Noise, Vibration, and Harshness (NVH) in Automotive Settings

The analog output ADXL356 and the digital output ADXL357 are low noise density, low 0 g offset drift, low power, 3-axis accelerometers with selectable measurement ranges. The ADXL356B supports the ±10 g and ±20 g ranges, the ADXL356C supports the ±10 g and ±40 g ranges, and the ADXL357 supports the ±10 g, ±20 g, and ±40 g ranges.

The ADXL356/ADXL357 offer industry leading noise, minimal offset drift over temperature, and long-term stability, enabling precision applications with minimal calibration.

The low drift, low noise, and low power ADXL357 enables accurate tilt measurement in an environment with high vibration. The low noise of the ADXL356 over higher frequencies is ideal for condition-based monitoring and other vibration sensing applications.

The ADXL357 multifunction pin names may be referenced only by their relevant function for either the serial peripheral interface (SPI) or limited I²C interface.

• Inertial measurement units (IMUs)/altitude and heading reference systems (AHRSs)
• Platform stabilization systems
• Structural health monitoring
• Seismic imaging
• Tilt sensing
• Robotics
• Condition monitoring

The ADXL1001/ADXL1002 deliver ultralow noise density over an extended frequency range with two full-scale range options, and are optimized for industrial condition monitoring. The ADXL1001 (±100 g) and the ADXL1002 (±50 g) have typical noise densities of 30 μg/√Hz and 25 μg/√Hz, respectively. Both accelerometer devices have stable and repeatable sensitivity, which is immune to external shocks up to 10,000 g.

The ADXL1001/ADXL1002 have an integrated full electrostatic self test (ST) function and an overrange (OR) indicator that allow advanced system level features and are useful for embedded applications. With low power and single-supply operation of 3.3 V to 5.25 V, the ADXL1001/ADXL1002 also enable wireless sensing product design. The ADXL1001/ ADXL1002 are available in a 5 mm × 5 mm × 1.80 mm LFCSP package, and are rated for operation over a −40°C to +125°C temperature range.

Applications

• Condition monitoring
• Predictive maintenance
• Asset health
• Test and measurement
• Health usage monitoring sytems (HUMS)

The ADAQ7768-1 is a 24-bit precision data acquisition (DAQ) μModule^® system that encapsulates signal conditioning, conversion and processing blocks into one system in package (SiP) design that enables rapid development of highly compact, high performance precision DAQ systems.

The ADAQ7768-1 consists of:

• A low noise, low bias current, high bandwidth programmable gain instrumentation amplifier (PGIA) that is also capable of signal amplification and signal attenuation while maintaining high input impedance
• A fourth order, low noise, linear phase anti-aliasing filter
• A low noise, low distortion, high bandwidth ADC driver plus an optional linearity boost buffer
• A high-performance medium bandwidth 24-bit sigma delta ADC with programmable digital filter
• A low noise, low dropout linear regulator
• A reference buffers
• Critical passive components required for the signal chain

The ADAQ7768-1 supports fully differential input signal with a maximum range of ±12.6V. It has an input common mode voltage range of ±12V with excellent common mode rejection ratio (CMRR).

The input signal is fully buffered with very low input bias current of 2 pA typical. This allows easy input impedance matching and enables the ADAQ7768-1 to directly interface to sensors with high output impedance.

The seven pin-configurable gain settings offer additional system dynamic range and improved signal chain noise performance with input signals of lower amplitude.

A 4th order low-pass analog filter combined with the user programmable digital filter ensures the signal chain is fully protected against the high frequency noise and out of band tones presented at the input node from aliasing back into the band of interest. The analog low pass filter is carefully designed to achieve high phase linearity and maximum in-band magnitude response flatness. Constructed with Analog Devices’s iPASSIVES™ technology, the resistor network used within the analog low-pass filter possess superior resistance matching in both absolute values and over temperature. As a result, the signal chain performance is maintained with minimum drift over temperature and the ADAQ7768-1 has an excellent device to device phase matching performance.

A high-performance ADC driver amplifier ensures the full settling of the ADC input at the maximum sampling rate. The driver circuit is designed to have minimum additive noise, error and distortion while maintaining stability. The fully differential architecture helps maximizing the signal chain dynamic range.

The analog to digital converter (ADC) inside the ADAQ7768-1 is a high performance, 24-bit precision, single channel Sigma-Delta converter with excellent AC performance and DC precision and the throughput rate of 256 kSPS from a 16.384 MHz MCLK.

An optional linearity boost buffer can further improve the signal chain linearity.

The ADAQ7768-1 is specified with the input reference voltage of 4.096V, but the device can support reference voltages ranging from VDD_ADC down to 1V.

The ADAQ7768-1 has two types of reference buffers. A pre-charge reference buffer to ease the reference input driving requirement or a full reference buffer to provide high impedance reference input. Both buffers are optional and can be turned off through register configuration.

