[ Home ]

Please note that the information contained was correct when it was published in July 1997. However, since that time further investigations and analysis by the consultants has resulted in minor amendments to some of the details following.

REPORT ON OPTIONS FOR THE STRENGTHENING AND MAINTENANCE OF HAMMERSMITH BRIDGE

1.0 EXECUTIVE SUMMARY

Hammersmith Bridge was closed to private vehicles in February due to a number of problems with the 110 year-old structure in its current condition. The London Borough of Hammersmith and Fulham have been working with consultant bridge engineers from Hyder Consulting Limited to address the problems and plan the best way forward for the bridge.

The late Victorian suspension bridge is required to carry modern levels of traffic loading. It has had one major programme of remedial works between 1973/1977 which was planned to give the structure at least a further 15 years of guaranteed life. Since then the bridge has had some repairs due to damage from overloading and a continuing pattern of maintenance work. Also design code loading requirements have increased in the period, reflecting the greater use and weights of commercial vehicles. It is now due for more strengthening and repairs which with regular maintenance will extend the life of one of London's finest bridges.

This report describes alternative options for the future of the bridge from replacement by a new crossing, to localised repairs on the old bridge in conjunction with building an independent footbridge alongside it, to extensive strengthening and repair works to enable the existing bridge to satisfy current requirements of loading from traffic and pedestrians.

The report focuses on the plan to strengthen the old bridge presenting options for construction measures and further detailed studies into each group of elements which needs attention. The riveted connections of cross girders to hangers need to be upgraded with new bolts replacing rivets. Stiffening trusses require plating to top and possibly bottom chords in localised regions. The Hammersmith tower top bearings must be replaced. Further studies are planned to examine closer the load effects on hangers and chain connections with possible improvements to hanger articulation. It is also recommended that corroded footway cantilever girders are refurbished along with external masonry on abutments and piers and internal masonry in anchor chambers. Finally the deck surfacing panels and some deck timbers must be replaced.

For works to the Hammersmith Bridge listed building consent needs to be gained for all construction measures which alter the character of the Grade II listed structure, and must be built into the programme.

Another important aspect of the design of strengthening is the maintenance implications that are associated with prolonging the bridge's life. A schedule of monthly, annual and six-yearly inspections is recommended for Hammersmith Bridge. At this stage an indication of the elements that will need maintenance in the future is given, and a full Bridge Maintenance Manual will form part of the strengthening design recommendations.

For the strengthening and repair option, there is an identified cost of up to £1.8 million for work which needs to be implemented in order to maintain the present load-carrying capacity of the bridge which restricts traffic to LT scheduled buses, emergency vehicles, motorcycles, bicycles and pedestrians. Further work, depending on the level of strengthening identified as a result of further study work, could add a further £2.2 million. These costs are based on information found in previous assessments. The extent of works necessary will be defined as design work progresses, and the programme updated. It is proposed that construction works are carried out in four phased contracts. At this stage it is estimated that the construction work will be completed in 1999. On completion of these works the bridge will be able to be re-opened to traffic not exceeding the 7.5 tonne weight limit.

2.0 INTRODUCTION

2.1 Historical Background

The first suspension bridge across the River Thames at Hammersmith built in 1827 was condemned in the 1870s due to, amongst other reasons, over-loading during the University Boat Race.

The existing Hammersmith Bridge was built on the same site in 1887 to Sir Joseph Bazalgette's design. A timber deck bolted to wrought iron cross girders is supported by longitudinal stiffening girders. The deck structure is suspended on wrought iron hangers from parallel double suspension chains. The chains are connected to saddles which slide on bearings at the top of wrought iron towers housed inside ornamental cast iron casing.

The bridge was refurbished in 1973 with replacement steel stiffening trusses, improvements to the articulation of some of the mid-span hangers and new tower-top roller bearings and deck expansion joints. New deck timbers in 16 foot panels replaced the existing deck structure and surfacing was changed from wooden blocks to coated plywood panels. The surfacing panels were subsequently replaced in 1987.

In 1984 the Barnes tower roller bearings failed under a heavy load and were replaced by elastomeric bearings to ensure longitudinal movement of the suspension chain saddles at the top of the tower.

