Product Name: Guardian, an NTN Monitoring Solution

Manufacturer: NTN Bearing Corporation of America

Product Category: Test, Measurement, Sensors and Software

Supporting Documentation (if available)

Effective predictive maintenance depends on collecting the right data with the right technology to deliver actionable insights. The need for scalable deployment has driven adoption of wireless condition monitoring, although many wireless solutions trade diagnostic data quality for ease of installation and use. Furthermore, modern industrial facilities increasingly rely on variable speed and intermittently operated equipment, creating challenges for traditional monitoring approaches that depend on assumed operating conditions or limited data context.

Guardian is a complete, wireless, cloud-based condition monitoring system designed to simplify machine monitoring implementation while delivering high quality, contextualized data for accurate fault detection, reliable diagnostics, and effective machine health monitoring across a wide range of industrial assets.

Key innovations and features of Guardian include:

3-in-1 Wireless Sensor – Tri-Axial Vibration, Temperature, and Running Speed
• Overview: Full waveform, tri-axial vibration measurement is essential for early fault detection and accurate diagnosis of issues that may be directional in nature. Temperature provides additional context on fault severity and progression. Accurate running speed is a critical supporting parameter for effective vibration analysis and machine health assessment.
• Technical Approach: Guardian Pro sensors employ MEMS accelerometers to capture full waveform, tri-axial vibration data with a frequency response up to 6 kHz, enabling early detection of bearing and gear defects. Configurable measurement settings allow users to adjust maximum frequency (Fmax) and spectral resolution, supporting resolution down to 0.125 Hz for accurate separation of closely spaced fault frequencies. Surface temperature is measured concurrently, and integrated magnetic flux sensing is used to compute true machine running speed.
• Performance Impact: By integrating vibration, temperature, and running speed into a single wireless sensor, Guardian simplifies installation while simultaneously collecting critical machine condition data in context of speed. This combination enables earlier and more reliable fault detection, improves diagnostic accuracy, especially on variable speed machinery, and reduces the complexity and cost of deploying predictive maintenance at scale.

Magnetic Flux-Based Speed Detection:
• Overview: Accurate running speed is critical for effective vibration monitoring and fault diagnosis. Variable speed machines challenge wireless systems that rely on assumed or estimated running speeds, reducing diagnostic accuracy and increasing false or missed alarms.
• Technical Approach: Guardian Pro wireless sensors use an integrated magnetometer to directly measure a motor’s magnetic flux field. This signal is processed to calculate the machine’s true running speed without relying on assumptions.
• Performance Impact: Magnetic flux speed detection enables reliable monitoring of variable frequency and VFD driven assets, improves diagnostic accuracy, and reduces false alarms caused by incorrect speed assumptions.

Smart Measurement Triggering
• Overview: Many industrial machines operate intermittently or under changing conditions. Fixed, schedule-based measurements may capture data when machines are not running or miss critical operating states entirely.
• Technical Approach: Guardian Pro sensors use edge-based detection at the sensor level to identify machine operating states. Measurements are triggered only when machines are running at defined speeds or conditions, ensuring relevant data capture. Sensors coordinate across the drivetrain to enable synchronized measurements at those operating states.
• Performance Impact: State-based measurements improve data relevance, allow for better fault detection, and reduce missed events and unnecessary alarms.

Machine Learning Analytics
• Overview: Condition monitoring systems generate large volumes of vibration data that can be difficult and time consuming to interpret. Traditional systems often rely on overall trend thresholds (e.g., RMS, peak, crest factor), which can be ineffective under changing speed or load conditions and require ongoing manual tuning.
• Technical Approach: Guardian machine learning models are trained on the unique operating behavior of each asset. The analytics evaluate more than 30 vibration features, including speed related frequencies, bearing and gear fault frequencies, vane pass frequencies, and band energies. These features are combined with running speed data to generate numeric asset and bearing health scores. Supervised learning options allow users to tailor models to specific machines and operating environments.
• Performance Impact: By analyzing machine specific frequencies using reliable speed data, Guardian’s machine learning analytics outperform traditional threshold-based methods and anomaly detection alone. The result is earlier fault detection, fewer false alarms, reduced setup effort, and more reliable health indicators under variable conditions.

Diagnostic Engine:
• Overview: Successful predictive maintenance requires not only detecting abnormal behavior but also understanding the nature of the machine problem and what corrective action to take. Alerts without diagnostic context reduce confidence and limit ROI.
• Technical Approach: Guardian’s diagnostic engine uses physics-based models and a library of rules to continuously evaluate vibration patterns associated with specific fault conditions, such as bearing defects, gear defects, misalignment, soft foot, imbalance, and flow-related anomalies. Users can also define custom diagnostic rules for specialized assets.
• Performance Impact: Guardian goes beyond anomaly detection to identify likely fault conditions and guide corrective action, improving maintenance planning, efficiency, and repair effectiveness.

Expert Vibration Analysis Tools
• Overview: Advanced vibration analysis tools are essential for validating automated analytics, investigating complex machine behavior, and supporting root cause analysis.
• Technical Approach: Guardian provides a comprehensive suite of expert analysis tools, including FFT spectra, time waveform analysis, waterfall plots, Cepstrum, demodulation, narrowband enveloping, and other advanced visualizations.
• Performance Impact: Integrated expert analysis tools enable analysts to investigate abnormal machine behavior and validate automated diagnostics directly within the monitoring platform, allowing for precise diagnosis and confident maintenance action.

Guardian Pro Wireless Sensor Specs:
Vibration
• Frequency Response: 2 – 6,000 Hz
• Measurement Range: +/- 2g to +/- 16g
• AD Conversion: 16 Bit
• Resolution: Up to 6,000 lines of resolution
• Sampling Rate: Up to 26.7 kHz

Connectivity & Integration
• 802.11 b/g/n Wi‑Fi (2.4 GHz), Bluetooth Low Energy
• REST APIs for CMMS, EAM, and historian integration

Security
• WPA3 / 802.1X network security
• TLS 1.2 with AES‑encrypted data transmission

Environmental & Mechanical
• IP67 enclosure, SS316 base
• Epoxy or stud mounting
• Operating range: −20°C to 75°C

Power
• Battery‑powered (2× CR123A), typical life 1–5 years