Website Banner. John Monash: Engineering enterprise prior to World War 1.

[Introduction.]   [Main JM Index.]   [Bridges Index.]   [People Index.]
[Abbreviations.] [Units & Currency.] [Glossary.]

Stawell Street Bridge, Ballarat.

Summary.

Situated in the Town of Ballarat (formerly 'Ballarat East'), this was Monash's first attempt at a T-girder bridge. He prepared a rough design in April 1903 and wrote up final calculations on 1 January 1904. The reinforced concrete deck sat on brick abutments and had a clear span of 20 feet (6.1m). It differed from most of Monash's T-girder bridges in having only two main girders, one under each footpath. These were joined by two end-blocks and four intermediate cross-girders, dividing the deck slab into five panels. Design and construction took place under difficult circumstances. Monash was in financial difficulties, desperate to find new work, and preoccupied with problems at the Barham-Koondrook bridge site. Of the two experienced staff he had selected for Stawell Street, the foreman declined the job just prior to commencement and the ganger failed to appear on the first day of construction. Pressured by the Council for early completion, Monash could find only a new and inexperienced assistant to take charge, and a labourer hastily promoted to act as foreman, for this critical new venture.

View through bridge, along bed of creek. Two massive brick abutments to left and right. A short, single concrete span. Underneath can be seen the two main longitudinal girders and four cross beams.

University of Melbourne Archives
Reinforced Concrete & Monier Pipe Co. Collection, BWP/24407.

University of Melbourne Archives UMA holds two other images of Stawell St Bridge with Location Numbers BWP/24406 and 24408. Another image, with Location No. GPNB/1251, is not labelled, but is almost certainly of this bridge, and gives a more complete view.

When the wedges between the deck and the supporting falsework were removed in March, the deck proved incapable of carrying its own weight. Diagonal cracks appeared in the stems of the main T-girders near the supports. This form of failure is commonly referred to as 'shear' failure. The wedges were replaced, and Monash designed additional, external, reinforcement to effectively truss the bridge. The awkward task of fitting and plastering it over was completed in April. In July the strengthened bridge was tested by the Council's 16-ton roller. The girders withstood the load, but one of the deck panels gave way under the main roller wheel. After further repair, the deck survived until 1910, when Monash decided to replace it at his firm's expense.

It is possible to reinforce a girder to prevent 'shear' failure; but it is not clear from evidence in the files whether this was done at Stawell Street. In any case, it appears that Monash over-estimated the shear strength of the girders, or was not fully appraised of the danger. After this experience, he went on to investigate the phenomenon in the latest texts and journals and in written discussion with Gummow and Baltzer in Sydney. His letter to von Emperger, editor of the serial Beton und Eisen, was published with a reply in the journal in 1905. Monash also conducted his own research tests at the University of Melbourne. He then went on to design and build many T-girder bridges with multiple closely-spaced main girders. These were satisfactory for their time, and many are still standing, though accepted theory of reinforced concrete remained over-optimistic about shear strength for several decades and many bridges of this era were later strengthened.

The bridge is described in the RCPMC file as carrying Stawell St over "Little Bendigo Creek". I have not yet tracked this down, but think it might be a creek shown on old maps (example from SLV) crossing what is now Stawell St South, and now canalised.

[Top.]

Background to the Stawell Street venture.

In 1900 Monash and his partner Anderson won a contract to build eight Monier arch bridges in Bendigo [story]. Later that year they set about gaining work in Ballarat. At the time, Ballarat consisted of two municipalities divided by the Yarrowee Creek which had formerly been Ballarat West and Ballarat East. These had become the City of Ballarat and the Town of Ballarat respectively, but the old name of East Ballarat continued in use. The two councils were contemplating a major scheme worth some £25,000 to extend the channelisation of the Creek. The project included replacement of the bridge carrying Grant Street over the channel. The engineer for the City of Ballarat was Arthur Farrer, and the Engineer for the Town was William J Robertson.

Monash & Anderson were keen to gain work in the two municipalities, but Farrer was very suspicious of the new form of construction, following collapses at Scott's Creek and King's Bridge Bendigo, and problems with spandrel walls at Wheeler's Bridge. Robertson was lobbied by M&A and by B H Gummow of Ballarat (brother of F M Gummow). In November 1900, B H Gummow announced "Robertson is a convert to Monier". Anderson noted: "Old Robertson at Ballarat East is Farrer's little enemy", and was therefore even more likely to champion M&A's cause. It appears that Farrer overshadowed Robertson and carried more weight with the Ballarat East Town Councillors than he did. M&A failed to persuade the two Councils to adopt Monier lining for the Yarrowee Channel, or the City to adopt a Monier arch for Grant Street, despite Robertson's efforts. This may have had much to do with Town's decision in 1904 to give Monash his first chance to design and construct a reinforced concrete T-girder bridge.

