Investigation of Low-Cost Accelerometer Performance for Vibration Analysis of Bridges 131 Fig. 1 Low-cost MEMS accelerometer with screw terminals attached and mounted in field enclosure. (a) Accelerometer with screw terminal block installed. (b) Accelerometer with leadwires in field enclosure. (c) Accelerometer enclosure sealed in electronics epoxy Section A-A Steel Plate 38.1 mm 3.175 mm 3 Spa. @ 203.2 mm = 609.6 mm 812.8 mm Instrument Grade Accelerometer (Typ.) Fixed Support Low Cost Accelerometer (Typ.) A A 1 2 3 4 Measurement DOF (Typ.) Fig. 2 Schematic of the cantilever beam test setup 3 Cantilever Beam Evaluation 3.1 Test Setup A small-scale steel cantilever beam structure was developed and instrumented for use in this study. The cantilever beam has a span length of 812.8 mm. A total of four PCB3701 accelerometers were installed with hot glue on the beam at 203.2 mm spaces from the free end. A total of four ADXL335 accelerometers were installed directly on top of the Model 3701 accelerometers with hot glue. A schematic of the cantilever beam setup is shown in Fig. 2 and a photograph of the test setup is shown in Fig. 3. The accelerometers were connected to a NI-9234, four channel input module from National Instruments for data acquisition. External 5 VDC and 24 VDC power supplies were used to provide the excitation voltage for the ADXL335 and PCB3701 accelerometers, respectively. The measurement data were acquired from the acquisition hardware using SigView software. Since the author only had access to a single NI-9234 input module for testing the cantilever beam, the data was collected in three separate test cases. All of the accelerometers were left in place on the beam throughout the testing to keep the dynamic characteristics of the beam constant during data collection. In Test Case 1, the four instrument-grade accelerometers (PCB3701) were connected to the input module and free vibration data were recorded. In Test Case 2, the four low-cost MEMS accelerometers (ADXL335) were connected to the input module and free vibration data were recorded. In Test Case 3, the two instrument-grade accelerometers and the two low-cost accelerometers located at Positions 1 and 2 as shown in Fig. 2 were connected to the input module and free vibration data were recorded. Free vibration testing consisted of applying an initial displacement to the free end of the beam and releasing. The test was repeated five times for each test case. The sampling rate used for all testing was 400 Hz, and the free vibration response was recorded in each case for 6 s.
RkJQdWJsaXNoZXIy MTMzNzEzMQ==