What is Patented Undersampling? Differences from Bearing Diagnosis (Envelope FFT)

In predictive maintenance of bearings, accurately reading "signs of anomalies" from vibration waveforms is crucial. Let's first look at what happens inside a bearing when damage occurs.

Relationship Between Bearing Damage and High-Frequency Impact Vibration

When damage such as scratches or flaking occurs on the inner or outer ring of a bearing, a "damped impact vibration" (a fine vibration that echoes like a tapping sound) in a high-frequency band of 2 to 8 kHz is generated every time a ball (rolling element) passes through the damaged part.
The interval at which these impact vibrations continuously occur is called "TS (characteristic cycle of bearing damage)." Since TS matches the pass frequency calculated from the bearing structure and rotation speed, if this "TS" can be found in the vibration data, the location of the bearing damage can be accurately identified.

Why is FFT of Raw Waveforms Insufficient? (The Need for Envelopes)

You might think, "If we have raw vibration data, can't we just run a frequency analysis (FFT) to find the TS?" However, diagnosis is not possible with just that.

As shown in the figure above (right), even if the fine high-frequency raw waveform in blue is analyzed by FFT as is, the information is buried in the cluster of fine waveforms, and the essential "TS (characteristic cycle)" cannot be found.

This is where "Envelope processing" becomes necessary.
As shown in the figure above, only the outer cover (envelope: red line) is extracted by connecting the peaks of the jagged blue raw waveform.
By performing FFT analysis (Envelope FFT) on this red envelope waveform, the peak corresponding to TS appears clearly as a "sharp single peak" for the first time, making it possible to accurately identify bearing damage. This is the foundation of conventional precision diagnosis.

The Breakthrough of the Patented "Undersampling" Technology

Conventional technology required an "expensive high-speed sampling sensor" to create a clean red envelope. However, conanair overturned this common sense with its patented "Undersampling" technology.

Undersampling is a technology that intentionally acquires (thins out) vibration data at a low sampling cycle.
When high-frequency vibrations are measured at a low cycle, the original waveform breaks down and is recorded as a different waveform called "aliasing noise" (the orange dotted line in the figure above).
Normally, this noise is avoided as a measurement error, but conanair actively utilizes it as a "sign that high-frequency abnormal vibrations are occurring."

The waveform containing aliasing noise (orange dotted line) is not the correct waveform, but the most important information—the timing of the "cycle where the impact occurs (TS)"—is not lost.
By applying envelope processing and then an FFT to this thinned-out alias waveform, the bearing damage cycle (TS) can be identified exactly as when measured with conventional expensive equipment (the green dotted peak in the figure above).

Summary: High-Precision Diagnosis at Low Cost with Lightweight Data

Through this paradigm shift, conanair achieves high-precision bearing diagnosis with a general-purpose component configuration while minimizing data size to the absolute limit.

• Drastically reduces costs by using inexpensive general-purpose sensors instead of expensive high-frequency sensors.
• Long-term wireless operation is possible on commercial batteries alone by thinning out (lightening) the data, and the low communication load makes it highly suitable for IoT and DX.
• Because raw data waveforms can be saved to the cloud or locally, it avoids becoming a black box, allowing engineers to verify the waveform at any time.

By focusing on the "cycle" of the anomaly rather than absolute numerical measurement, it realizes an optimal lightweight, low-cost, and high-precision system for on-site predictive maintenance.

30-Day Free Trial & Inquiry

Try conanair for free for 30 days. Here is what our users are saying:

• Easy installation for immediate use
• No app required; operate via web browser
• No cloud required for fully automatic measurement
• Affordable, including dedicated software

Try equipment anomaly detection and predictive maintenance for free. Please feel free to contact us.

30-Day Free Trial & Inquiry

Try conanair for free for 30 days. Here is what our users are saying:

• Easy installation for immediate use
• No app required; operate via web browser
• No cloud required for fully automatic measurement
• Affordable, including dedicated software

Try equipment anomaly detection and predictive maintenance for free. Please feel free to contact us.

前のページ

Patented "Under-sampling" Technology: Mechanism and Clear Technical Limits

次のページ

Overcoming Manufacturer Dependency via CSV Raw Data & Python