Causeway Bridges - 2012 Booklet
Situation prior to the construction of the present bridges
The navigation channel was bridged by a humped structure, situated at the eastern end of the causeway where the present road roundabout is situated. For many years after construction was finished the presence of piles cut off during the demolition of this bridge was revealed by lumps in the road surface as the fill placed in the channel settled.
The present channels replaced two shallow channels through mudflats roughly in the same location. These were deepened with improved alignment to provide for navigation and to accommodate floods. The spoil was used to construct Heirisson Island.
The 1864 bridges were of one lane, probably 12 feet wide. The original deck timbers of those bridges were still there in the widened timber structures at the time of construction of the new bridges. They and the widening were covered with a bituminous concrete of considerable depth in which the rails for the tram were situated on the newer widening on the eastern side. Deck planks were frequently broken.
As a point of interest, treated karri was used for both pier half caps and decking for a number of bridges including Garratt Road Bridge and the Canning River Bridge on Albany Highway. As noted by Lloyd Margetts, in his history of the Canning Bridges, the timbers were eaten from the inside out by the termites entering through bolt holes. They gained access to the piers through the wandoo stringers, most of which are hollow. So the first time anyone was aware of the termite attack was when a half cap was crushed or broken.
The Causeway Bridges
The eastern bridge is 725 feet (221 m) long and the western one 376 feet (115 m) long. The longer bridge consists of eleven 61 feet (18.6 m) spans and a relieving span of 27 feet (8.2 m) at each end. The shorter bridge at the western end has five 62 feet (18.9 m) spans and relieving spans of 33 feet (10m) at each end. Provision was made for a vehicular way 27 feet (8.2 m) wide (three lanes in each direction) and pedestrian footways 8 feet (2.4 m) wide with accommodation for all services such as water and gas mains under the deck.
Construction of Piers for the Existing Bridges
These were built inside coffer dams consisting of driven timber piles supporting guides through which 75 mm thick karri birdsmouth sheeting planks was driven. The planks were 225mm wide and were 900 pointed on one edge and a 900 slotted on the other. They were pointed with a splayed cut at the bottom to wedge the plank being driven hard against the preceding one by the pressure of the soil being displaced. Walings were needed at frequent intervals during dewatering and excavation of the mud from within the dams as the pressure of the water and mud outside was likely to bend the sheeting below the bottom waling inwards extruding the mud inside upwards.
The pile groups under each line of deck girders were then driven, pile caps cast, a trapezoidal extension cast on top narrowing down to the size of the columns a little below water level. The columns are contained within curtain walls and transverse top and bottom beams. The piles in my recollection were karri, not jarrah. They would be immune to rotting and borer attack being encased in fully saturated mud. The abutments for the relieving spans at each end of each bridge were founded on reinforced concrete piles rather than timber piles as a considerable length of pile was above permanent ground water level and would be subject to rot attack. The relieving span piles were encased in embankment fill which was over a considerable depth of highly compressible soft clay. This created a problem at the western relieving span abutment of the western bridge for two contributing reasons. Firstly negative friction on the piles would have loaded the piles to such an extent that they were driven further into the founding stiff clay below and, secondly, the embankment load would have led to consolidation of the foundation clays below the toe level of the concrete piles exacerbated by a greater depth of relatively soft foundation clay below as the piles reached their design penetration per hammer blow well above the level of the timber piles of the adjacent bridge abutment piers. No doubt the bridge abutment piers would have settled a little due to the effect of the nearby embankment, but it would have been millimetres compared with centimetres of the relieving span abutment.
A further problem may have been created by the use of rapid hardening cement for the relieving span piles of one of the relieving span bridges due to lack of normal cement at the time. Rapid hardening cement is low in alkali, which normally passivates the steel, so delaying corrosion after penetration of water. As the river water is salty for much of the year the problem is exacerbated. To my knowledge no problem due to corrosion of the steel in these piles has yet become apparent.
Construction Plant
Timber pile driving frames, with steel channel faced pile driving hammer guides, were used. These were often built for each job, the pile driving gangs becoming so skilful they could knock one up very rapidly without any plans. The hammers consisted of a trapezoidal shaped lump of cast iron with a tongue down the back to fit between the guides and generally weighed 2 tonnes for timber plies and up to 6 tonnes for concrete piles. Double drum winches were located in a convenient place to lift and drop the hammer. The coffer dams provided in the case of the Causeway bridges were used to support the gear for driving the foundation piles inside the coffer dams, the coffer dam support piles having been driven from a barge mounted pile driving rig.
