Chandra Jalluri

An Advanced Vibration Based Real-Time Machine Health and Process Monitoring System

The ever-increasing emphasis on product quality with increased productivity has been driving the automotive manufacturing industry to find new ways to produce high quality products without increasing production time and manufacturing costs. In addition, automotive manufacturing plants are implementing flexible manufacturing strategies with computer numerical control (CNC) machining centers to address excess capacity, shifting consumer trends and future volume uncertainty of products. Over time, plants have used several preventative and predictive maintenance methods to address machine reliability. Such systems include, but are not limited to, scheduling machine down times at regular intervals to check/replace bearings and other spindle/slide components before they can have an adverse affect on part quality. However, most of these methods and traditional systems are not cost effective and cause significant machine down-times, safety concerns and labor overheads and do not reliably monitor other process issues, such as, clamping, incoming stock variations and thermal phenomena. This paper describes an advanced real-time vibration based machine health and process monitoring system that has been developed to address the above issues. The system, called Condition Indicator Analysis Box for CNC (CIAB™-CNC), is easily configurable, and provides real-time data and historical trends of machines, processes and tooling, enabling manufacturing plants to make accurate predictions regarding future production runs. The system also aids in the optimization of preventative maintenance tasks in a cost effective manner. The developed system monitors machine spindle and slide for unbalance, misalignment, damaged/spalled bearings, mechanical looseness, and ball screw issues. Additionally, it performs in-process monitoring during machining as well as non-machining by individual tool and/or feature to detect tool breakages, quality issues and other gross process or machine anomalies. Innovative statistical trending algorithms enable the system to automatically adapt to valid process/parameter changes and significantly reduce the chances of false alarms and warnings. The developed system provides manufacturing plants with a tool to analyze machine tools and their associated components in an effort to gather information they can use effectively to make decisions regarding flexible machines, processes and tooling.Copyright

Tuned Vibration Based Gear Checker for Gear Profile Anomaly Detection

Significant cost and quality metrics can be realized through the no-faults-forwards approach. This philosophy has put emphasis on early detection of defects through innovative signature analysis techniques within Powertrain Operations production lines. This paper captures the unique approach adopted on gear machining line for functionality testing of individual gears and detection of subtle gear profile anomalies leading to Noise, Vibration and Harshness (NVH) issues in fully assembled transmissions and vehicles. Gears are tested in a Vibration Based Gear Checker (VGC) in single flank engagement and operating under a torque load. The short cycle time enables feasibility for high volume measurements. The structure of Gear Checker is tuned for optimum gear profile anomaly detection**. The tuning is accomplished by structural modifications such that the Gear Checker structural natural frequencies and modes are shifted to enable a high level of excitation at harmonics of Gear Mesh Frequency (GMF) during operation. Impact testing was deployed to guide the structural adjustments for tuning. Gear checker vibration signatures for smooth and defective gears are provided; capability for detection of defective gears is demonstrated. The Gear checker frequency band vibration levels exhibit excellent correlation with the dimensional gear profile measurements by Analytical Gear Checker Coordinate Measuring Machines (CMM’s).Copyright