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

Miscellaneous Topics. Page 1.

[Main JM Index.]   [Buildings Index.]
[Abbreviations.] [Units & Currency.]

Safety and Risk Management.

Engineering design of Mitchell's Building.

Under this heading are discussed two issues that led to a difference of opinion between Monash and the Melbourne City Building Surveyor, E H Morton, regarding the engineering design of Mitchell's Building No.2.

JM's refusal to design the floors for 164 psf live load.

There are two issues here: (a) the contravention of regulations, which cannot be condoned; (b) the actual safety of the building as constructed. In the mid-20th Century the British Standard recommended that warehouse floors be designed for either 100, 150, or 200 psf depending on the proposed usage. JM's 120 psf for this building is comfortably within the range. His design for the New Zealand Wool Store allowed for 224 psf. No doubt he had good reason for choosing 120 for the Mitchell Building. The building is still standing, used now as apartments for which the recommended design load is very much lower.

JM's arguments for not increasing the size of the basement columns.

Again this raises interesting questions of safety or 'risk management'. In designing a structure we estimate to the best of our ability the nature and magnitude of the loads to which it might be subjected. (Some we simply forget about - although trucks do crash into houses, few of us are willing to pay for and live in a bunker designed to resist truck impact.) We do our best to estimate the forces and stresses that will normally occur in the structure. We do our best to estimate and control the quality of the materials that will be delivered to the job or made on site. We do our best to control the quality of the work that goes into assembling the structure. Because we are aware that there will be significant variability in all these factors, we aim to make it stronger than our cogitations suggest it needs to be.

Should we ever consciously encroach on this factor of safety? If we aim to make a structure twice as strong as we think it needs to be, it might be, for the sake of argument, 2.5 times as strong, or 1.5 times. We do not know. Any additional erosion of the factor of safety takes us a bit nearer to 0.99999, which means disaster. It is ironic that widespread knowledge of the fact that engineers design structures with a margin for safety may encourage managers and workers to cut corners and encourage users to abuse the structure. From time to time a sufficient number do this to whittle away the safety margin, and the structure collapses. Safety is as much an attitude of mind as a question of arithmetic.

That said, it is unlikely that JM's action in this case was particularly significant, though it is difficult to be precise. The stresses quoted are extremely low, being roughly one third of those used even a few decades later. On the other hand, understanding of concrete was in its infancy in JM's time. (According to Bussell [p.303] it was 1917 before the importance of compaction was realised and 1918 before it was realised that a high water-cement ratio meant lower strength.) Although the 'Europeans' [i.e. excluding the British] used higher permissible stresses, they measured the proportions of cement, sand and aggregate by weighing rather than by volume, ensuring less variation in the quality of concrete. JM was aware of this, which may be why he said he would not himself have pushed the local product to 15 kg/cm².

JM and the Expanded Metal Controversy.

Late in 1907, the journal Building published an article favourable to the use of expanded metal as reinforcement in concrete construction. This competing technique presented a serious threat to Monash's business. Building described itself as "The Magazine essential to the successful Builder, Architect, Craftsman, Property Owner, and Merchant" and its letterhead carried the banner "Under the auspices and cordial support of the Master Builders' Association of NSW and other leading Building and Architectural authorities in Australasia". On 21 December 1907, JM wrote to the editor, George A Taylor, complaining that the use of expanded metal was "unscientific" and protesting about the positive tone of the article. Taylor had part of JM's letter copied and sent to people who had a stake in the rival system.

1. Refer to text below. 2. Refer to text below.

1. Diagonal expanded metal. (Christophe p.21.)
2. Kahn expanded metal. (Bussell p.299.)

Expanded metal is manufactured by making a pattern of slits in a steel plate which is then stretched sideways to open up the slits. The type most commonly seen nowadays in gratings and grills is shown in (1) above. The slits are expanded to a diamond shape. The type patented by Kahn is shown in (2). This resulted in rectangular openings. It left continuous strips of metal running in one direction, joined by weaker links at right angles. Both were used as reinforcement in floors and bridge decks in JM's time.

Monash's original letter is reproduced below. The paragraph that Taylor said he had sent to supporters of expanded metal is shown in italics.

Dear Mr Taylor, I trust you will not regard what I am going to say as presumptuous, but I have the interests of your Journal at heart and desire, from a scientific point of view, to do all I can to keep its matter on a high plane. Now, in your last number, I see you have devoted a good deal of space to the employment of Expanded Metal. Of course, there cannot be the slightest objection on any ground to the full illustration of actual practice in this regard, and I have not a word to say in disparagement of the action taken in giving a full account of what has been done in this field. But I notice that in the letterpress, in one or two places the writer speaks in high approval of Expanded Metal in preference to other methods of reinforcement. It is this aspect which I think is very debatable.

You may be aware, but if not I may explain that I am not an advocate for any particular system, as I practise Reinforced Concrete in its broadest and most scientific sense, and make use of the best practice I can find in any system, or can devise for myself. I am as free to use Expanded Metal in my designs and works as anyone else, or as I am to use any other type of reinforcement, but I hold very strong views upon the question of the radical unsoundness of Expanded Metal in all those cases where it is professed to be used as a positive tensile reinforcement. By this I mean that while Expanded Metal is harmless for such purposes as wall work or lining where it is merely used to hold the concrete together, it is very objectionable where it is relied upon to supply the tensile ingredient in any member or part subjected to cross bending [i.e. flexure].