Mobile cranes, except for small capacity fixed (non traversing) truck mounted cranes, were unavailable at the time, though we had the services of a steam driven floating crane from the Harbours and Rivers Department. This was used to place the bridge girders but didn’t have the reach to place concrete on the deck. There was no such thing as ready mixed concrete, so the mixing plant was situated on an approach embankment to each bridge so the concrete could be delivered on the level or downhill in hand pushed bottom dump skips mounted on narrow gauge rail trolleys.
Concrete Mix Design
Main Roads Chief Bridge Engineer Ernie Godfrey knew how to make good concrete. His bible was the US Bureau of Reclamation Handbook. The Bureau was very clued up on concrete durability. The aggregate cement ratio of the concrete used in the Causeway Bridges was 4.5 and the water cement ratio if I recall correctly was 0.4.
As cement was in short supply, supplies were double ordered from all over the world to try and ensure timely deliveries. We had cement from England, Sweden, Japan, South Africa and the Eastern States as well as local Swan Cement. The cement from UK, Sweden and Japan had a high alkali content and we found later that it reacted expansively with the aggregate available from the local quarries. This aggregate was epidiorite quarried from seams within the granite of the Darling Scarp. Most of the aggregate production at the time was produced for road surfacing and granite or grano diorite was not then used as they are acid rocks to which bitumen won’t stick without an additive which I guess was not then available or it was cheaper to use the basic igneous rock. The trouble with the epidiorite was that it was extremely fine grained due to it having cooled rapidly as it was intruded in thin seams into the pre existing granite. It appeared that the large surface area of these grains allowed the cement to attack it. This was a bit of a surprise as crystalline rock was supposed to be immune to cement aggregate reaction. We had done no testing as a consequence, which we always did if the presence of amorphous silica was suspected.
There was no record of any trouble with the local cement which had a lower alkali content. In any case the pier surfaces below high water level had been painted with a thick bitumastic coating to retard water penetration. The underside of the bridge deck was also painted with this coating to prevent penetration of water condensation.
The Composite Deck
Steel beam/concrete deck composite construction was first used in Tasmania where Alan Knight introduced it in the 1930s. Knight was the Chief Engineer of the Public Works Department of Tasmania. I believe Knight was the first in the world to use the technique. It was subsequently widely used in the USA and also notably in Germany on the reconstruction of many large bridges destroyed during the second World War. In the USA the technique was further developed by the use of steel studs in place of the reinforcing hooks initially used to bind the bridge deck to the steel beams.
As far as I am aware, the Causeway bridges were different from any of the Eastern States composite bridges in that the concrete slab was prestressed by securing the ends of the simply supported steel girders and jacking up in the centre (at the third points), casting the concrete and when cured lowering the jacks, transferring longitudinal compression into the concrete. Jacking trusses spanning between piers beneath the girders supported the jacks. This overcame the tendency which had occurred in many of the Eastern States bridges for shrinkage cracking to develop into more serious cracking under the effect of heavy traffic.
Such prestressing was widely used in Germany to control cracking which would otherwise allow the intrusion of salt laden water from de icing salt with serious corrosion consequences, particularly in the negative moment regions of continuous bridge decks. Most of the bridges there were of continuous construction to avoid deck joints which often present maintenance problems. This prestress involved jacking up the whole length of the bridge girders in huge arcs before concreting. In long bridges jacking heights were as much as 4 m or more.
We in WA developed a span by span technique for continuous bridges, by jacking and casting from ¾ point to ¾ point of the next span which only involved raising and lowering the girders at the piers on small hydraulic jacks by 75 mm in 18 m, regardless of the length of the bridge. The prestress varies along each span but is heaviest where it is most wanted in the negative moment region. A number of bridges were built using this technique in the Pilbara and it was very economical. Heavy shrinkage reinforcing though, can be used to control cracking in bridge decks. Normal shrinkage reinforcing is inadequate in this situation as shown in one of the bridges in the Pilbara and in several bridges in Victoria that were not prestressed.
Construction Time
The long construction time of the existing bridges (1947-1952) was dictated to a large extent by the shortage of the materials required. The bridges were constructed by the Main Roads bridge construction crews with about 50 men being employed.
Official Opening
The bridges were officially opened on 19 September 1952 by the then Premier of Western Australia, The Hon Sir Ross McLarty, MLA, assisted by his Minister for Works, The Hon David Brand, MLA. Mr J D Leach was the commissioner of Main Roads and Mr E. W. C. Godfrey the Design and Construction Engineer. At a time of financial austerity prevailing when the bridges were built it was a tribute to the skill of Main Roads bridge engineer Ernie Godfrey that innovative bridges could be constructed economically in the state. It was the first bridge in the Western Australia to utilise composite steel/concrete construction, shortly after its pioneering use in Tasmania.
Author: Gilbert Marsh supplemented with material provided by Lloyd Margetts
