The overall program to minimize defects in the final coated product consists of several steps,
• Optimizing the formulation and coating and drying conditions in the scale-up process from laboratory to manufacturing
• Maintain product quality by statistical quality control procedures and on-line measurement of coating weight and surface inspection
• Utilize a troubleshooting solving procedure to determine cause of defects and eliminate them, when they occur in routine production
• Modify process after defect incidents to prevent their reoccurrence
This blog will only discuss the troubleshooting procedure.
The troubleshooting procedure starts when a defect is detected either by visual observation of the on-line surface inspection system in the coating line, analysis of yield data indicates a problem, defects detected in converting process or by the customer complaints. At this point there should be agreement by all functions that the observed defect needs to be eliminate After the agreement to eliminate an estimate of the resources, people, analytical testing and data analysis that are need should be made and then allocated. A decision not to proceed can also be made if the expense is not warranted. It can be better to just scrap defects,
The second step is to define the characteristics of the observed defect. A start middle and end sample should be obtained samples of the roll stock containing the defect. If this is not possible then sheet samples should be used. A good practice is to save roll stock sample so that defects are detected after the coating can be traced back to their location on the web. Based on the observed defect a description of the defect in as much detail as possible should be developed. This would include size, type of defect, spots, linear and direction shape, frequency and severity. The final step is to prepare a program statement that describes the defect and the anticipated results form the subsequent steps. The defect should be given a name until the defect is characterized in the next step. The reason for this is that giving the defect a name implies a cause and direction for the program to eliminate the defect. This may not be accurate while defects appear to be similar; they have very different causes defects. The incorrect name can lead to wasted effort and a longer time then necessary to eliminate the defect.