Quantifying the strength of 40-year old lattice towers and pyramid foundations

The overhead transmission line network in England and Wales utilises more than 22,000 steel lattice towers. These towers have anticipated technical asset lives that range between 30 and 120 years, depending on their operating environment (eg coastal proximity) and maintenance history (painting frequency). The life-limiting process of the foundations is not fully understood but the foundations are expected to last the life of the tower in all but exceptional circumstances. The original conductors typically have a 40-to 50-year life and many routes are now being refurbished by stringing new conductors on existing towers. The strength of the existing towers and foundations is therefore being re-analysed using modern design standards and techniques to allow for new, higher capacity (and hence often physically larger) conductors, and to ensure that they will remain in generally 'good' condition for a further 40 years. In assessing the amount of work required to recover towers to this condition, a visual grading system has traditionally been used. On-site and laboratory assessment subsequently revealed that this visual grading system is not reliably indicating the hidden condition and residual life of steelwork. Therefore, a new technique has been developed. A computer model is used to evaluate the conditions that affect corrosion throughout the life of a tower at its actual geospatial location. The model combines environmental parameters (weather and pollution data) with the tower's construction and maintenance history to calculate the likely degradation year-on-year since construction. For each tower, the historic and current condition of the tower steelwork can be determined via a probabilistic prediction of the loss of galvanising and subsequent loss of steel section. This model is run for every year of the life of the tower to assess current condition, and can be projected forward. The value of the model is further increased by the inclusion of a risk module, which defines the consequence of a tower failure based on the terrain over which the adjacent spans pass and the tower-specific loading factors that increase the risk of failure. The output of the model (in terms of the mean predicted loss of steel section on a tower) can then be used in a finite-element analysis model of the tower structure to assess current strength and predict future performance. This allows an estimate to be made of the future date at which individual tower members will fall below the strength required for their operating environment, allowing remnant life to be predicted. In the case of foundations, modern design standards indicate that the uplift capacity of standard "pyramid"tower footings is over-predicted using the conventional UK 'frustum' design. However, at present, failure of these foundation systems is extremely rare, prompting investigations to identify additional factors not considered in these simplified design methods that may provide increased resistance to uplift. Determination of the uplift capacity of foundations and foundation systems under typical in-service loading conditions is difficult. Previously, Clark et al [1] demonstrated that the rate at which loads were transmitted to a foundation following a simulated broken wire event would lead to a fully undrained response in the clay beneath a footing. Lehane et al [2] demonstrated, using scaled centrifuge modelling techniques, that suctions in excess of 70 kPa could be generated across the base of the footing when subject to load uplift rates sufficient to ensure fully undrained conditions. Further enhancements to footing uplift capacity were identified by Rattley et al [3] who quantified loading rate effects on soil strength and stiffness. These factors lead to significantly increased foundation capacities that are not currently considered in assessing the resilience of such foundations to "fast"events. This paper presents aspects of the tower assessment and predictive modelling methodologies, and a simplified numerical modelling procedure to predict foundation capacity for both drained and undrained conditions.