I have learned some practical pointers from recently performed integrity surveys at several stages of construction and immediately after completion of the following lining system, from the top down:
- 24 in. (600 mm) compacted < 0.75 in. (19 mm) stone
- Geotextile/geonet/geotextile composite (geocomposite)
- Primary 60 mil (1.5 mm) HDPE geomembrane
- GCL
- Geocomposite
- Secondary geomembrane
- GCL
- Tertiary geomembrane
- Cushion geotextile
- Prepared subgrade
The geocomposite over the subgrade, although wetted prior to covering with the tertiary geomembrane, was not sufficiently wetted (to make it electrically conductive) to give uniform leak indication signals. Nor was there sufficient condensation under the liner to provide a conductive layer. Where not adequately wetted, it took approximately 2 or 3 seconds for the water from the water lance to penetrate the geotextile (through a small hole/cut in the liner) for a signal to register. With some wrinkles in the liner, a survey on the tertiary liner would be unreliable, so it was discontinued.
To ensure that the GCL under the secondary liner would be adequately conductive, it was lightly irrigated with water immediately before placement of the geomembrane. While GCLs usually are reasonably conductive, we needed to ensure there was no doubt about adequate conductivity in this critical project. A water lance survey was performed over a small hole placed in the geomembrane, with the current return electrode (4 in. square plates) clamped to the edge of the GCL between tertiary and secondary geomembranes. The GCL was wetted to ensure good contact with the plates. The calibration hole would not adequately register using the water lance, even with 500 VDC applied between water in the secondary sump and the GCL. The potential was then applied between the GCL and another electrode placed at the edge of the GCL about 200 ft away. There was very little current flow. The initial GCL electrode was then moved to another location in the corner of the cell where the GCL had not been exposed to the sun – there was good current flow and the calibration hole could easily be "seen." Thus, placement of an electrode on a GCL is important, as is assurance of surrounding conductivity.
A test pad was constructed using the top four layers on a soil subgrade to define survey parameters for the final soil surface survey. Calibration holes of 0.125 in. and 0.25 in. were made in the geomembrane prior to covering. Both the GCL and the geocomposite were wetted prior to covering. Again, we found that the placement of the current return electrode on the edge of the GCL was important to generating adequate current flow. At the location of largest current flow, there was no difference when the electrode was placed in the soil subgrade. However, even when there was adequate current flow, it was not possible to see a sufficient leak signal above the known holes.
The depth of soil was decreased to 6 in. (150 mm), and rainwater collected on the liner. All holes could be adequately identified and marked to within 0.5 in. of their actual locations. The soils were wet as was the geocomposite – silty rivulets were on the geonet strands. However, there was no standing water on the geocomposite. It was concluded that the geocomposite would have to be very wet, as would the overlying soil, and that perhaps a maximum soil thickness of 12 in. (300 mm) could be tested.
The subsequent production surveys were successful in finding small leaks.
For more information, questions or comments, please contact Ian Peggs.
