Modal identification of machine tool spindle units by output only operational modal analysis

Area of Science:

• Mechanical Engineering
• Vibration Analysis
• Condition Monitoring

Background:

• Accurate modal characteristics are crucial for machine tool spindle condition monitoring.
• Conventional experimental modal analysis (EMA) is invasive and impractical for production settings.
• Impact hammer testing, a key EMA method, disrupts operations.

Purpose of the Study:

• To propose a novel operational modal analysis (OMA) technique for machine tool spindle identification.
• To enable condition monitoring using output-only vibration measurements during operation.
• To validate the proposed OMA method against conventional EMA.

Main Methods:

• Utilized output-only vibration measurements during milling with variable spindle speed.
• Applied two standard OMA methods: stochastic subspace identification (SSI) and frequency domain decomposition (FDD).
• Compared OMA results with conventional EMA from impulse hammer testing on static and operational spindles.

Main Results:

• The proposed OMA method showed less than 10% difference in natural frequencies compared to conventional impulse hammer testing.
• For operational spindles during cutting, the OMA method yielded less than 3% difference.
• Demonstrated the practical applicability of the OMA technique in production environments.

Conclusions:

• The developed OMA technique provides accurate modal identification for machine tool spindles.
• This method offers a non-invasive and practical approach for real-time condition monitoring.
• The findings support the implementation of OMA in industrial production settings.