Date of Award

12-2011

Document Type

Thesis

Degree Name

Master of Science (MS)

Legacy Department

Mechanical Engineering

Advisor

Mocko, Gregory

Committee Member

Mears , Laine

Committee Member

Vahidi , Ardalan

Abstract

Equipment employed in a manufacturing environment must be able to operate as long as possible having as little downtime as possible. Therefore, maintenance is crucial in order to allow for the equipment to perform its designated tasks without failure, especially on critical systems. In a CNC machine, if the spindle fails, the machine is useless. Having the ability to detect spindle degradation to the point where a replacement spindle installation can be planned, via condition monitoring, is invaluable to a manufacturer who utilizes these types of machines.
An early warning monitoring system for CNC spindle bearing failure has been developed to be utilized directly on a CNC machine's controller employing an open architecture structure. The main system uses an ultrasonic sensor as its primary sensing component and provides a singular value as to the spindle condition. The system allows for both real time data recording as well as provides a trending history for the machine. Additionally, the system allows for the data to be seen remotely via the internet. Accessory devices can be added to perform an in-depth bearing failure analysis. The total system (including accessories) costs just under $2,400, allowing for a very effective system at a very low price. A few thousand dollars towards a predictive and preventive maintenance monitoring solution can prevent tens-of-thousands of dollars in lost production and unnecessary maintenance costs if the system is utilized as intended.
System performance was tested to investigate sensor measurement applicability. Spindle speed was found to have an effect on the sensor's output, however excessive vibration did not. Therefore, the same spindle speed must be used each time a measurement is taken. Measurements while the machine is cutting can be performed, however, a test mode is recommended for the most accurate results. The amount of variation for an in-process reading was found to be lower for a harder material (ie: steel vs. aluminum), for the same spindle speed and depth of cut. The system was tested to see if it could detect the various stages of bearing failure. It was unable to detect a plastic/resin bearing cage degradation failure until it was too late as the failure was too quiet to detect.

Share

COinS