2.2 Inspection, Assessment and Load Testing of the Bridge

Recent investigations on the Hammersmith Bridge revealed some areas of immediate concern. A Principal Inspection in September 1996 highlighted some areas where the 110 year-old structure was not functioning in the way it was designed to do, namely bearings on the Hammersmith tower, deck expansion joints and suspension chain and hanger connections. Structural elements of the bridge were also found to be corroded or worn, in particular cross girders, hangers and deck surfacing as well as some areas of masonry on piers and inside anchor chambers for the suspension chains.

In a Load Assessment the structure in its current condition was found to be unable to carry traffic with a 7.5 tonne weight restriction in combination with pedestrian loading as specified in the mandatory design manual Highways Agency Standard BD 21/93 (Technical Requirements for the Assessment and Strengthening Programme for Highway Structures). Elements which failed the 7.5 tonne assessment were the stiffening trusses, hangers, chain connections, cross girder cantilever connections, and bracing members in the towers.

A series of further load assessments for particular loading scenarios relevant to the current use of the bridge was then carried out by Hyder Consulting's Special Structures group. The results of these further assessments indicated that the bridge was capable of carrying only limited road vehicles and pedestrians with outstanding problems in two critical areas:

1. cross girder cantilever connections

2. bracing connections in the Hammersmith tower.

As there were no obvious signs of distress in the tower bracing connections a load test was requested by the London Borough of Hammersmith and Fulham to find exactly how the tower behaved under test loads.

The practical load test on the bridge revealed lower stresses in the tower elements than predicted, due to small relieving movements of the roller bearings on the Hammersmith tower.

In all other elements the practical load test measurements confirmed the findings of previous assessments. It was found necessary, therefore, to close the bridge in February 1997 to traffic except a single bus in each lane, emergency vehicles, bicycles, motorcycles and pedestrians, which corresponds to the maximum safe load that can be caused by the bridge in its current condition.

3.0 ALTERNATIVE PLANS FOR HAMMERSMITH BRIDGE

3.1 Options Examined

Since closure several alternative scenarios for the future of the bridge have been examined. The main options are described below highlighting the major implications of environmental impact, cost and programme.

One option to replace the old bridge with a new river crossing was explored by the Council. A new bridge to a modern design with a design life of 120 years capable of carrying unrestricted traffic and pedestrians is estimated to cost approximately £10-£20 million. There would be serious environmental problems associated with increasing traffic flows and removing weight restrictions in an already congested area. Construction of a new crossing would maximise disruption to both existing traffic and public transport routes and to local residents. In addition the loss of an attractive Grade II listed structure made this option unacceptable.

A proposal to provide alternative temporary structures using truss-type sections on five piers supported on piled foundations in the river was considered. However, as well as having similar environmental problems to a replacement bridge and the difficulties of realigning the approach roads, at a cost of £4-£5 million this temporary measure was found not to be cost effective and environmentally unacceptable.

Several options were studied with a view to limiting the use of the bridge to specific vehicle groups and loadings and hence limiting the necessary strengthening works. A possible "do-minimum" solution is to limit the use of the bridge to emergency vehicles, buses, motorcycles, bicycles and pedestrians as presently enforced on the bridge during its present closure. However, as stated in section 1.0 above, there is an identified cost of up to £1.8 million for work which needs to be implemented in order to maintain the present load capacity of the bridge.

Another possible scenario is to allow black cabs (taxis) to be added to the list of exempted vehicles. There are, however, over 15000 of these vehicles and, regardless of the manual or automatic methods considered, the practical problems of ensuring that the scheme was enforceable and that the permissible loads on the structure were not exceeded, were such that this proposal was rejected.

A further option of reducing the traffic weight limit to 3.0 tonnes with allowance for single midi buses was also rejected. Weigh-in-motion systems of traffic management to divert vehicles exceeding the limit away from the bridge route were thought to be problematic with long-term maintenance and operation implications. The existing structure would still require strengthening, but to a lesser extent than for the previous 7.5 tonne limit.

One-way tidal flows of traffic going north in the morning and south in the evening were also deemed inappropriate for the Hammersmith crossing.

