition (IMEC software) + data analysis (test.lab LMS software) 38 copyright LMS International

39 Future activities automotive engineering Comparison results with a wireless and a wire sensor over a rotary machine Fault Simulator System (MFS2004) Data acquisition (IMEC software) + data analysis (test.lab LMS software) 39 copyright LMS International

40 Piezo-electric Sensors PIEZO Wired MEMS Wireless 40 copyright LMS International

41 Radio link: nrf2401a Transceiver. The Nordic Semiconductor nrf2401a ultralow power 2.4GHz transceiver FEATURES True single chip GFSK transceiver in a small 24-pin package (QFN24 5x5mm) Data rate 0 to1mbps Only 2 external components Multi Channel operation 125 channels Support frequency hopping Channel switching time <200µs. Power supply range: 1.9 to 3.6 V Address and CRC computation ti Shock Burst mode for ultra-low power operation and relaxed MCU performance DuoCeiver for simultaneous dual receiver topology Low supply current (TX), typical 10.5mA -5dBm output power Data slicer / clock recovery of data 100% RF tested No need for external SAW filter World wide use Low supply current (RX), typical 18mA peak in receive mode "Green" lead free alternative 41 copyright LMS International

42 Hardware tools: SPARTAN3 XC3S200, XC3S400 XILINX Spartan 3 Evaluation Kit 42 copyright LMS International

43 Hardware tools: Resampler AD copyright LMS International

44 Sensor level: Battery 44 copyright LMS International

45 Interface Block Diagram FPGA X,Y Resampler Input X,Y (2.5 KHz) RESAMPLER BOARD Extracting data Axes (X,Y,Z) YZ) Z Resampler Input Z (2.5 KHz) RESAMPLER BOARD X,Y,Z SPDIF encoder Z (20 KHz) Uart Decoder SPDIF encoder X,Y (20 KHz) X,Y,Z Data Receiver (2.5 KHz) X,Y (20 KHz) Z (20 KHz) SCADAS III (QDA) 45 copyright LMS International

46 Physical implementation AD FPGA Upsampler p 46 copyright LMS International

47 Physical implementation 47 copyright LMS International

48 Interface Definition and components SENSOR LEVEL: MEM sensor (Kionix KXP , Freescale MMA6233Q), Battery (VARTA LPP CE), Radio (Nordic nrf2401a), 3D Stacking and microcontroller (IMEC technology) RECEIVER LEVEL: Radio (Nordic nrf2401a), FPGA (XILINX EVL1500), resampler (AD1896EB), measurement system (LMS SCADAS) Nordic nrf2401a RF communication Nordic nrf2401a 3D Stacking Battery XILINX EVL1500 Analog devices AD1896EB KIONIX: KXP Freescale: MMA6233Q Sensor Level SCDAS: SPDIF Module Receiver Level 48 copyright LMS International