ADAQ7768-1 supports three clock input types: crystal, CMOS or LVDS.

Three types of digital low pass filters are available on the ADAQ7768-1. The wideband filter offers a filter profile similar to an ideal brick wall filter, making it a great fit for doing frequency analysis. The Sinc5 filter offers a low latency path with a smooth step response while maintaining a good level of aliasing rejection. It also supports an output data rate up to 1.024 MHz from a 16.384 MHz MCLK, making the Sinc5 filter ideal for low latency data capturing and time domain analysis. The Sinc3 filter supports a wide decimation ratio and can produce output data rate down to 50 SPS from a 16.384 MHz MCLK. This combined with the simultaneous 50/60Hz rejection post filter makes Sinc3 filter especially useful for precision DC measurement. All the three digital filters on the ADAQ7768-1 are FIR filters with linear phase response. The bandwidth of the filters, which directly corresponds to the bandwidth of the DAQ signal chain are fully programable through register configuration.

The ADAQ7768-1 also supports two device configuration methods. The user has the option to choose to configure the device via register write through its SPI interface, or through a simple hardware pin strapping method to configure the device to operate under a number of pre-defined modes.

A single SPI interface supports both the register access and the sample data readback functions. The ADAQ7768-1 always acts as a SPI slave. Multiple interface modes are supported with a minimum of three IO channels required to communicate with the device.

ADAQ7768-1 also features a suite of internal diagnostic functions that can detect a broad range of errors during operation to help improving the system reliability.

The ADAQ7768-1 device has an operating temperature range of −40°C to +105°C and is available in a 12mm x 6mm x 1.6mm, 84-ball BGA package with 0.8mm ball pitch which makes ideal multiple channel application. The ADAQ7768-1 utilizes only 75sq mm of board space, 10 times less than the discrete solution that utilizes 750sq mm.

APPLICATIONS

• Universal input measurement platform
• Electrical Test and Measurement
• Sound and Vibration, Acoustic and Material Science R&D
• Control and Hardware in Loop Verification
• Condition monitoring for predictive maintenance
• Audio Test

The AD4000/AD4004/AD4008 are high accuracy, high speed, low power, 16-bit, Easy Drive, precision successive approximation register (SAR) analog-to-digital converters (ADCs) that operate from a single power supply, VDD. The reference voltage, V_REF, is applied externally and can be set independent of the supply voltage. The AD4000/AD40004/AD4008 power scales linearly with throughput.

Easy Drive features reduce both signal chain complexity and power consumption while enabling higher channel density. The reduced input current, particularly in high-Z mode, coupled with a long signal acquisition phase, eliminates the need for a dedicated ADC driver. Easy Drive broadens the range of companion circuitry that is capable of driving these ADCs (see Figure 2 in the data sheet).

Input span compression eliminates the need to provide a negative supply to the ADC driver amplifier while preserving access to the full ADC code range. The input overvoltage clamp protects the ADC inputs against overvoltage events, minimizing disturbances on the reference pin and eliminating the need for external protection diodes.

Fast device throughput up to 2 MSPS allows users to accurately capture high frequency signals and to implement oversampling techniques to alleviate the challenges associated with antialias filter designs. Decreased serial peripheral interface (SPI) clock rate requirements reduce digital input and output power consumption, broadens digital host options, and simplifies the task of sending data across digital isolation. The SPI-compatible serial user interface is compatible with 1.8 V, 2.5 V, 3 V, and 5 V logic by using the separate VIO logic supply.

APPLICATIONS

• Automated test equipment
• Machine automation
• Medical equipment
• Battery-powered equipment
• Precision data acquisition systems
• Instrumentation and control systems

The Wireless Vibration Monitoring Platform is a system evaluation solution for a wireless signal chain for MEMS-accelerometer based vibration monitoring. The system solution combines mechanical attach, hardware, firmware, and PC software to enable rapid deployment and evaluation of a three-axis vibration monitoring solution. The module can be directly attached to a motor or fixture, via a ¼-28 stud. It can also be combined with other modules on the same wireless mesh network to provide a broader picture with multiple sensor nodes, as part of a Condition Based Monitoring (CbM) system.

Solution Overview

The CbM hardware signal chain consists of a three-axis ADXL356 accelerometer mounted to the base of the module. The outputs of the ADXL356 are conditioned and converted to digital with three AD7685 16-bit daisy chain ADCs. The ADC outputs are buffered and transformed to the frequency domain in the ADuCM4050 low power microcontroller and from there streamed to the SmartMESH IP mote. From the SmartMESH chip time domain and FFT data are wirelessly streamed to the SmartMESH IP Manager. The manager connects to a PC and visualization and saving of the data can take place. Data is displayed as raw time-domain data, and FFT data. Additional summary statistics information on the time-aggregated data are available. The full Python code of the PC-side GUI, as well as the C firmware deployed to the module is made available to enable customer adaptation.