3.2 Feasible Options

The Council's requirement to restore the existing structure to a condition where it is capable of carrying traffic with a 7.5 tonne limit in two lanes with single buses and emergency vehicles plus pedestrians would be satisfied by carrying out strengthening and remedial works on the bridge.

Constructional measures will be required for stiffening trusses, hangers, crossgirder connections, tower-top bearings and deck expansion joints, with further detailed design studies into chain connections, towers and deck elements. In addition repair measures would be needed to cross girders, deck surfacing and some timbers, external masonry on piers and abutments and internal masonry at anchor chambers.

The estimated cost of the works is between £2-£4 million.

The strengthening of the existing structure would mean small-scale changes to the details of the bridge whilst not altering its overall appearance. There would be no foreseeable change to traffic using the bridge once reopened. There would however be some disruption in the surrounding area due to closure while the works were being implemented. An outline of the work required for the strengthening option is given in Section 4.0.

After strengthening had been put in place the 110 year-old structure would need to be inspected at regular intervals to monitor conditions into the future in order to investigate and plan further maintenance and repairs as necessary. A guide to the continuing maintenance liabilities of the bridge is given in Section 6. 1.

Estimated time for completion of the works based upon investigations so far is during 1999 as shown on the programme in Section 8.0.

An alternative feasible option would be to build a separate footbridge alongside the existing bridge. This would relieve the old structure of significant design footway loading requirements, thus minimising the work needed to be carried out on the Hammersmith Bridge. Under 7.5 tonne traffic loading with allowance for buses and emergency vehicles it is likely that repairs would be limited to the Hammersmith tower bearings, deck replacement panels and some areas of masonry. The estimated cost of these items is approximately £1 million.

The Grade II listed structure would therefore be visually unchanged, as would the road traffic on the crossing. Maintenance requirements would be reduced by closing the footways which are currently corroded and in need of refurbishment, although some minor remedial work would be required to prevent further deterioration.

A new footbridge is estimated to cost £1-£3 million which would in the long term provide a saving over strengthening. The new structure would take approximately six months to construct, so depending on the time needed to gain planning permissions, this option could allow the Hammersmith bridge to be opened earlier than the strengthening option. Additional land-take would be needed for siting the new structure with access onto existing pavements. The visual effect upon the river in this area would greatly depend upon the design of the new footbridge. This option is described in section 5.0.

4.0 STRENGTHENING AND REPAIR OPTIONS

Load assessments of the existing bridge found the following structural elements to be inadequate for 7.5 tonne traffic loading in combination with buses, emergency vehicles and pedestrian loading. These were:

1. Cross girder riveted connections

2. Longitudinal stiffening truss

3. Suspension hangers

4. Suspension chain connections

5. Tower bracing

In addition the Principal Inspection of the bridge in 1996 revealed several areas where maintenance and repair work was necessary. These were:

1. Cross girder cantilever flanges

2. Hammersmith tower bearings

3. Piers and abutments external masonry

4. Abutment anchor chambers internal masonry

5. Deck surfacing panels

This section outlines options for the implementation of strengthening and repair measures for the above items, giving advantages and disadvantages for alternative methods with recommendations for preferred solutions.

It should be noted that as with any historic structure repair work will inevitably uncover unforeseen problems on site such as corrosion or cracks in steelwork or deterioration in timber. The possibility of such unforeseen problems needs to be considered in terms of possible additional costs and disruption of the programme.

The strengthening details are currently being discussed with English Heritage and the London Borough of Richmond-upon-Thames prior to the Council making applications for listed building consent.

4.1 Cross Girder Riveted Connections

Wrought iron cross girders supporting the road deck and cantilevered footways are connected to suspension hangers largely by original rivets. When the footway cantilevers are assumed to be fully-loaded with local crowd loading in the worst case the existing connections fall assessment and are in need of strengthening. Increased capacity can be achieved in these connections by replacing rivets with new bolts.

Recent inspection using the maintenance gantry suspended below the bridge deck showed the extent of bolting that has already been placed in cross girder connections on the upstream side during the 1973 modifications. Additional pairs of new bolts are required in each of the web connections on the upstream side to resist the current footway loading as shown in Figure 1.

