CIPP Collapse: 3D Diagnosis of Failure Due to Poor Curing

One month after a trenchless repair using Cured-in-Place Pipe (CIPP), a sewer pipe catastrophically collapsed. The initial forensic inspection pointed to a premature structural failure. To determine the root cause, a 3D analysis was implemented using laser inspection cameras, generating high-precision point clouds of the interior of the collapsed conduit and adjacent areas.

Forensic methodology: From point cloud to infiltration diagnosis 🔍

The process began with data capture using a laser scanner mounted on an inspection robot. The resulting point clouds were processed in Leica Cyclone to register and clean the 3D model of the pipe. Subsequently, they were exported to CloudCompare to perform a comparative thickness analysis. The distance calculation tool allowed identifying an area at the base of the pipe where the CIPP lining showed irregular thickness and anomalous point density, indicative of a porous structure. This data was cross-referenced with soil temperature records, revealing that cold groundwater had infiltrated the area during the epoxy resin curing process. Finally, the cross-section of the failure was modeled in AutoCAD Civil 3D, quantifying the material strength reduction at the collapse point.

Lessons for buried infrastructure: Preventing the silent collapse 🛠️

This case demonstrates that LiDAR technology and point cloud analysis serve not only to document failures but also to establish real-time quality control protocols during CIPP installation. Early detection of curing anomalies, such as those caused by cold infiltration, can prevent costly collapses and health risks. Integrating 3D scanning as part of the certification process for trenchless repairs is a necessary investment to ensure the long-term integrity of our sewer networks.

Which key parameters of the CIPP curing process, such as temperature, exposure time, or catalyst distribution, should be analyzed through 3D scanning to predict and prevent structural failures like the collapse documented in this case?

(PS: Simulating a collapse is easy. The hard part is not crashing the program.)