From arches to T-girders.

From late 1899, when Monash was involved in the design and construction of the Bendigo Monier arch bridges, he became increasingly aware of their disadvantages. In February 1900, he asked W J Baltzer, of the Sydney firm Gummow Forrest & Co, to send details of the current theory on strength of reinforced concrete beams and slabs. In January 1901, considering a bridge for a factory in Bendigo, he proposed a "flat plate type" with "brick and concrete abutments of usual style". This would provide adequate waterway without the need to ramp the road up over an arch. He offered slabs (or 'flat plates') for several more projects in 1901 and 1902.

In January 1903, JM asked Baltzer to come to Melbourne to help solve problems in the manufacture of Monier pipes. He took the opportunity to ask if Baltzer could introduce him to the latest French and German texts on theory and practice of reinforced concrete. Although a note in his diary reveals he was "worried by an appalling mass of military work" he ordered two French texts (Christophe and Berger & Guillerme) and von Emperger's Beton und Eisen with back copies. In March, Gummow sent him a copy of the 'Wayss Handbook' containing an account of German experience and theory developed under the Monier brand.

In April 1903 Monash's attention was drawn to an article in one of the Ballarat evening papers listing expenditure approved by the Town Council following the raising of a loan. It included "Stawell St Monier Bridge, Little Bendigo, £250". JM must have been surprised and pleased, writing to Robertson: "I do not know whether to regard this as a misprint or whether you propose to favor us with a trial of the system". Robertson replied: "As far as I know it is the intention of the Town Council to erect a "Monier" bridge over the Yarrowee Creek at the points named … there are several places where the Monier system could be applied". On 21 April, JM travelled up to Ballarat with J B A Reed, Resident Engineer of the Koondrook project. As well as visiting Robertson, he called on the City Engineer, Arthur Farrer in connection with the Grant St project.

Back in Melbourne JM wrote to Baltzer that he was making a careful study of the overseas textbooks in his leisure time. "I am particularly interested in the flat bridges, and the T form construction." He asked Baltzer to rough out a quick design for a T-girder bridge of about 24 foot span, as a check on his own attempts. JM's diary entry for Sunday 26 April reads: "All day home, studying Monier textbooks". On 28th he offered fully costed alternatives to bridge a creek in the Shire of Benalla, and the next day sent firm costed proposals for four bridges and a stormwater channel to Robertson. These included the Stawell St Bridge, which he proposed at that stage to carry on three parallel main girders.

In May 1903, Gummow Forrest & Co sent sample calculations by Baltzer for a typical T-girder bridge, as requested by JM. They mentioned that they had just constructed a "test object". JM thanked them and replied that he also had "inaugurated a similar test, but on a much smaller scale, of a double T plate 10 ft span, designed in accordance with the principles given in Wayss's recent handbook". He stated that he had good prospects of a T-girder bridge on existing abutments in Ballarat. He told Gummow: "Your brother is rendering me considerable assistance in canvassing the Councils". JM went on to prepare an alternative T-girder proposal for Grant St and another in June for Ford's Creek, Mansfield. In August he branched into the field of building construction with a proposal for T-girder floors supported on reinforced concrete columns. (Malthouse project for architect de Lacy Evans.) By December 1903, when urging W J Muntz, Shire Engineer at Alexandra, to adopt reinforced concrete for a significant bridge project, JM told him that M&A "now use girder bridges".

In August 1903, Monash bid unsuccessfully to supply T-girder bridges at Peel and Rowe Streets in East Ballarat and concluded that the Council did not "seem disposed to give us a start with Monier Bridge work". However, it stuck to its decision regarding Stawell Street, and early in December, Town Clerk A Fraser wrote asking for drawings and specifications. Monash posted these on 7th, including a quote for £230. He was obviously very keen to win the contract, and gave Robertson a free hand to make any changes he thought necessary to ensure the documents were entirely to his satisfaction. On 14th the Council wrote accepting the tender.

Design and Construction of the Stawell Street Bridge.

Initial preparations.

Monash then proceeded with design of the falsework and formwork. A drawing dated 15 December shows he planned to use large members from the old bridge for the main props. The Mayor was keen to see work start immediately, and JM had to ask him to wait until after Christmas.

