Timken Wind Energy Knowledge Center - Blog Details
 Ron Kittle

High Accuracy At Low Speeds: Timken's New Online Intelligence System For Wind Turbines Acts As 'Early Warning' System

3/28/2011 9:00:00 AM | Sustainable Solutions for Wind Energy | Comments | Ron Kittle |

What if you could identify a wind turbine bearing problem before it leads to failure, especially in low-speed applications? Think of the cost savings and convenience early detection would provide.

This is where condition monitoring systems come into play. By tracking your turbine’s performance on an ongoing basis, you can detect anomalies that can signal potential problems long before they lead to critical failures.

This allows repairs to be implemented on your schedule so the difficult logistics of wind turbine maintenance can be streamlined.

For the last 40-plus years, vibration analysis on rotating equipment has been a vital part of condition monitoring. However, the technology used in this analysis has not advanced significantly over this time. Traditional vibration measuring techniques remain unsuitable for machines running at speeds of less than 60 RPM. With typical rotor speeds between 10 RPM and 25 RPM, monitoring wind turbine main shafts simply isn’t feasible with traditional techniques.

Timken has a successful history of meeting condition monitoring challenges with new innovations. Several years ago, the industry identified a need for a condition monitoring system that could be used in places and applications in which hard-wired sensors were not feasible. Timken responded with its highly successful StatusCheck® wireless condition monitoring system. Similarly, we have now addressed the need for monitoring of low-speed machines by introducing our new Online Intelligence System (OIS).

The new Timken OIS utilizes a revolutionary, high-frequency measurement technique that is especially suited for low-speed applications. It can accurately detect faults, such as bearing and gear damage, and offers four clear advantages over conventional vibration techniques:

• The Signal. Measurements specifically tuned to 32 KHz provide optimum detection and filter out unwanted mechanical noise from the lower frequency range. By utilizing a proprietary-tuned transducer, you get an excited occurrence of resonance at the frequency of interest. This increases signal sensitivity by seven times that of a traditional vibration sensor. The sensor is designed to detect shock or stress waves that are indicative of bearing or gear damage.

• Quality Data. The sample rate through the extended data acquisition period is constantly auto-adjusting to handle any variations in machine speeds. This prevents skewing and provides high quality data for post processing. The data sample rate is controlled by the turning speed instead of time, thus increasing the ability to detect relevant data. A 24-bit A/D converter is used to capture extremely high resolution data needed for processing.

• Intelligent Processing. Advanced algorithms reduce random signal noise and enhance repetitive signals of interest, producing text book-like patterns of data for precise analysis and fault recognition. A patented method of focusing on repetitive signals yields extraordinary signal-to-noise ratios that provide a previously unachievable detailed look into time signals and spectrums.

• Absolute Levels. Easy-to-interpret scalar values are provided, which include a maximum value for accurate trending and alarming, and a carpet value essential to bearing lube film assessment. The evaluated results appear in green, yellow and red formats for at-a-glance assessment of turbine components.

These advantages provide extremely clear and precise spectrum and time signals that allow for easy interpretation and early detection of problems, even at low speeds that traditionally have not been feasible. See for yourself! The following charts will show you examples of very early detection of inner race bearing damage at just 14 RPM, while the shaft was still in rotation. Due to data supplied by Timken's OIS, the bearing was replaced before failure and before any resulting collateral damage to the shaft or other components could occur.

Time signal late April

Time signal before replacement December

Time signal after replacement December

Wind Blog Band Trend Values

In wind applications, it’s important to capture condition-monitoring data during a consistent RPM and power range. This way, you can be sure the data is repeatable and reliable. Timken has the complete condition monitoring solution to uniquely satisfy your wind-energy needs.

What systems do you use to identify potential wind bearing problems in an effort to ensure valuable uptime?



Post a Comment
Submit a Comment

* Required field.

User Name:*

E-mail address (Note: will not be visible to other users.)*


If you can't read this number refresh your screen.
Enter the code shown above:
(Note: If you cannot read the numbers in the above image, reload the page to generate a new one.)




RSS Sign-up

Get real-time blog updates.  
        What's RSS?


Meet Our Bloggers


Brad Baldwin  Gerald Fox

Mike Kotzalas  Dirk Neumann

Hans Landin  Doug Lucas

Chris Marks 



Newsletter Sign-up

Timken Wind Energy Knowledge Center NewsletterGet blog and resource updates via e-mail.


Suggest A Topic

Timken Wind Energy Knowledge Center BlogHave something you want us to blog about? Let us know.