Prestressing Understrength Walls and Parapets

PRESTRESSING UNDERSTRENGTH WALLS AND PARAPETS
Patrick Jansen and Dr Graham Tilly

PAPER PRESENTED TO ‘STRUCTURAL FAULTS AND REPAIRS - 1999’

PRESTRESSING UNDERSTRENGTH WALLS AND PARAPETS
Patrick Jansen and Dr Graham Tilly
Gifford and Partners
Canton House, Ringwood Road
Woodlands
Southampton, S040 7HT

INTRODUCTION

There are many brick walls and parapets, supported on elevated structures and located at the sides of roads and railways, that are now over 100 years old, some over 150 years, and in need of maintenance. One of the main causes of deterioration is through ageing and degradation of the mortar which is invariably lime based in these older structures.

The walls must be able to withstand wind loading, and where they are located beside roads or railways, there are extra windage effects caused by passing traffic. Additionally, there are traffic induced vibrations which can exacerbate the live loading and accelerate deterioration of the walls by loosening the old mortar. ft is also not uncommon for walls, particularly those beside railway lines, to experience additional dead load effects due to utility pipes being bolted on, see Figure 1

Figure 1 Parapet wall with bolted on utilities

Currently, old walls are failing strength assessments and it is necessary to undertake remedial measures by a suitable method. In addition many older walls are prone to inadequate stability. Differences in flexibility between the walls and their supporting structure, thermal movements, lateral loading effects, bed joint degradation and loss of adhesion can all lead to the walls becoming detached from their substrata. The supporting structure’s contribution towards stability is thus lost, resulting in a reduced factor of safety for stability.

Any strengthening proposal must satisfy the requirements for both strength and stability as necessary. This papers presents such a strengthening solution using Cintec anchors, a system of post-tensioning that has been tailored to meet the requirements of brick walls. An in situ test to confirm the performance of a post-tensioned 100 year old wall in the field is described.

OPTIONS FOR STRENGTHENING

The appearances of old brick walls do not necessarily have intrinsic value on their own, but taken alongside other structures such as adjacent bridges and buildings of similar age, the collection often merits heritage status. It follows that methods of refurbishment and strengthening should be acceptable in a cosmetic as well as structural sense.

In the normal course of events, deep raking out and repointing of the mortar joints using a stronger cementitious material is the most straightforward refurbishment. The lateral bending strength of the wall can be raised some 70 per cent using this method. However, using conventional methods of analysis and assessment, it is difficult to justify that such repairs provide adequate strength and factors of safety. Furthermore, repointing is a time consuming activity made additionally expensive by costs of access and the need to have lane closures or track possessions to satisfy safety requirements. In any event, repointing is unlikely to satisfy the requirements for stability.

In situations where walls are located on top of retaining structures (a fairly common occurrence), strengthening is sometimes carried out by bolting vertical channel-section steel girders to the brickwork and retaining structure, see Figure 2. This is an unsightly method requiring regular maintenance painting and unpopular with heritage bodies. Furthermore, it is often impracticable to fit the girders when utilities are bolted to the walls.

Figure 2 Externally strengthened wall

Post-tensioning the brickwork into the substructure provides a mode of strengthening having none of the above objections. It is quick to carry out, leaves no external evidence on the brickwork and is economic. The post-tensioning can be designed to suit local conditions and strength and stability can be calculated with adequate accuracy.

POST-TENSIONING SYSTEM

The required level of post-tensioning, spacing of tendons and lengths of anchorages are calculated to meet the requirements of the local conditions. The main components of the Cintec system comprise stainless steel tendons, cementitious grout and a sock to contain the grout. Stainless steel tendons are required to provide long term durability in an environment that is akin to post-tensioned segmental bridges where serious corrosion can occur at the mortar joints. Cementitious grout is used in preference to an epoxy based material as it is considered important to have materials that are compatible with the wall. The sock has been designed to contain the grout and prevent any loose brickwork being displaced by the injection pressures of 3 to 4 bar. It also prevents unsightly leakage through cracks that may be present. The sock permits controlled leakage of grout to enable a structural connection to be formed with the surrounding brickwork.

The construction activities are similar to conventional post-tensioning:

Remove capping pieces and bore vertical holes through the centre of the wall to the required depth in the substructure.
Introduce the sock and stainless steel tendon into the anchorage length within the supporting structure and inject grout to form the anchorage.
When the grout has fully hardened, post-tension and lock off against a plate fitted to the top of the wall.
Inject grout into a second sock occupying the remaining space in the masonry wall.
When the second injection of grout has hardened, remove the end piece and plate, replace the capping pieces to leave a cosmetically acceptable appearance to the wall.

STRENGTHENING DESIGN PARAMETERS

No specific standards are available for the assessment or refurbishment of existing masonry parapets. The following parameters have therefore been developed for this Cintec anchor strengthening system:

The post-tensioned wall is designed in accordance with BS562:Part 2 assuming bonded tendons.
Due account is to be given to the make-up and condition of the wall in deriving an appropriate characteristic compressive strength of the masonry.
In accordance with BS5628:Part 2, at the Serviceability Limit State, no tension is to be permitted in the masonry.
Wind loading to BS6399:Part 2 and additional loading due to dynamic pressure and suction from railway traffic derived from European Rail Research Institute Report ERRI D 189/RPI.
Consideration of crowd loading.
Bond stress between the Cintec anchor and masonry or other material is based on pull-out tests on similar materials.
The prestress force must be sufficient to ensure a factor of safety against overturning greater than 2.