The next problem was to find a foreman. JM's first choice was J G Anderson, brother of his partner J T N Anderson. JGA had had extensive experience of Monier arch bridge construction and had acted as foreman for the geometrically awkward 3-arch bridge on the Upper Coliban and for the arch at Ford's Creek, Mansfield. Late in December 1903, he agreed to take on Stawell Street and was asked to report there after Christmas.

Monash paid his second visit to the site on 31 December. An M&A drawing dated the same day shows that the timber bridge was situated on a bow in the river, with abutments about 40 feet apart. Robertson's intention was to straighten the creek and narrow the waterway to 20 feet.

Plan showing the new abutments, situated closer together than the old ones, narrowing the creek and concentrating it, but still within the old water course.

Plan of location.
Based on a drawing in the Reinforced Concrete & Monier Pipe Co Collection,
University of Melbourne Archives

Monash wrote up his "Final Stress Calculations" for the bridge deck on 1 January 1904. They allow for the load of a 16-ton roller travelling on the centreline of the roadway. An undated engineering drawing, which is presumably the one sent to Council with the quote, shows a bridge with only two main (longitudinal) beams of 20 feet clear span, and four intermediate cross-beams. In accordance with JM's and Gummow's concern to protect their intellectual property, the drawing indicates the general arrangement of reinforcement, including main bars and some stirrups, but omits details of size and spacing. A clause in the accompanying specification reads: "The contractors will be entirely responsible for the sufficiency and proper arrangement of the steel reinforcing rods, and that they will have sufficient cross-section area to take up the whole of the tensile stresses in every part of the structure".

JM asked David Mitchell & Co to reserve 50 casks of cement for the project.

Staffing Problems.

On 8 January, Monash learned that J G Anderson was no longer available for the position of foreman and wrote to offer it to Charley Savage, who had been JGA's ganger at Scott's Creek. As JM knew that Savage "did not care to take sole responsibility", he promised to look for an Engineering Assistant to take official charge.

He first offered this post to H L Wilkinson in Sale, who had worked as a student in M&A's office some 15 months previously, just before completing his degree. JM explained that the job would be of short duration, but would provide great experience. He continued: "Of course you are, no doubt, thoroughly seized of the fact that the young engineer beginning his practice has to concern himself more with the drudgery of book-keeping and simple mensuration and correspondence rather than with exalted scientific work". Wilkinson would be "in a position of responsible charge with a thoroughly reliable and practical working foreman to assist and advise on practical points". As M&A's system was "to concentrate the control of our contracts in the Melbourne office", the Assistant Engineer's functions would be "almost wholly those of keeping in active touch with the Melbourne office and exercising local supervision under daily instructions from Melbourne". M&A would normally use a non-professional man for these tasks, but the job afforded a good opportunity to start. The wage would be £2-10-0 per week. [This was half the wage of a good foreman.] If Wilkinson took the job, JM would spend two or three days inducting him into his duties and assembling tools and plant, and then take him to Ballarat to receive final instructions on site.

Wilkinson must have declined, and on 11th Monash wrote to his former assistant J S Gregory in Lorne, making the same offer and promising work for four weeks. Gregory would have been glad to accept, but happened to be in Melbourne at the time. By the time the letter reached him, JM had offered the job to H F Tisdall who had accepted it. Tisdall commenced work with M&A on 19th.

JM then wrote to tell Savage that arrangements for Stawell St were complete. "We will leave in responsible charge an Assistant Engineer, who, however, has not very much practical experience." Savage was to meet JM at the bridge on the morning of Monday 25 January 1904. "Get out at Ballarat East Railway Station and proceed easterly along Humffray St till you come to Taylor's Brickyards. This is about 1¼ miles from the Station."

It was only then that a start was made on dismantling the timber bridge. The arrival of JM and Tisdall was of sufficient interest to warrant an item in the Ballarat Star the next day. Excavation for the new northern abutment was soon underway. Carters began to deliver materials, commencing with sand, stone and cement for mass concrete foundations, and bricks for the abutments.

However, there was no sign of Savage. M&A's clerk, John McNaught, sent him a telegram to which there was no reply. It was established that Savage had left Geelong on 20th headed for M&A's Melbourne office, and JM told Tisdall: "There is every reason to fear that he has come to some harm, as he is a perfectly sober man". To replace him, a man called O'Brien was to come up - a "steady labouring chap, but as far as we know, not fit for any responsibility". His pay would be 7/- per day. On 29th, JM advised Tisdall: "It now seems pretty certain that something unusual has happened to Savage, and we have given up hope of his joining you on this bridge … We must now leave you to do the best you can under our instructions."

