Holdfast Training Services Trial Blog

Site Two Fifty One – Prop Installation

Despite site activities being focused on the remaining piling works (17 CFA piles to go), as always there is a push to do multiple other activities all at the same time, but luckily not quite in the same place.

Groundforce Shorco prop installation. The aim is simple – put in 9 props and remove about 1500m3 of water and about 10,000m3 of sand and gravels (and some clay).

Prior to now sections of capping beam have been poured, shear stubs were embedded and prop end plates were fastened.

Shear stubs installed

End plates being fastened to the capping beam

The general principle is to excavate to underside of capping beam, install a prop, excavate in order to dewater, excavate more, dewater more, install the next prop and repeat, 9 times (9 props), all while avoiding piles and segregating the exceptionally beach like sand and gravels (removal of this is less than half the cost of clay muck).

The first 2 props were installed this week and aside from problems with crane time there are a few other interesting points to consider.

Problem 1. End plates not meeting with shear stubs. Despite very careful installation of the 4 x M30 bolts embedded 250mm into the beam using resin, the end plates did not quite have full contact with the shear stubs. Options: weld a connection, use a grout pack, place shims or do nothing (assuming with the applied prop pre-stress the gap would close) to provide a load path. A good weld is difficult to achieve in a 30mm deep recess and the plates may well be damaged when the weld is ground off at the end of the works. Using a grout pack is possible but with movement of the prop (thermal expansion in the day) this might crack. Shims – the best and simplest option – this was chosen and on referral to the designer, the recommendation was to use S275 steel.

Shims used to provide path for load transfer

Problem 2. Not all of the capping beam is complete. The capping beam (which will form an entire ring around site) is used to transfer vertical load (from the buildings above) to the piles below. It is also being used to prop against horizontally.

Diagram shows line of completed capping beam (in red) with future props to be installed (dashed green arrows) in areas of capping beam yet to be cast.

The question is, is lateral load transferred along it (as if it were a steel beam) to the ground at the corners or to the secant piles below acting in a sort of individual cantilever/domino fashion. The secant piles have 900mm of reinforcement cage embedded into the beam and are at 1.2 centres with male C30/40 piles and alternate C10/15 female piles. On further consideration, the female piles are only there to provide water resistance, so the load is transferred into the top of each pile down into the ground (embedment anything up to 23m for the permanent condition).

This can be modelled as follows:

Plan view of capping beam showing prop loads applied to beam

By splitting the capping beam into sections around the props (10m spacing between props in this case) each pile within the section can be considered to carry part of the prop load. It is proposed the nearest two piles will see more load than the next two and so on. If each prop and pile section behave in this way it implies that as long as the capping beam is completed to at least midway between the next prop the system should work. That’s the theory anyway!

Plan view showing prop load transferred horizontally and vertically into the capping beam and onto the male pile reinforcement

Props 1 and 2 in place

Excavation between bearing piles

Problem 3. Remaining props are likely to be installed prior to the capping beam reaching 50N/mm2. Similar situation to above, capping beam is designed for permanent loading, temporary design is for 30N/mm2. Looking at the cube results, this strength is reached between 15-25 days. Therefore, in theory, this is when the props can be loaded. To be more sure additional cubes will be cast and crushed to understand the rate of strength gain between 7 and 28 days a little better. Additionally the plan is to use thermocouples within the beam to understand the actual temperature of the concrete. This could then be replicated in the curing tank and rather than adhering to the 20 degrees standard, which should produce more realistic cube strengths similar to actual conditions.

Contract Update: Work on site is currently progressing at risk (no formal contract) – the monthly instructions to continue work ran until 20 Jul 15 and the next one (as far as 20 Aug 15) has not yet been signed off. The client has hand shaken on the main contract with Ray O’Rourke at £119M so clearly all will be fine!