Years ago, I was plagued with web breaks on a large, high-speed paper coater. In the motor-control room, we had a full-time, 8-channel chart recorder. It was set to chart at slow speed continuously at all times and chart at a faster speed for every splice.
The chart recorder showed that tension ran along steadily and then suddenly dropped to zero suddenly just at the time the web broke. Most of the breaks occurred at the second coating station.
It didn't take a genius to determine that the problem clearly was the drive system creating the breaks. I spent hundreds of hours monitoring everything on the drive system.
Finally, we upgraded the drive and made progress with solving the web-break problems. The drive upgrade consisted of purchasing four high-speed video cameras and recorders. These recorders were coordinated with web speed and simultaneously showed the same section of the web at different points along the web path. Of particular interest were the cameras at the two coating stations.
The first day the cameras showed a small web defect (hole) passing successfully through the first coater. The hole would then tear apart at the second coater.
The drive system was fixed! The effort then moved to the paper machine, which produced the defects and the web inspection system, which ignored the defects.