Thus at a time when Monash was battling problems of flooding at the site of the Barham-Koondrook Bridge project, and shortly to experience a major crisis there, he was obliged to inaugurate his first essay in an unfamiliar form of construction with an inexperienced Assistant Engineer and a foreman freshly elevated from the rank of labourer.

Construction.

By 30 January the old bridge was dismantled. Supplies of bricks, sand and gravel were delivered and work commenced on the foundations. Tisdall reported that Robertson had "suggested" alterations to provide extra depth and extra protective wing walls for the abutments. JM replied: "We are not there to adopt suggestions, but want very clear orders for extras." He was annoyed that they had struck water in the foundations. Had it really been necessary to go that deep? Pumping was a very costly solution. Had Tisdall made it clear that this additional work would be charged as extras?

By 3 February, the ground had been cleared of old timber, the creek turned to its new course, and most of the earth shifted. Two thousand bricks had been laid and nine cubic yards of concrete poured. However, Monash suffered more annoyance when local funds ran out due to what he saw as "lack of foresight". His detailed advice to Tisdall had so far covered selection and testing of sand for concrete; how to use heat to bend the ends of reinforcing bars; details of postal communications; how to avoid having men on stand-by; and how to order and measure materials needed for extra work. Tisdall should include more detail in his reports - how many casks of cement were used and how many men engaged each day. Also, he should get new carbon paper. "Your letters are barely legible - your sketches are not."

Old foundation timbers had been discovered, interfering with new work. It would be possible to avoid these by altering the position of the new wing walls. JM told Tisdall: "If your local investigations support the desirability of this from our point of view, you have a free hand to try and cajole Robertson into requesting the change. But it must be done, as far as possible, as a suggestion wholly from him … While all this work is extra it does not follow that we will be properly recompensed for all the labor that we put into it. It is always so in extras, and the fight to recover bare expenditure incidental to a disturbance of programme always leads to rows. Therefore, although this work is extra we are still anxious to keep the cost as low as possible, so that a trade rate for the brickwork and concrete will cover our costs."

On 7th Monash wrote that he was obliged to go to Koondrook. The reports would be forwarded to him via Melbourne. "As I must be present for Monier work [at Stawell Street], & cannot be present during coming week, & assuming all concreting of foundations is finished, you had better pay off all hands except the bricklaying staff, O'Brien and one or at most two handy labourers." Tisdall should continue quietly with constructing the 'centering' and placing the reinforcing grids. "Am sorry Baker [bricklayer] is not doing too well; am alarmed about the 'mistakes' you merely hint at."

The word 'centering' is properly applied to the timberwork necessary to support a stone or concrete arch while it is being constructed. Presumably out of habit, M&A applied this term for the falsework necessary for girder bridge construction.

By 13 February, all bricks were in place (at least 12,000 of them) and attention turned to the deck. The first task was to build the falsework and formwork. Tisdall found this task difficult and made slow progress.

JM planned to arrive on the day of concreting in time for a 1 p.m. start. This would allow 5 or 6 hours of daylight. Two cubic yards could be cast per hour with a team comprising:
4 men gauging and turning [measuring the ingredients and hand-mixing],
1 boy watering,
1 man wheeling [the barrow],
2 men ramming &c,
1 or 2 men spare for emergencies.

On 21st Monash told Tisdall he was overwhelmed with work and would not be able to supervise concreting on 22nd as planned. "A disaster threatens at Koondrook by our having struck a blow hole right in the centre of our large coffer." He might be needed there, and if so - "regret it as I may and you may, your job will have to stop, the hands being paid off pending developments". However, if Tisdall felt confident, he could make tentative arrangements as detailed above.

Monash eventually reached Stawell Street on 25th for the casting of the deck. The Ballarat Star reported the event with an article commencing "The biggest day".

Disappointment.

Having overseen this task, JM attempted to supervise the remaining steps by mail. On 26th he wrote explaining the routine for curing and stripping the deck. It should be possible within two days to remove the formwork supporting the cantilevered footpaths, but Tisdall should judge the maturity of the mortar, and whether it could stand the weight of the handrailing and kerbs. It would not be fair to put much weight on it for a week. However, it seemed that not all was well. "You do not say where the two "drummy" patches are, nor what area they cover. I assume those cracks have got no worse." Probably, JM have to go again to Koondrook on 1st or 2nd March, so Tisdall should anticipate all questions that might arise in the next 10 days.

