Uncovering Vibration Monitoring VOYAGER4 Essentials for Practicing Engineers

From automation to industrial systems, electric motors are crucial for driving essential processes in a wide range of applications. Any failure or performance degradation in a motor can result in unwarranted downtime, which can hamper productivity on the factory floor, triggering substantial delays and disruptions in the manufacturer's supply chain and creating substantial losses for the company. Besides losing time and money, the undesired downtime also tarnishes the manufacturer's image in the market.

Consequently, to ensure that the motor functions appropriately throughout the life cycle of the system, the health and performance of these machines must be constantly monitored in systems where they are deployed. This kind of predictive maintenance of machines minimizes failures, improves reliability, and increases productivity on the factory floor. All this translates into significant savings for the company.

While there are several rotating machine parameters to monitor, vibration is the most important and useful characteristic to examine and determine the health of the rotating machine. It is a key predictive variable that can be used to monitor and detect potential faults like soft footing, bearing, and other similar issues within rotating machinery. Although vibration is not difficult to monitor, gathering data and reporting it meaningfully are not trivial. It requires data analysis, novel algorithms, and wireless connectivity.

Monitoring motor vibrations

For such an application, Analog Devices, Inc. (ADI) has developed a wireless vibration monitoring sensor using microelectromechanical systems (MEMS) accelerometer sensing technology. Recognized for its small size, low-power consumption, and a broad frequency response that goes up to 8 kHz, MEMS sensors are the preferred technology for a wide array of industrial rotating machines.

Designed for condition-based monitoring (CbM) in robotics and industrial applications, ADI’s new generation MEMS sensor, labeled VOYAGER4, incorporates edge artificial intelligence (AI) for more intelligent data analysis at the sensor level. In fact, it is a complete solution that includes supporting ICs, components, and other devices like accelerometers, processors, and power management ICs (PMICs) (Figure 1).

Figure 1: The complete VOYAGER4 system block diagram. (Image Source: Analog Devices, Inc.)

VOYAGER4 evaluation kit

To simplify engineers' understanding of a wireless condition monitoring system, ADI has prepared the VOYAGER4 wireless vibration monitoring evaluation kit, EV-CBM-VOYAGER4-1Z. This kit is a complete, low-power vibration monitoring platform that enables engineers to rapidly deploy a wireless monitoring solution for an electric motor or a similar test setup. It incorporates the following:

• Intelligent, smarter, and secure decisions at the edge
• An AI algorithm for making decisions at the edge
• Mechanical mounting and measurement capability up to 8 kHz bandwidth
• 3-axis ultra-low power, ultra-low noise MEMS accelerometer technology
• Ultra-low power microcontroller and robust low-power Bluetooth low energy (BLE) technology         

ADI ICs and other parts mounted on the printed circuit board (PCB) of the kit (Figure 2) include the ADXL382 and ADXL367 triaxial digital-out MEMS sensors, the BLE MAX32666, MAX78000 AI microcontrollers, PMIC MAX20335, and power devices MAX17262, and MAX38642. The populated PCB is vertically mounted on an aluminum base with a battery attached to a standoff. It also includes an M6 threaded hole in the base for screw stud mounting to a motor casing. The entire unit is then encased in an aluminum enclosure with a diameter of 46 mm and a height of 77 mm.

Figure 2: The populated printed circuit board of the EV-CBM-VOYAGER4-1Z on its aluminum base. (Image Source: Analog Devices, Inc.)

To avoid shielding the antenna of the BLE connection, the enclosure uses a lid made of ABS plastic (Figure 3). It is a strong non-metallic material that allows radio signals to pass through with minimal interference.

Figure 3: The VOYAGER4 sensor unit’s mechanical assembly with its aluminum enclosure and ABS plastic lid. (Image Source: Analog Devices, Inc.)

Employing modal analysis, ADI engineers have designed a sound mechanical enclosure that enables the VOYAGER4 sensor to precisely extract sensitive vibration data from the motor or the rotating machine under test. For that, it uses an edge AI algorithm to detect anomalous motor behavior and trigger a call for machine diagnostics and maintenance. However, before the software begins its diagnostic process, the 16-bit, 8 kHz triaxial MEMS accelerometer sensor ADXL382 is deployed to gather the vibration data. The raw vibration data collected is then processed using the MAX78000 AI processor. If the AI algorithm detects a fault or suspects the vibration data is errant, the system sends a vibration anomaly alert to the user over the MAX32666 wireless BLE radio.

Sensor system operation

In principle, the VOYAGER4 sensor system processes the initial vibration data in a well-defined method (Figure 4). As depicted in the figure, the raw data gathered by the MEMS sensor follows path (a) to the BLE processor. However, before sending it to the user via the BLE radio or USB connection using the FT234XD-R USB to basic serial UART interface IC, this MEMS data is sent to the processor with edge AI via path (b) to predict faulty machine data. If the AI algorithm predicts faulty or suspects the vibration data, the system uses path (c) to alert the user regarding the anomalous data over the BLE radio. If a fault or no anomaly is predicted, the VOYAGER4 system uses path (d) to send the MEMS sensor into sleep mode until the next detection event.

