Date of Award

2008

Document Type

Master Thesis

Degree Name

Master of Engineering (Research)

Department

Manufacturing, Biomedical and Facilities Engineering

First Advisor

Mr. Daithi Fallon

Abstract

Boston Scientific produces the Guglielmi Detachable Coil in their Cork based facility. These are used in a surgical procedure to eliminate aneurysms. The first step in the process is to form a coil by wrapping a platinum wire around a mandrel. These coils have outer diameters ranging from 190 pm to 520pm. This is called the primary wind and is inspected to find any defects introduced into the coil from the winding process. These defects may be the result of wire defects in the raw material or possibly poor handling by the operators. These primary winds are processed further to form aneurysm embolisation coils. Any defect in the coil could result in coil breakage during the procedure and cause a piece of the coil to flow to an undesired part of the body. The current method of inspection involves manually examining the coils under a microscope. The operator examines the coils for differences in the reflection caused by non-uniformities of the coil. This method is quite subjective and is also very demanding on the operator. A more robust and reliable method of inspection is required. Direct optical microscopy falls short and fails to provide a reliable medium for automated inspection.

Diffraction occurs when the coil is illuminated by a collimated laser beam. In this project relationships between the Fraunhofer diffraction pattern and the coil have been mathematically deduced. These relationships are exploited in order to use diffraction as a means of examining the coils. This diffraction methodology also allows for automated analysis. The diffraction method overcomes the shortfalls experienced in direct optical microscopy. This diffraction system will produce more accurate results than those found in direct magnification, when components of similar quality are used. This is due to the sensitivity of the diffraction pattern to small changes in the coil size.

Images for all available coils were captured using a customised diffraction rig. Those sections of the coil which created the diffraction images were also physically measured for calibration and validation of the system. The measurement outputs for the calibrated diffraction system are compared to the measured parameters. The difference in these measurements is less than 0.1%. This project also uses an image comparison algorithm to find small defects which are not necessarily detectable through dimensional measurement. A suite of algorithms were developed within Lab VIEW for this purpose.

All the available coils have been inspected and measured; the system successfully identified all the defects and measured the coils accurately. Diffraction is therefore a viable method for measuring coils of this size which are difficult to measure with direct microscopy. This project has resulted in a system which will automatically inspect a single point on the primary wind coil. Further recommendation work should involve developing an automated system for moving the coil past the inspection point thus enabling inspection of the coil along its full length.

Access Level

info:eu-repo/semantics/openAccess

Included in

Manufacturing Commons

Share

COinS