On 1 March, Monash recommended removing the formwork. "By now the concrete should be very hard and ring under the blows of a light hammer. It should be difficult to dislodge particles of gravel with the point of a knife." The sample in Melbourne was setting excellently. If the rest was as good, all the formwork could be stripped, except the boards under the girders. A few days later, JM wrote: "The showing of wires underneath is a pity. Where the exposure is slight, a thick wash of liquid cement upon the metal would be better than plastering which is apt to fall off. But where there is any recess or cavity, plastering is best."

Robertson was informed on 10 March that the bridge was complete and almost ready to open to traffic. Monash would travel up within a few days to have the last of the framing removed, and formally hand over the bridge. However, he was still worried. He told Tisdall to avoid using a roller while laying the road surface, and if he could not avoid it, "it had better be done with a horse roller before the striking of the centres". A photograph received on 12th seemed to show "a decided sag in the line of the main beam". JM asked "Is this really so?"

A meeting was arranged with Tisdall at Craig's Hotel in Ballarat, at 7.40 p.m. on 15th March. Monash told him to bring all the information he had about extras and finance, so that a settlement could be thrashed out with Robertson. The visit must have convinced JM that the deck was in serious trouble. Back in Melbourne he made calculations for steel components to be inserted into the structure to effectively "strap up" the cracked girders. Shortly after, a visit to the bridge by Reed and Robertson passed without incident. JM told Tisdall: "As far as we could judge they suspected nothing, although Reed tells us Robertson remarked and commented upon the longitudinal sag in the overhang and the handrails."

The major problem was what is commonly known as 'shear failure' - the development of inclined cracks, generally near the end of a beam, as illustrated in the diagram below. Because it is caused by tensile stresses acting across the line of the cracks, the phenomenon is more accurately described as 'diagonal tension failure'. The girders at Stawell Street evidently had inadequate strength in this respect, even to carry the weight of the deck alone.

It is possible to guard against this type of failure by selecting appropriate dimensions for the concrete (in this case the web of the T girder) and/or including vertical or diagonal reinforcement. The records leave open the possibility that the girders, as constructed, were not provided with such reinforcement. If it was included, it was far from sufficient.

Strengthening, Payment and Testing.

Remedy.

Monash's solution was to add the reinforcement externally. Two sloping 'tang bolts' were installed at the end of each main girder, as shown at the left of the diagram below. The bolts were anchored behind the mass concrete block that ran across the width of the abutment. They crossed the shear cracks in such a way as to prevent them from re-opening next time the wedges were removed. The ends of the sloping bolts were attached to a pair of vertical tang bolts, and a U-shaped piece passed under the bottom of the web. These components were connected together by a threaded rod that passed right through the web of the girder. A flat steel bar, hooked around the bar of the U piece, ran under the soffit of the girder, to connect up with a similar arrangement at the other end (not shown in the diagram).

A tang bolt consists of a rod welded to a flat bar. The free end of the rod has a thread and the free end of the bar has a hole for a bolt.

Diagram showing the situation described in the text and caption.

Monash's method of strengthening the main girders.

In the diagram above, added components are shown in brown. For clarity, they are shown at the left-hand end only. Tang bolts at the right-hand end mirrored those at the left. Also for clarity, the position of the cross girders is not indicated, and only typical 'shear' cracks are shown.

Monash's calculations for this arrangement date from 16 to 19 March. On 23rd he wrote to Tisdall: "On the question of the occurrence of those shearing cracks, we have very exhaustively examined all our technical textbooks on the subject, and the more the matter is investigated, the more mysterious the occurrence becomes, as we find that in all respects our design was at least stronger than numerous examples of similar work described in our books of reference. It will be very interesting for you to go into this matter more fully with us on your return to Melbourne. Meanwhile our investigations give us increased confidence in the efficiency of the bridge upon the completion of the added reinforcement." He complained that the "metalwork suppliers" [Cowley's of Ballarat] were taking their time.

In Monash's more formal letters to Tisdall he used the first person plural because he was writing on behalf of the firm. In private communications he began "Dear Tisdall", which in the conventions of the time was more friendly than "Dear Mr Tisdall".

Installing the steel "frame" as JM called it, necessitated drilling through the end block and the deck of the bridge and then fitting the tang bolts through the holes and around the web. Correspondence suggests that the drillers struck reinforcing bars and were obliged to re-drill alongside the original attempt. Monash described the slow and fumbling progress as "heart-breaking" and was concerned the washers would now bear on damaged concrete. He warned Tisdall to use washers of ample size and bearing area. "If this not be done, there will be a great danger of the main anchorage of the new system yielding and permitting sufficient deflection to make the shearing cracks recur. This would be fatal to the whole scheme." In addition, the long delay was becoming dangerous to the reputation of the firm. Tisdall should not wait for instructions by mail - this was a legitimate occasion for use of the telephone.