Figure 4: The VOYAGER4 system operating principle. (Image Source: Analog Devices, Inc.)

The system uses two MEMS accelerometers for a reason. While the high-performance ADXL382 MEMS accelerometer is used to capture vibration data, the ultra-low-power, 14-bit, 100 Hz ADXL367 can be utilized to wake up the BLE radio from a deep sleep mode when a significant vibration or shock event occurs. This wake-up device consumes only 180 nA, contributing to substantial power savings to extend battery life. Concurrently, the MEMS raw vibration data is routed to either the MAX32666 BLE radio or the MAX78000 AI microcontroller via a single-pole double-throw (SPDT) analog switch, the ADG1634BCPZ-REEL7. This analog switch is controlled by the BLE microcontroller.

Other peripheral devices connected to the MAX32666 BLE microcontroller include the MAX17262 multi-cell fuel gauge IC , the MAX3207EAUT+T transient voltage suppression (TVS) diode array, and the DS28C40ATB/VY+T secure authenticator device. While the Li-ion fuel gauge IC implements the Maxim ModelGauge m5 EZ algorithm to monitor battery current, the low-input capacitance TVS diode array provides ±15 kV ESD protection per human-body and air-gap models. Likewise, for data integrity, the secure authenticator provides a core set of cryptographic tools derived from integrated asymmetric (ECC-P256) and symmetric (SHA-256) security functions.

Managing power consumption and battery life

To minimize power consumption, the VOYAGER4 intelligently manages the operation of on-board PMICs with respect to the operational modes of the BLE microcontroller and the edge AI processor. In essence, the BLE microcontroller enables or disables the individual outputs of the MAX20335 PMIC for different VOYAGER4 operating modes. The MAX20335 offers two ultra-low quiescent current buck regulators and three ultra-low quiescent current low-dropout (LDO) linear regulators (Figure 5). The value of each output voltage can be programmed using the PMIC’s I²C interface. If additional power is needed, the kit provides an adjustable, single-output, positive-voltage buck regulator, the MAX38642AELT+T, that can source up to 350 mA.

Figure 5: The MAX20335 block diagram. (Image Source: Analog Devices, Inc.)

For minimal power consumption, the VOYAGER4 sensor adjusts its power mode features between active and inactive states, depending on the BLE and AI operating modes. For instance, in training mode, the BLE microcontroller must first advertise its presence in the BLE network and then make a BLE connection with the network manager. The VOYAGER4 then streams the ADXL382 MEMS raw data over the BLE network for training an AI algorithm on the user’s PC. In normal AI mode, the BLE radio advertising, connection, and streaming features are disabled by default. Concurrently, at periodic intervals, the MAX78000 wakes up and runs an AI inference. If no anomaly is detected, the VOYAGER4 returns to deep sleep mode (Figure 6).

Figure 6: The VOYAGER4 sensor’s average power consumption with time between events. (Image Source: Analog Devices, Inc.)

Figure 6 shows that when a sensor does not transmit raw data over the BLE radio, it consumes up to 50% less power. In training mode, approximately 0.65 mW of power is consumed when the BLE radio is active, advertising, connecting, and transmitting data once per hour. When the sensor operates in normal AI mode, the system consumes 0.3 mW, even when the sensor is active once every hour. Data analysis indicates that with 0.3 mW power consumption, a single 1,500 mAh battery can provide operation for two years. However, using two standard AA, 2.6 Ah batteries can extend battery life to about seven years. For longer durations, using battery cells that feature low base currents and periodic pulses is recommended.

VOYAGER4 GUI and firmware

The VOYAGER graphical user interface (GUI) is written in Python and uses key libraries, such as bleak, asyncio, and Tkinter to enable an interactive interface that connects to the VOYAGER4 sensor over the BLE radio.

The VOYAGER4 evaluation kit includes two microcontrollers and several peripheral devices, including sensors, PMICs, flash memory, and communication interfaces. ADI offers tools to develop the code needed to control and communicate with the host PC. For instance, engineers can leverage CodeFusion IDE for overall embedded development and the VOYAGER SDK for AI application deployment. Specifically, for the MAX32666 and MAX78000 microcontrollers, dedicated developer resources are available to program these devices.

Conclusion

ADI’s wireless vibration monitoring sensor, VOYAGER4, is an effective tool for condition-based monitoring of motors in robotics and other rotating machines within industrial systems. ADI’s evaluation kit enables engineers to understand and apply the MEMS sensor by offering a complete, low-power platform for the rapid deployment of wireless vibration monitoring.