Figure 1, click to view fullscreen in separate window, file size 44KB

The existing downstream connections have previously been modified. Additional strut and tie elements have been bolted on each side of girder webs to support a temporary footway in 1973. On the downstream side 4 new pairs of bolts are required in each cross girder web connection as well as 2 pairs of bolts in each cross girder/hand plate connection to suspension hangers.

At all these locations existing 3/4 inch rivets must be removed one at a time with the relevant footway closed, holes reamed out and M20 High Strength Friction Grip (HSFG) bolts placed in the arrangement shown in Figure 2.

Figure 2, click to view fullscreen in separate window, file size 49KB

Existing tie-rod assemblies will have to be removed to accommodate the new bolts in the downstream connections. It is recommended that tie rods are reinstated after bolts have been installed.

Dome-headed bolts which partially resemble rivets on exposed faces of the connections may be used instead of hexagonal-headed HSFG bolts if it is thought necessary on conservation grounds to maintain the appearance of the riveted connections. However these will not be an exact match of the existing rivets and a conventional nut and washer is still required on inside faces.

Since standard HSFG bolts have already been used extensively in 1973 modifications to these connections, the new bolting proposals are thought not to alter their existing character.

Approximately 1250 No. new bolts are needed to replace existing rivets.

4.1.1 Construction Problems

The use of dome-headed bolts may add access problems for proprietary powered wrenches used to tighten this type of bolt. Standard HSFG bolts would be recommended, as previously used. Gaining access to connections below the bridge deck may be problematic in construction.

Existing rivets may have to be drilled out. There have been associated problems when similar work was carried out on other Thames Bridges, such as misaligned holes in plates and corroded sections. Such problems will be solved on site as they arise.

4.2 Longitudinal Stiffening Truss

The longitudinal stiffening truss was found in assessment to have insufficient chord buckling resistance. Strengthening of truss chords is required in the midspan region of side spans and quarter-point regions of the main span. Further detailed analysis of the global bridge structure will provide information on the full extent and level of construction measures needed to strengthen the trusses.

Inspection of the condition of sliding bearings at the ends of the trusses will be carried out to determine the need for work there.

The preferred solution for strengthening trusses would be to add plating to the existing chords. Plate material could be added above or below existing chord flanges since new material does not need to be continuous to resist buckling between restraint points. The most unobtrusive method, Option A, would be to fix plates in two strips on the top of bottom chord flanges and beneath top chord flanges in discrete sections between hanger positions as sketched in Figure 3.

Figure 3, click to view fullscreen in separate window, file size 56KB

In Option B plating could be added above top flange and below the bottom flange plates between hangers after grinding flat existing flange seam welds, as sketched in Figure 4. This method however, would visually increase the depth of the trusses and is not preferred.

Figure 4, click to view fullscreen in separate window, file size 53KB

The trusses were replaced on the bridge in 1974 and the material, Grade 43 steel, is suitable for welding. In-situ welding is recommended for top chord plating which is visible from footway and road carriageway, where access to the trusses is good. Bolting plates to bottom chords if necessary, may be acceptable where trusses are below deck level over the river and access for welding is more difficult.

4.3 Suspension Hangers

Wrought iron hangers suspend the deck from the steel chains. A number of the short hangers near the middle of the main span were found to fail assessment where rotation of their pinned connections was impeded and the bending stresses induced in addition to axial tensile stress exceeded their capacity.

Global analysis of the suspension structure will be carried out for a range of loading conditions to find the extent and severity of bending stresses induced in the critical hangers. A number of hangers have previously been modified due to corrosion of hand plates and articulation problems as indicated in Figure 5. Strengthening measures will need to be considered for the different types of existing hanger.

Figure 5, click to view fullscreen in separate window, file size 105KB

Several have been worn near to the truss level connections and measurements of each hanger will be taken into account.

Extensive investigation into the behaviour and strength of the existing hangers is necessary at this stage.

The bending effects on hangers is dependent on the fixity of tower-top bearings and the deck expansion joints both of which are known to have restricted longitudinal movements. It is therefore essential to examine the combined effects of altering the behaviour of these elements and to plan a suitable construction sequence.

The stress history of the hangers is unknown and further investigation may lead to the need for fatigue testing of a hanger removed from the bridge with a temporary replacement installed.

Further analysis must be carried out into the effects of the failure of a single hanger to ensure that progressive collapse does not occur.