Having battled the leaking cofferdam and other problems at Koondrook, while juggling a court case and other engineering projects from the Melbourne office, Monash now warned Tisdall that he would be absent from Thursday 31 March to Tuesday 5 April in order to attend an Easter military camp. "We are beginning to be anxious lest you will not be able to complete the added reinforcement, and the plastering over of same well before the Easter holidays". JM wanted Tisdall to complete the work and strike the wedges, so as to be assured that the strengthening program had worked. However, "in the event of the worst happening", and "any further alarming symptoms", Tisdall should put the wedges back, as JM's other commitments would prevent him from coming to Ballarat.

Tactics.

The work of strengthening the bridge was finally completed about 6 April. Monash was confident that it would safely carry the steam roller, but feared that anything more than normal traffic loads might re-open the shear cracks. He told Tisdall: "The best thing is to know nothing whatsoever about the test; there is no test specified, and as far as I am concerned I do not want to be party to a test. Of course if they care to test the bridge they are welcome to do so". A heavy wooden waggon would seem to be about as good a test as they could apply, certainly favourable for M&A, as it would not be likely to exceed 2½ tons per wheel. JM added that "it would be sensible of them" to test the bridge before the falsework was taken away, but this would interfere with the plastering, so the best policy would be "to take away the under-structure, complete the plastering and clear out, leaving them to test the bridge in any way they like and when they like". Tisdall reported the following day that there had been no "unforeseen effects" following striking of the wedges.

To minimise the danger of the cracks reappearing, JM gave instructions to have the road surface as smooth as possible to minimise vibration. Tisdall must try to persuade Robertson to have the roller pass down the centre of the road, so that its load was shared equally between the two main girders. It should go as quickly as possible consistent with "minimum vibration", and make only one or two passes.

However, it now appeared that the Council was in no hurry to test the bridge. Monash even wondered whether Tisdall was chasing the Council for the test, rather than the reverse. He told him: "the undoubted proper policy is to get the under-structure [falsework] away, return the plant, and clear out of Ballarat as fast as possible. Call on Robertson, and formally and briefly intimate to him that the work is done and he may open the road whenever he likes." JM and Tisdall could then make up the final account and submit it to Council. "That will soon bring matters to a head. When they are compelled to talk "test" on paper, we can put a different complexion on the matter." He sent a final instruction on 12th telling Tisdall to close down the job and return to Melbourne.

Claims and hand-over.

In the meantime, JM had been trying to settle the question of 'extras'. With his meticulous record-keeping, he was able to set out the history of the contract. The drawing submitted in December 1903 with the specification and quote had become part of the contract when the Council accepted. This drawing showed no wing walls whatsoever. It had been Robertson's idea to change the location of the abutments, and when JM had pointed out that this would expose the southern approach to scour, Robertson himself had suggested that additional wingwalls be built and later confirmed the new arrangements in a letter. JM declared that this should leave no doubt about the soundness of M&A's position regarding the claim for extras. A similar case was presented for work on the enlarged foundations.

On 13th, the firm sent the Council its final account for Stawell St, "taken over by you on the 12th inst.". This claimed, in addition to the original price of £230, extras of £85-10-00 for work on the wingwalls. Progress payments to date had amounted to £220, leaving £95-10-00 still to be paid.

Back in Melbourne, Tisdall carried out the customary detailed analysis of the costs involved in building the bridge.

Waiting.

The Town Clerk, Alex Fraser, remained reluctant to settle the account before the load test. After some confusion, it was agreed that the Council would conduct the test with Monash in attendance. During this time Tisdall made two trips to the bridge and reported that the largest loads crossing it were wood wagons weighing up to 9 tons (3 or 4 trips per day); hay wagons of up to 4 tons; and a clay cart on two wheels (20 trips per day). Other vehicles, which included tradesmen's carts, residents' traps, and market gardeners' carts, brought the total crossing per day to around 40. There were fine cracks under the deck and "cracks and splintering at abutments". However, Tisdall thought the latter were due to the settling-in process. A sketch suggests that surface cracking in the deck was due to inadequate cover to reinforcement.