The governing criteria in the design of post-tensioned masonry structures is usually the restriction of no tension at the Serviceability Limit State. Accurate values for material properties are therefore not always necessary. This is fortunate since little data appear to be available for typical strengths and stiffnesses of old brick walls.

SUPPLEMENTARY LOAD TEST

A load test was carried out to confirm that the performance of the strengthened wall had been correctly calculated and provide assurance on the method. It was debated beforehand whether to do the test in the laboratory or in situ. A laboratory test has the merit that it can be accurately controlled and loaded to collapse so that the full non-linear load-deflection relationship is recorded. It is not affected by bad weather and a laboratory environment is conducive to good workmanship. On the other hand it was felt that a test on a freshly constructed model would be unlikely to reproduce the true conditions presented by a deteriorated wall containing weakened mortar and weathered bricks. It was therefore decided to test an existing wall in situ and accept the various shortfalls.

The load test was carried out according to the guidelines published by the National Steering Committee for the Load Testing of Bridges1. Although written for bridge testing, the principles of the guidelines are fundamental and generally applicable to other structures. The guidelines define three types of load testing; supplementary, proof and proving tests. In this investigation the load tests were supplementary and, as the name implies, were planned to supplement the structural analysis. The level of loading was to be sufficient to produce measurable responses from the structure without causing any permanent damage.

The available space and access constrained the load test to being as simple as possible. The supplementary load test was undertaken on a section of wall identified as being understrength for wind loading. The brick wall was believed to be constructed from London stocks with a lime mortar and the strengthening scheme was therefore designed assuming a characteristic compressive strength of the masonry of 2.3N/mm2. Built in English bond, the wall was supported on a mass concrete retaining wall. For the purposes of the test a 2m panel was separated from the rest of the wall by vertically saw cutting the parapet down to the top of the retaining wall.

Details of the wall together with the strengthening scheme using two 16mm diameter Cintec anchors are shown in Figure 3.

Figure 3 Details of Strengthening Scheme for Test Panel

The vertical Cintec anchors were each tensioned up to 57kN and, during jacking, the behaviour of the wall and anchors was monitored, see Figure 4. The elongation of the anchor between the jack and the anchored end in the retaining wall was as expected. However, the wall itself was also found to compress by about 4mm at the top which was significantly greater than the 0.2mm expected. The wall was also found to deflect by about 4mm towards the platform.

Figure 4 Post-tensioning Anchor in Test Panel

For the supplementary load test, the applied wind load was simulated by the application of a lateral point load on a hothontal spreader beam positioned vertically at the centroid of the wind pressure. The lateral load was applied with a hydraulic jack pushing against a jacking frame anchored to the retaining wall supporting the parapet, see Figure 5.

Figure 5 Testing Arrangement

The behaviour of the test panel under lateral loading was monitored using twelve 140mm (5½ inch) strain gauges located on both the tensile and compressive faces and six dial gauges on the tensile face. The test was undertaken with an incremental increase in applied lateral load up to 3.5kN/m, equivalent to 1.6 x nominal wind pressure. The maximum deflection at the top of the wall was 0.38mm while the maximum tensile strain at a position 500mm above the base of the wall was 48 micro strain.

The test results are plotted in Figure 6 against the predicted values which were based on a characteristic masonry strength fk of 2.3N/mm2 and an elastic modulus E of 2.O7kN/mm2 (=0.9fk). The measured results identified that the loading was not uniform across the full panel so, for comparison with predicted values, weighted averages were determined in proportion to the tested area. The measured results demonstrate linear behaviour but give greater strains and deflections than those predicted. This suggests that the masonry strength and/or stiffness assumed for the strengthening calculations were too high. Nevertheless, considering the possible variation of material parameters for old masonry walls, the measured results match the predicted very well.

Figure 6 Comparison of Test Results against Predicted

At the end of the test there was no evidence that the loading had caused any damage such as cracking or spalling.

CONCLUSIONS

The 2m long test panel of 100 year old parapet wall was post-tensioned against wind and dynamic pressure and suction loading, using two Cintec anchors. The supplementary load test with an applied load up to 3.6kN/m (1.6 x nominal loads) demonstrated a linear elastic response.
The predicted response of the strengthened wall, calculated beforehand and based on assumed values for the material properties, were within 30% of the measured values. Bearing in mind the wide range of uncertainties in relation to the wall stiffness and strength, this is surprisingly close. On completion of the test there was no damage such as cracking or spalling.
It is concluded that the supplementary load test was successful in demonstrating the efficacy of strengthening an old brick wall in a poor state of repair. The strengthening scheme presented is an economic and aesthetic solution to the refurbishment of understrength and unstable masonry walls and parapets.

ACKNOWLEDGEMENTS

The authors would like to thank Dr S Mehrkar-Asl for his assistance during the supplementary load tests and analysis of the results.

REFERENCES

ICE National Steering Committee for the Load Testing of Bridges ‘Supplementary Load Testing of Bridges’. Thomas Telford 1998.
Commercial-in-Confidence
Report No. B2056A/R01
May 1999