Possible construction measures range from a minimum action option whereby hangers are inspected at regular intervals and replaced as necessary if problems occur, to full replacement of the group of hangers at the centre of the main span.

The first option has ongoing maintenance implications, whilst the second would be expensive and may not be warranted for a structure of this age. In between these two extremes there may be scope for solving the problem of the fixed connections by adding articulation in the form of new bearings which would not lose their ability to rotate if properly maintained. Construction measures would require temporary support of the bridge deck at each defective hanger and fixing of new bearings to the old hangers.

Further investigation into the problems and the options available needs to be carried out before recommendations about the most appropriate and cost effective solution can be made.

4.4 Suspension Chain Connections

The original suspension chain links were found to have adequate tensile capacity to carry traffic and pedestrian loading. Inspection of the chains inside the anchor chambers will be carried out to examine their condition below road level. The existing suspension chain pinned connections, however, do not currently satisfy design code requirements for either shape or the onerous limits of bearing stress. Investigations into the basis of the relevant code clauses are recommended at this stage. Further analysis of the detailed arrangement of chain links and material testing may be needed in order to avoid difficult and expensive construction measures to improve the pinned connections.

New pinned connections would involve temporary chain links being put in place while the old connections were dismantled, then rebuilt with new components. Additional attachments to the existing chain links would be very unsightly and are to be avoided if at all possible.

If the loading on the bridge were reduced the chain connections could be found to satisfy design requirements by analysing alternative load paths whereby the chain links flex rather than the pins rotate, and no additional material would be needed.

4.5 Tower Bracing

Bracing elements in the Hammersmith tower were found to fail assessment under full 7.5 tonne loading as a result of tower-top saddle roller bearings being "seized".

Tower bracing elements could be strengthened by adding plating and replacing all the one inch rivets with 24mm diameter HSFG bolts at their connections. However, it is recommended that seized bearings at the top of the Hammersmith tower described in Section 4.7 are replaced in order to restore the functioning of the structure to its original design. In this way construction works would not be needed on longitudinal bracing elements.

Further analysis of the effects of wind on the bridge is required to examine the effectiveness of transverse bracing elements.

4.6 Cross Girder Cantilever Flange Refurbishment

4.6.1 Results of Gantry Inspection

The below deck survey of the cross girder cantilevers carried out in April 1997 revealed severe corrosion in localised areas of the top flanges of approximately 90 No. cantilevered girders. In those areas water and salts have corroded the wrought iron angles forming the top flanges reducing them to less than 50% of their original thickness. 85 No. of the corroded areas were between the exposed hanger hand plate and the 12 inch low pressure gas pipe supported on top of the girders. Another 44 No. corroded areas were found directly beneath the gas pipe supports which are mounted directly off the flanges creating a moisture trap. A further 23 No. severely corroded areas were found at the tip of the cantilevers around the connection of the parapet posts. Photographs in Figure 6 show examples of the corrosion.

Figure 6, click to view fullscreen in separate window, file size 298KB

On 9 No. girders the top flange was completely corroded through leaving rough-edged holes from which further corrosion may perpetuate the loss of section despite being painted.

In addition 5 No. of the wrought iron angles connecting girders to hanger hand plates were found to be cracked adjacent to one of the top rivets as shown in the photograph on Figure 7. And in a few top flange locations existing rivets appeared to be cracked or missing.

Figure 7, click to view fullscreen in separate window, file size 303KB

A further 55 No. other girders were observed to have less severe areas of corrosion which has reduced the section of the top flange by less than 50%. This appeared as a rough and dimpled top surface which it is assumed was grit blasted back to good metal prior to the application of recent protective paintwork. In other areas flange angles have been bent locally.

4.6.2 Proposals for Remedial Action

In assessment it was found that the section of the cross girder cantilever was adequate for supporting the applied loads with a condition factor of 0.0 applied to the top flange. However, as corrosion still appears to be progressing it is important to implement measures to prevent the spread of girder deterioration which could affect the webs. Several options have been explored using different materials and applications which have a range of maintenance and cost implications as described below.