Throughout this waiting period Monash continued to search for an explanation of what had gone wrong. He analysed the strength of a T-beam tested at Gorinchem in the Netherlands, and compared its stresses with those at Stawell Street. He also checked and re-checked his calculations for the strength of the added frame, accepting now that the roller might travel over any part of the bridge and attempting to work out which position would induce the greatest stresses in the added frame. He also checked the strength of the slab panels for the load of an individual wheel. His calculations for 19 June end with: "Conclusion: Bridge perfectly safe for test of 16 tons".

JM was also ready to assist Tisdall in his career, writing references for job applications and eventually appointing him for a while as a travelling representative for the firm, drumming up work in country Victoria.

The test.

The date of the test was finally set for 11.00 a.m. on Tuesday 26 July 1904. Monash was busy and asked Tisdall to attend, saying he would wait all afternoon for a telephone call with news of the result. This was not entirely good. As the Argus reported:

A slight mishap occurred while the new Monier bridge in Stawell-street, Little Bendigo, was being tested this morning. The bridge, which was built at a cost of £325, and has a span of 20 ft., is constructed on the Monier parallel girder principle. It was arranged to subject it to the stipulated test this morning, and the corporation steam roller, which is 16 tons weight, was utilised for the purpose. The roller had almost passed over - in fact, the back wheels were above the northern abutments - when the decking cracked, and the 20-inch wheels subsided on to the brickwork at the side of the creek. In an instant the ponderous locomotive came to a dead halt and the driver hastily shut off steam. Mr William Robertson, the engineer for the Ballarat East Council, and Mr H F Tisdall (representing Messrs Monash and Anderson) decided to raise the back of the roller with screw-jacks, and after placing planks beneath the wheels it was brought back over the bridge to safety. The girders and bearers stood the test without the least move, and the whole of the decking, except the portion near the northern wall, remained perfectly solid. Mr Tisdall states that the repairs necessary will only cost £2, and the bridge will be ready for traffic in a few days.

Again, Monash was baffled by the outcome. His notes attempting to understand the failure include several pages of sketches and calculations. He now knew enough to check not only the bending strength, but also the shear strength and the adhesion (bond) between bars and concrete. On 3 August, Tisdall reported that the bridge had been repaired. It appears that he had been ill and Robertson had supervised the work. The Council forwarded the remaining £95-10-0 on 17th August. In September 1905 The Age newspaper reported that Robertson was not speaking to the Ballarat East Councillors and had served them one month's notice of resignation.

Replacement?

Late in 1910 Alex Lynch examined the Stawell Street Bridge and sent a confidential memorandum to Fairway. After listing faults such as permanent sag, cracking, and vibration under light traffic, he concluded: "In my opinion this bridge is only safe for light traffic for a few more years. A heavy load, say over 5 tons, could be dangerous." Monash estimated the cost of replacing the bridge at £70 and wrote to Gibson: "As I entirely concur with you in the desirability of having this work done, I shall be glad if you will make an appointment at your early convenience to discuss the project of getting Mr Fairway to negotiate the reconstruction".

Although the above suggests a clear intention to replace the bridge, I have not come across any evidence in the RCMPC files to confirm that it was done. The present bridge at this location is obviously newer than c1910. A.H.

Analysis.

What went wrong?

The evidence I have found in the RCMPC files is not complete or consistent enough to allow a definite conclusion. However, it appears that at the time he designed Stawell St, Monash was not fully appraised of the danger of 'shear' failure, or of the theory and practice required to prevent it. JM prepared tentative designs for several other T-girder projects about the same time, and none of them contain calculations for shear strength.

This can be understood, given the state of knowledge at the time and the fact that Monash had come to T-girder bridges from arch and slab bridges in which shear (or diagonal tension) was not a significant danger. An examination of the contemporary texts that I have been able to lay my hands on suggests they did not treat the phenomenon fully, or presented it piecemeal. In an English-language text published the following year, A W Buel stated: "shearing stresses in reinforced concrete beams have not as yet been satisfactorily analysed, and few if any experiments have been made that throw much light on the question".

If JM had checked the shear stress in his girders during his calculations of 1 January 1904, he would have found that self weight alone produced 75 psi (0.52 MPa), equal to Buel's maximum recommended figure without shear reinforcement. Late in July he did check shear stress allowing for the roller, and came up with 130 psi (0.90 MPa). However, he compared this with the girder at Gorinchem which had apparently withstood 223 psi without distress, and remained mystified. In later practice, the recommended maximum shear stress without reinforcement was reduced to 30 or 40 psi (0.21 or 0.28 MPa), and it is now considered necessary to provide stirrups in all bridge girders regardless of the level of shear stress.

