A visitor to Watney Market in East London might comment on the odd steel framework on the top of Winterton House, a 26 storey height brick clad building but would be unlikely to know the reason for it.

Winterton House was built in the 1960s as a steel frame building clad in glass reinforced plastic panels. The steelwork design led to awards for its efficiency but sadly towards the end of the century the cladding was failing. The building was one of three, two of which were demolished but the plan was to rehabilitate this one to modern standards. This decision posed a problem for structural engineer Bryn Bird of Whitby Bird as the steel frame could not support the extra load of concrete to be cast for the floors to improve the acoustic isolation between apartments. CERAM was first involved when the Information Centre was asked to do a literature search into the interaction between the steel frame and masonry infill in the Empire State Building.

As Bryn's plans developed it became clear that some structural testing was required and CERAM was commissioned to do it. The concept was to build a freestanding "sleeve" of brickwork about the frame for its complete 26 storey height. A steel frame would then be erected above the roof and sitting on the new brickwork. The frame was to be jacked vertically upwards off the brickwork and steel ties from the frame were attached to the tops of the original steel columns. In this way all of the extra load on the frame was transferred via the ties and frame into the brickwork.

Although the approach is novel there remained a number of unanswered questions to complete the design.

Clearly the 26 storeys of brickwork will expand and contract with changes in temperature, also with moisture effects and also with time dependent load effects (creep). Long term shrinkage would have the effect of reducing the force in the ties to the columns and hence would overload the existing frame. All of the movements referred to are inter-related and so simply taking the sum of all the movements was not going to be accurate enough. As a result CERAM built a storey height structure outside and it was subjected to a fixed prestress for a period of months. Separate tests were made to establish the likely extent of the individual movements and the outdoor structure was used to estimate how the effects might combine in reality.

Twenty-six storeys of brickwork represents a significant cost and brick selection was an issue. The further one could come down the building before thickening out the walls the greater the savings in cost. CERAM therefore carried out compressive strength tests on walls of a number of brick types so that an accurate estimate of the strength was available rather than use the inevitably conservative Code of Practice approach.

Rain penetration tests proved it would not be necessary to render (parge) the inside of the brickwork to resist rain penetration.

Special corner connections were designed so that any wind load that was transferred into the frame by horizontal ties would be efficiently transferred into the buttressing walls parallel to the wind.

Overall a comprehensive programme of brickwork testing was carried out to fully characterise the brickwork and support the design decisions. The savings made in the cost of the brickwork more than paid for the test work.

Winterton House is a good example of how "design by test" can work and of how CERAM can devise experimental programmes to support real design situations.