Where girder web angles forming part of the connection to hand plates are cracked and new bolts are to be installed as described in Section 3.1 it is proposed that new angle sections are bolted over the cracked wrought iron members as shown in Detail C of Figure 8. Where rivets are cracked or missing it is proposed to replace them with HSFG bolts.

Figure 8, click to view fullscreen in separate window, file size 48KB

Option 1

One option to restore the cantilever girder flanges to their original condition would be to cut out the severely corroded sections of flange angles and replace them with new steel angles bolted to existing webs and spliced with bolted steel plates onto the remaining wrought iron flanges.

This extensive repair work would be the most expensive option, would have many associated construction problems and may be unwarranted since it would undoubtedly outlast other areas of the bridge. The gas main would have to be temporarily relocated during construction work. Access to girders would have to be gained by means other than the underslung gantry, and cutting out 110 year-old members may uncover unforeseeable problems with the existing girders.

Option 2

A second option involving adding extensive new steel plating to all the severely corroded areas of girder flanges could be adopted. Corroded surfaces would have to be filled and prepared to receive mild steel plates in approximately 130 No. separate areas on 90 No. cantilever girders. An estimated 6-8 tonnes of new steel would be fixed by bolting as shown in the sketches of Figure 8 in the affected zones. Plate bonding techniques were explored for fixing the new plates, however the life of the epoxy cements used in service require clamps to be left in place to prevent failure. There is therefore little advantage of this method over bolted connections

The advantages of this option would be a long design life with relatively low maintenance implications.

Disadvantages are the high cost for mainly cosmetic repair work which add little gain in structural strength. A large amount of drilling of bolt holes through corroded wrought iron members may create unforeseen problems. Access for drilling and plating is restricted, particularly by the gas main which would have to be independently supported to allow the placing of new materials beneath it. British Gas have stringent requirements for working around the pipes. Another disadvantage would be the visual effect of many additional plates and bolts on the underside of the bridge deck.

Option 3

A less extensive repair option would be to prepare and fill severely corroded areas and add steel plating to only those 9 No. girders where flanges have been completely corroded through. Plating measures would be similar to option 2 with plates bolted above and below filled existing flanges. For this option research has been carried out into suitable filler materials which have properties suitable for bonding to metal, which may be trowel-applied, hardening in time to give adequate compressive strength to receive bolts and which may be painted. Approximately 1 tonne of new steel would be added with an area of 78000cm² covered by filler material of varying depth.

Advantages of this option are that costs would be relatively low. Further corrosion due to ingress of water to roughened holes in existing flanges would be prevented by plating. The impact and weight of additional steelwork would be minimised, as would difficult construction measures when connecting to the old flanges.

Disadvantages are that at 9 No. locations plating would be needed beneath the gas main, where access is problematic. The extensive use of filler materials would have maintenance implications of inspection and reapplication where cracked or debonded from girders in the future.

Option 4

The final option involves no steel plating, but would be simply to apply suitable filler material to cleaned corroded areas of top flanges, shaped to restore the form of original girders and painted to colour match existing paintwork. Advantages are low cost, ease of application with no need to move or support the gas main as the filler could be applied from beneath. There would be no drilling for bolt holes through existing flanges. Repairs would be lightweight adding minimal dead load to the structure. And filler patches could be shaped to look like original girders.

Disadvantages are that the materials available have limited life and so would require inspection and replacement maintenance operations in future. A major disadvantage would be that there are some areas of completely corroded flanges where it is unlikely that filler material could effectively fill the holes to prevent the ingress of moisture. Corrosion would therefore not be halted in these spots.

Recommended Option

For the cross girder flange refurbishment we recommend implementation of Option 3 as the minimum feasible repair option for the corroded girder flanges.

In addition new angle sections where the original angles are cracked are required. We also recommend replacing all cracked or missing rivets with new HSFG bolts.

4.7 Hammersmith Tower Bearings

Original steel roller bearings in the Hammersmith tower were replaced by new rollers in 1973. Access for observation and maintenance of the bearings is extremely tight between the cast iron casing and the wrought iron tower structure. In 1995 we understand the bearings were cleaned and lubricated. However, during load testing on the bridge in January this year, the movements of the bearings were found to be severely restricted.