The failure of the main girders at Stawell Street at such a low load suggests that either the stirrups were too widely spaced, or that none were installed. While the undated drawing which appears to be the tender document does show some stirrups, it gives no indication of size or spacing. The undated large-scale sketch showing the main reinforcement, probably prepared to help Tisdall avoid the bars when drilling to install the added components, shows no stirrups whatever.

A further consideration is that, although JM went to Ballarat for the casting, he was working through an inexperienced crew, and the concrete might have been of inferior strength. There are references to "drummy patches" and to reinforcement lacking cover. It is quite likely that the relatively dry concrete employed was not properly compacted underneath the main reinforcing bars, a task that was then done by hand.

Why did Monash choose to patch up the bridge with the "added frame", rather than demolish it and rebuild, as at King's Bridge, Bendigo? Possible answers are that he was still experiencing severe financial difficulties; a further bout of bad publicity might have wrecked the business; and the public and Council were anxiously awaiting the overdue opening of the bridge.

The experience at Stawell Street spurred JM to more intensive study of the problem of 'shear' in reinforced concrete beams, including correspondence with Fritz von Emperger, the European authority on the reinforced concrete and editor of the key serial Beton und Eisen. JM organised research of his own, carrying out tests at the University of Melbourne. He continued to design T-girders for buildings, e.g. the roof at Raveloe and floors for office buildings, in which the concentration of load is less severe than in bridges. He soon felt confident to initiate the series of T-girder road bridges over the Canal at Elwood (Melbourne). One of these is still in service under modern traffic at the time of writing (2006).

More technical detail.

Monash's information on reinforced concrete came from four main sources: what he called the 'Wayss Handbook' the serial Beton und Eisen; Christophe's two volume text; and JM's colleague W J Baltzer in Sydney. I understand that the 'Wayss Handbook' was the text written by Emil Mörsch and published by Wayss & Freytag in 1902.

I have not been able to examine copies of Beton und Eisen, but have seen an English translation of the theoretical part of Mörsch. Early in the book he leaves no doubt that shear strength is important and implies that stirrups should be included in beams as a matter of course, based on practical experience. However, he is obviously feeling his way in his theoretical treatment of the subject, describing tests on beams specially modified to induce vertical and longitudinal shear cracks. There is no diagram showing an inclined diagonal tension crack. He suggests a permissible shear stress of 6 kg/cm2 or about 85 psi (0.59 MPa) - high by later standards - and suggests in an example that the shear force could be shared equally between concrete and stirrups.

In Christophe's text, information on diagonal tension failure is scattered widely through the 755 pages of Volume 1. (Volume 2 consisted of drawings of built examples, in most cases providing full details of reinforcement.) Emphasis is placed on the horizontal shear stress caused by the tendency for layers of the beam to move horizontally relative to one another. In this approach, the stirrup was seen to act as a dowel, rather than as a vertical hanger. Mörsch's development of the formulae, and his wording in a worked example of stirrup design, suggest that he adopted the same approach in his 1902 edition.

The author of a paper in the (British) Concrete Institute's journal as late as 1919 felt it necessary to warn that the concept of dowel action was untenable, because the adjacent concrete would crush long before the steel reached its yield stress in shear. The concept must have misled many designers, because it provided apparently scientific values for size and spacing of stirrups, while drawing attention away from the fact that they work in tension; need to be properly anchored in the compression zone of the beam; and should be spaced closely enough to intercept all potential diagonal tension cracks.

I have not examined a copy of Berger & Guillerme, which Monash also bought; but I cannot recall an instance of JM referring to this book.

Buel, also, did not include a diagram showing a typical diagonal tension crack. He suggested that under normal conditions a calculated shear stress of 75 psi could be allowed in a concrete beam without stirrups. Although Buel provided details of experiments on beams containing stirrups, he provided no method of calculation for their size or any guidance on their spacing. He states that he drew on contemporary texts and papers by authorities such as Considère, Christophe and von Emperger.

JM's choice of a system comprising two main girders with cross-girders meant that the entire load was carried to the supports mainly by four webs 9 inches thick and 25 deep, making it vulnerable to small patches of poor quality concrete. He conceived this scheme when preparing proposals for bridges of two spans across the Waranga irrigation channel, in December 1903. He thought at the time that it would be necessary to place a column under each girder, and that by minimising the number of girders he could minimise the number of columns in the central pier. In the Elwood bridges of 1905 he returned to a system of multiple main girders spaced at about 4 or 5 foot centres. Under this scheme, Stawell Street would have had perhaps five main girders.