Output comparison of Analog vs Digital mems accelerometer
Analog vs. Digital MEMS Accelerometer Output Comparison
Hello,
We are developing a condition monitoring device and are currently testing its performance using the SPM Vib-15 as a reference, which is a well-known industrial condition monitoring device.
SPM Vib-15 Vibrameter (Analog Device) :
Link to datasheet: Here
Our Device with ST Accelerometer Sensor (IIS3DWB Digital Sensor):
Link to the datasheet: Here
IIS3DWB Configuration :
Accelerometer ODR: IIS3DWB_XL_ODR_26k7Hz
FIFO BDR: IIS3DWB_XL_BATCHED_AT_26k7Hz
Effective ODR: 26.9 kHz
Full Scale: ACCELERO_FS_16
Hardware Filter: IIS3DWB_HP_ODR_DIV_800
Issue :
We are measuring the acceleration RMS and velocity of a compressor/chiller/motor in an industrial setting during normal operation. Despite the IIS3DWB sensor providing output in all three axes, we are considering only the Y-axis, which aligns with the SPM Vib-15.
We have taken readings for acceleration peak, acceleration RMS, and speed RMS using both our device and the SPM Vib-15. However, our device’s readings are significantly different from those of the SPM Vib-15, particularly at higher vibration levels (almost double). The acceleration RMS values have been converted to G to match the units used by the vibrameter, and velocity RMS values are already in mm/s for both devices.
Problem :
Despite similar measurement conditions and proper mounting of both devices, our readings do not match the SPM Vib-15 readings. Given that both devices serve the same application, should we expect the digital sensor data from the IIS3DWB to closely align with the analog sensor output from the SPM Vib-15?
Are we making an error by comparing the outputs of an analog sensor to those of a digital sensor? What could be causing these discrepancies?
We have also attempted linear regression for calibration, but it did not resolve the issue.
Additional Information :
Images:
First from the right side is the SPM vibrameter, and last one (biggest one) is our device.
mounted in vertical direction
mounted in horizontal direction
(in this image at the center is our device)
the third device is for refference (blue one) , which has same sansor that we are using in our device.
Data brief of that device: https://www.st.com/resource/en/data_brief/steval-stwinkt1b.pdfAnalog vs. Digital MEMS Accelerometer Output Comparison
Hello,
We are developing a condition monitoring device and are currently testing its performance using the SPM Vib-15 as a reference, which is a well-known industrial condition monitoring device.
SPM Vib-15 Vibrameter (Analog Device) :
Link to datasheet: Here
Our Device with ST Accelerometer Sensor (IIS3DWB Digital Sensor):
Link to the datasheet: Here
IIS3DWB Configuration :
Accelerometer ODR: IIS3DWB_XL_ODR_26k7Hz
FIFO BDR: IIS3DWB_XL_BATCHED_AT_26k7Hz
Effective ODR: 26.9 kHz
Full Scale: ACCELERO_FS_16
Hardware Filter: IIS3DWB_HP_ODR_DIV_800
Issue :
We are measuring the acceleration RMS and velocity of a compressor/chiller/motor in an industrial setting during normal operation. Despite the IIS3DWB sensor providing output in all three axes, we are considering only the Y-axis, which aligns with the SPM Vib-15.
We have taken readings for acceleration peak, acceleration RMS, and speed RMS using both our device and the SPM Vib-15. However, our device’s readings are significantly different from those of the SPM Vib-15, particularly at higher vibration levels (almost double). The acceleration RMS values have been converted to G to match the units used by the vibrameter, and velocity RMS values are already in mm/s for both devices.
Problem :
Despite similar measurement conditions and proper mounting of both devices, our readings do not match the SPM Vib-15 readings. Given that both devices serve the same application, should we expect the digital sensor data from the IIS3DWB to closely align with the analog sensor output from the SPM Vib-15?
Are we making an error by comparing the outputs of an analog sensor to those of a digital sensor? What could be causing these discrepancies?
We have also attempted linear regression for calibration, but it did not resolve the issue.
Additional Information :
Images:
First from the right side is the SPM vibrameter, and last one (biggest one) is our device.
mounted in vertical direction
mounted in horizontal direction
(in this image at the center is our device)
the third device is for refference (blue one) , which has same sansor that we are using in our device.
Data brief of that device: https://www.st.com/resource/en/data_brief/steval-stwinkt1b.pdf Analog vs. Digital MEMS Accelerometer Output Comparison
Hello,
We are developing a condition monitoring device and are currently testing its performance using the SPM Vib-15 as a reference, which is a well-known industrial condition monitoring device.
SPM Vib-15 Vibrameter (Analog Device) :
Link to datasheet: Here
Our Device with ST Accelerometer Sensor (IIS3DWB Digital Sensor):
Link to the datasheet: Here
IIS3DWB Configuration :
Accelerometer ODR: IIS3DWB_XL_ODR_26k7Hz
FIFO BDR: IIS3DWB_XL_BATCHED_AT_26k7Hz
Effective ODR: 26.9 kHz
Full Scale: ACCELERO_FS_16
Hardware Filter: IIS3DWB_HP_ODR_DIV_800
Issue :
We are measuring the acceleration RMS and velocity of a compressor/chiller/motor in an industrial setting during normal operation. Despite the IIS3DWB sensor providing output in all three axes, we are considering only the Y-axis, which aligns with the SPM Vib-15.
We have taken readings for acceleration peak, acceleration RMS, and speed RMS using both our device and the SPM Vib-15. However, our device’s readings are significantly different from those of the SPM Vib-15, particularly at higher vibration levels (almost double). The acceleration RMS values have been converted to G to match the units used by the vibrameter, and velocity RMS values are already in mm/s for both devices.
Problem :
Despite similar measurement conditions and proper mounting of both devices, our readings do not match the SPM Vib-15 readings. Given that both devices serve the same application, should we expect the digital sensor data from the IIS3DWB to closely align with the analog sensor output from the SPM Vib-15?
Are we making an error by comparing the outputs of an analog sensor to those of a digital sensor? What could be causing these discrepancies?
We have also attempted linear regression for calibration, but it did not resolve the issue.
Additional Information :
Images:
First from the right side is the SPM vibrameter, and last one (biggest one) is our device.
mounted in vertical direction
mounted in horizontal direction
(in this image at the center is our device)
the third device is for refference (blue one) , which has same sansor that we are using in our device.
Data brief of that device: https://www.st.com/resource/en/data_brief/steval-stwinkt1b.pdf predictive_maintenance, condition_monitoring, spm_vib_15 MATLAB Answers — New Questions