There may be a possible future option of on-going specialist lubrication and monitoring movements of these bearings if sufficient freedom is found after initial controlled cleaning and application of suitable lubricating agent. There is a risk in this case that even if initially freed, lubrication may not be a lasting solution for the bearings, keeping in mind this option also has ongoing monitoring and maintenance implications.

Another option would be to replace the roller bearings with new steel rollers providing a solution in keeping with the original design. However, in view of the facts that replacement roller bearings were added in 1973, those on the Barnes tower failed in 1984 and those on the Hammersmith tower are now not working as they were designed to do, there would be a risk involved with this option. The accurate installation of new rollers within the confines of the tower casing would be a difficult specialist operation.

It is recommended that the "seized" bearings are replaced with suitable elastomeric bearings similar to those placed on the Barnes tower in 1984 which were observed in the load test to be working satisfactorily. There is invariably a limited design life associated with moving bridge bearings and elastomeric pads as shown in Figure 9 would be most easily replaceable. Elastomeric bearings would also be significantly cheaper than steel roller bearings.

Figure 9, click to view fullscreen in separate window, file size 35KB

4.7.1 Construction Problems

The problems inherent in replacing the bearings are that a complicated jacking operation is needed to lift the chain saddles off the existing bearings. The steel rollers would then have to be removed with very restricted access and new bearings mounted in their place.

We anticipate that this would be in the order of several days. Some further total closures for shorter periods of time may be necessary for installation of scaffolding and the removal and replacement of sections of tower casing.

4.8 External Masonry on Piers and Abutments

It has been noted in the Principal Inspection of 1996 that external masonry in some areas of the exposed river piers and on abutments on either bank is in need of attention. A simple solution may be to re-point the masonry in these areas subject to closer inspection. Re-pointing would be a relatively low cost solution to a minor problem.

Access to piers can be gained from the river bed at low tide. Abutments are accessible from dry land. This work would be programmed to take place during the summer months to suit application of materials.

4.9 Internal Masonry at Abutment Anchor Chambers

It was also noted in the Principal Inspection that some water has entered the anchor chambers. It may be appropriate to prevent the ingress of water and hence deterioration of internal components by resin injection and pointing of the abutment masonry. Further inspection of internal masonry is needed to determine the extent of work necessary.

4.10 Deck Surface Replacement Panels

The running surface of the existing bridge deck is in need of replacement. An initial investigation will be carried out into alternative materials for surfacing the bridge. This will include research into the effectiveness of different panel types in service on other bridges and possible trials on Hammersmith Bridge. Inspection of the condition of sub-deck tongue and groove boarding will be made to assess the extent of the timbers that need replacing. The study will examine the effectiveness of alternative new materials both in reducing the running noise of the existing plywood panels and improving the life-time costs of the deck. Where necessary, improvements of the fixings to sub-deck timbers and cross girders shall be made at the same time as replacing the panels.

Kerbs on the road carriageway may also be moved at that time to restrict the lane width as appropriate for 7.5 tonne 2-lane loading.

It is recommended that the expansion joints are repaired at the same time as the Hammersmith tower bearings replacement to avoid further overstressing hangers due to relative movements between deck and suspension chains.

It would be most prudent to replace the deck surfacing panels after all the necessary strengthening works have been completed to avoid having to relay any areas where other work may interfere with the deck.

4.11 Global Structural Analysis

The design of the main structural elements of the bridge will be developed from global analysis of the complete structure. For a geometrically non-linear suspension structure the relative stiffness of the chain, hanger and deck members greatly influences the performance of the bridge. All critical load combinations will be applied to computer models of the bridge to give ranges of forces for designing truss strengthening, for finding the extent of hangers which need improved articulation, and for calculating design loading to be applied to the tower structure and tower-top bearings.

When changes are made to the stiffness of longitudinal trusses and dead weight of the structure by the addition of strengthening materials these will be incorporated into the computer model. Similarly when new articulation is designed for hanger or chain connections the altered non-linear behaviour of the structure will be checked by further computation, in a cyclical process of design.

The combination of releasing the Hammersmith tower saddle bearings as well as freeing the longitudinal movement of the deck expansion joints has a critical effect on the bending stresses induced in the hangers. The combined effect and sequence of construction measures must be carefully examined as a vital part of the design process.