The EV-CBM-VOYAGER3-1Z evaluation kit consists of the following items:

• SmartMesh IP Manager (DC2274A) USB Dongle
• Wireless Vibration Monitoring Platform
• JTAG Cable (only needed for firmware upgrade)

The EV-CBM-VOYAGER3-2Z evaluation kit consists of the following items:

• Wireless Vibration Monitoring Platform
• Batteries

Condition-based monitoring (CbM) is one form of predictive maintenance that uses sensors to assess status of equipment overtime while the equipment is operating. The collected sensor data can establish baseline trends, such as, diagnose or even predict failure. Utilizing CbM, maintenance is performed when needed as opposed to the conventional periodic preventive maintenance model, saving both time and money.

Vibration monitoring is a common type of CbM measurement because changes in vibration trends are potentially indicative of wear or other failure modes. To measure vibration data, high bandwidth (10 kHz and more), ultralow noise (100 µg/√Hz or lower) piezoelectric sensors were historically used to satisfy these requirements. The established sensor interface for piezo sensors is integrated electronics piezoelectric (IEPE), as a result, IEPE has become a de facto interface in CbM vibration ecosystem.

With recent advancements in the microtechnology processes and fabrication techniques, MEMS accelerometers have caught up with piezoelectric sensors low noise levels and supersede in other many specifications, such as dc to low frequency response, thermal stability, shock resistance and recovery, and cost. The output of MEMS sensors, however, are either conventional 3-wire analog (ground, power, and signal) or digital if integrated with an ADC. Neither output are directly compatible with IEPE, the preferred CbM industry sensor interface.

This reference design enables a direct piezoelectric sensor IEPE replacement with benefits of high bandwidth, ultralow noise MEMS accelerometers. This circuit allows customers to easily evaluate a MEMS accelerometer for CbM applications.

The reference design shown in Figure 1 shows a high resolution, wide bandwidth, high dynamic range, Integrated Electronics Piezoelectric (IEPE)-compatible interface data acquisition (DAQ) system that interfaces with IC Piezoelectric (ICP^®)/IEPE sensors. The most common IEPE sensors are usually found in applications measuring vibration, but there are many IEPE sensors that measure parameters such as temperature, strain, shock, and displacement.

The focus of this circuit note is on the application of this solution to vibration applications, especially in the area of condition-based monitoring, but there is a large set of applications in instrumentation and industrial automation that use IEPE sensors in a similar way and that are served by similar signal chains.

Condition-based monitoring, in particular, uses sensor information to help predict changes in the condition of a machine. There are many methods of tracking the condition of a machine, but vibration analysis is the most commonly used method. By tracking the vibration analysis data over time, a fault or failure can be predicted, along with the source of the fault.

Vibration sensing in an industrial environment adds additional challenges because of the need for robust and reliable sensing methods. Knowing the condition of a machine helps increase efficiency and productivity and creates a safer working environment.

Most solutions that interface with piezoelectric sensors in the market are ac-coupled, lacking dc and subhertz measurement capabilities. The CN-0540 reference design is a dc-coupled solution in which dc and subhertz precision are achieved.

By looking at the complete data set from an IEPE vibration sensor in the frequency domain (dc to 50 kHz), the type and source of a machine fault can be better predicted using the position, amplitude, and number of harmonics found in the fast Fourier transform (FFT) spectrum.

The data acquisition board is an Arduino-compatible form factor that can be interfaced and powered directly from most Arduino-compatible development boards.

Machine condition-based monitoring (CbM) by means of vibration sensing is growing in importance in industrial applications. Companies are seeking to optimize machinery lifetime and performance and to reduce the cost of ownership, while some are looking to develop new business models around the provision of such information. To provide an accurate representation of machinery that needs monitoring, large datasets must be collected to determine a baseline operating point for the equipment in both normal operating modes and failure conditions. Once this data is collected, an algorithm or threshold detection routine can be created to provide the correct analysis for this equipment.

CbM requires capturing full bandwidth data to ensure that all harmonics, aliasing, and other mechanical interactions in both the time and frequency domain are accounted for. This data collection requires a high performance sensor and data acquisition (DAQ) system that can provide high fidelity, real-time data into a data analysis tool or application.

Using established tools like MATLAB^® or newer Python-based tools like Tensorflow, analyzing the data, profiling the machinery, and creating algorithms for smart decision making is greatly simplified.

Vibration sensing has traditionally dominated most CbM applications because of the availability of sensors, and the science behind the analysis is better understood. The integrated electronic piezoelectric (IEPE) standard is a popular signaling interface standard for high end microelectronic mechanical systems (MEMS) and piezo sensors that are prevalent in the industry today.