5.0 FOOTBRIDGE OPTION

Current design load intensifies exceed those ever expected from normal use when the bridge was designed in 1887. Design footway load intensities comprise a large proportion of the total. Historically this is a recurring problem as the original 1850s structure was replaced due to worries about the bridge's capacity to carry large crowds of pedestrians during river events.

Provision of a separate footbridge is a feasible option for Hammersmith. This would be a completely independent structure sited alongside the existing bridge. Footways currently in use on the bridge would be closed and pavement approach routes directed onto the new footbridge.

The main advantage of providing an alternative pedestrian crossing is the significant relief in design loading to be applied to the existing bridge. The Highways Agency's latest code of practice for assessment of highway bridges, BD 21/97, specifies footway loading which amounts to approximately 400 tonnes of pedestrians on the bridge. This is the current requirement for footway loading and takes account of reduced intensity of loading and the span length is increased to that of Hammersmith Bridge. No further reduction in intensity of loading could be justified unless pedestrian usage were to be permanently controlled in such a way as to always guarantee a limit to the number of pedestrians on the bridge at any time. If however the footways were closed earlier assessments indicate that very little strengthening work would be necessary for the bridge to carry 7.5 tonne traffic alone, with allowance for buses and emergency vehicles. Seized bearings on the Hammersmith tower would require replacement, as would the road surfacing which is currently due for repair along with some minor masonry refurbishment on piers and abutments. The further items of strengthening described in Sections 4.1 to 4.6 would be avoided thus leaving the character of the listed structure unchanged. The life expectancy of this fine late Victorian structure could be extended both by relieving applied loading globally, and by taking away the burden from critical elements. Value for money could be gained in the long term.

The design life of a new footbridge would exceed that of the old bridge and so a pedestrian crossing would be guaranteed in the future if the time should come to replace the existing road bridge. In terms of life-time costs, the cost of providing footways on a new bridge in the next century would be saved if a footbridge were to be provided now.

Maintenance costs for future inspection and repairs to the worst-affected area of the bridge, the footway cantilevers, would be cut. By comparison, the additional costs of maintaining a new bridge would be small.

The environmental impact of building a new footbridge alongside the existing bridge would need to be considered in detail, however it is our view that a new lightweight structure may be designed in sympathy with the 110 year-old suspension bridge to enhance rather than detract from the character of the river in the area.

Additional land-take would be required for siting of the new footbridge with diversion of the existing footways onto the crossing.

Further investigation into the costs of a new footbridge would be needed to provide detailed comparison with strengthening options. An initial estimate of the cost would be in the range of £1-3 million depending on the form of structure chosen. It is our view that a footbridge could be designed and built within a budget of £1.5million. Time for construction would be approximately 6 months, design and contract preparation may take six months, however this does not include time for gaining planning permission and the required approvals and public consultation if this proposal is to be pursued.

6.0 ADDITIONAL ITEMS

6.1 Future Maintenance

It is difficult to estimate the expected life of the bridge once the strengthening and repair work has been carried out. In order to maximise the life of the listed structure it will be necessary to continue maintenance operations after the proposed strengthening works are completed. A programme of inspections will provide the records and information needed to implement both preventative and reactive maintenance measures to rectify any problems as they develop.

It is important to recognise that whatever form of strengthening is carried out, major maintenance and repair work will be necessary to prolong the life of the bridge. We note below the various routine and major maintenance activities. For the major maintenance activities we would include replacement of deck panels, deck timbers, expansion joints, tower saddle bearings and we estimate that these elements would need to be replaced approximately every 10-12 years, 25 years, 20-30 years and every 50 years respectively. It is also possible that some major work could be generated by unforeseeable increases in the loadings stated in new or revised Codes of Practice. Based on the above information and assuming that the bridge is repainted after 15 years, we would expect that the expected design life of the bridge without further major works would be approximately 25 to 30 years.

As with all Victorian structures corrosion and deterioration are increasing concerns which can be best managed by frequent inspection and an optimised programme of maintenance.s

6.1.1 Programme of Inspections

It is recommended that a schedule of regular inspections be followed in order to spot possible problems as they happen so that suitable maintenance operations may be implemented early. This will allow the Council to plan remedial action before major problems arise and budget for future repair work.

Inspections may be carried out at three levels.