The vendors of expanded metal naturally put their wares before the public in the best light and care little whether it is rightly or wrongly used, so long as it is used in large quantities. But sooner or later the ignorant use of Expanded Metal in places for which it is unfit will lead to disasters and collapses here, as it has done in America. I have had occasion to investigate and report upon many instances of the use of Expanded Metal in long span floor construction where the actual quantity of metal introduced into the concrete was only one-fifth or less in actual weight than the case properly required. Many Architects, entirely ignorant of stress calculations, imagine that a sheet of Expanded Metal embedded in a mass of concrete constitutes Reinforced Concrete. This view must be discouraged as much as possible if Reinforced Concrete is to be practised on sound scientific lines.

When you are next in Melbourne, if you care to discuss this matter with me, I shall give you abundant arguments which will appeal to your reason and judgment for the statement that Expanded Metal is not a sound scientific or economic form for use in floor plates, girders and the like. No European Building Regulations would permit its use. It is only where it is vigorously exploited commercially that it has found a footing.

While this letter is not intended for publication, you are quite at liberty to use the information it contains in any way you think fit. I have no object to serve in submitting these views to you, except a desire to give you the benefit of the best scientific thought which is at my command on this subject. Both Mr Gummow, of Sydney, and Mr Robertson, of the Ferro-Concrete Company in New Zealand, who with myself are the only three Engineers who have seriously and scientifically practised Reinforced Concrete in Australasia, will thoroughly confirm the views which I have expressed.

It is interesting that JM made no mention of Baltzer.

On 6 January 1908, Edwin C Elliott of Elliott, MacLean & Co, Structural Engineers, 6 Dean's Place, Sydney wrote to JM:

Dear Sir, As yet I have not the honor of your personal acquaintance, but hope when I next visit Melbourne the opportunity may occur of my securing that pleasure. We are both interested to an important extent in Reinforced Concrete, but until recently my efforts have been expended principally in this somewhat conservative State. About a year ago, however, I took a trip through the other States and New Zealand, when I was rather surprised to hear, in more than one or two instances, of disparaging remarks respecting Expanded Steel, alleged to have been made by a reinforced concrete expert. Apparently no tangible statements had been made, but only that sort of loose general remarks, such as I have sometimes found used towards each other by the lower types of business rivals in this part of the world, no doubt with the idea of smartness, but which are the more annoying because not easily traceable.

My business then called for a trip to England from which I returned only a few days ago, when I found to my intense surprise that you had, doubtless without due consideration, fallen into the same injudicious course, and to the further extent of putting it into writing.

Now the object of this letter is to suggest to you as a fair minded business man, that the pushing of Reinforced Concrete being our common purpose, it is unwise as well as altogether unnecessary to traduce competitors' systems, particularly those which have an immense record of entirely satisfactory work behind them.

That we are each within our rights in lauding the excellence of our own methods goes without saying, but disparagement of those of others is tactically bad, lowering to the status of the person descending to it and likely to produce mischievous results in boomerang fashion.

The gratuitous assertion of being 'the only etc.' is purely a matter of taste and I thought was the sole privilege of the Universal Provider Storekeeper, but I can assure you that amongst the five or six million of people in Australasia there are other intelligent and capable men who have studied the principles of reinforced concrete and put them into successful practice.

I must apologize for troubling you with this long letter, but I feel that it is best we should understand one another and am confident that if we had the advantage of personal acquaintance there would have been no possible suspicion of 'hitting below the belt'.

I desire to state that neither myself nor any one connected with my firm had anything to do with writing the article in Building which appears to have troubled you, nor had any knowledge of it before it appeared in print.

On 23 January, George Taylor wrote to JM:

The stress of issuing my January issue precluded an earlier reply to yours of 15th inst. I do not follow that there has been any breach of confidence, as already pointed out, and a perusal of your letter does not convince me of any indiscretion in desiring some verification of the statements. I may point out that I have been connected with Journalism for some twelve years, having had my contributions published in …

Coming from such a capable source as yourself, any remarks as to danger of using any questionable reinforcement carried considerable value, and as I am continually advocating reinforced concrete as the latest and best in constructional science, it does not need me to remind you that any possible question of failure has to be avoided, particularly as reinforced concrete is under the watchful eyes of every qualified Architect and professional man in the Commonwealth … My letter of 13 Jan covers everything that need be said.

On 31 January, Taylor wrote a letter marked "Confidential" to JM:

I hasten to reply. Your letter of December 21st was of such importance that I called for particulars from the expanded metal representatives. In fairness to you they were only acquainted with the extracts from your letter enclosed. [Taylor here refers to the portion of JM's original letter (above) distinguished by italics on this web page.] There has not been any breach of confidence, as your letter was not marked confidential, and you gave me the liberty to use the information in any way I thought fit. So having mentioned expanded metal in my notes I was just as willing to report against it if deemed necessary in the interests of the public. I thank you for your information of today which I am using with kind acknowledgement. I am also printing extracts from your paper read at the Victorian Institute, so that I consider you will be pleased with the forthcoming number.

Expanded metal first appears in our research notes in April 1905 when Monash told Gummow that the Expanded Metal Co of New York had "active agencies in Victoria vigorously exploiting the use of expanded metal for floor plate construction". At current prices JM did not fear competition but thought it would be fair to him to exclude expanded metal from the classes of work on which he had to pay royalty.(Gummow agreed to this.) In February 1906 Gummow told JM that he had lost a job in Sydney for NMLA because the architect's specification permitted alternatives to the Monier floors designed by GF&Co, and he had accepted a "cheaper" and "inferior" floor system with expanded metal. JM continued: "It will serve no useful purpose to commiserate with you on the shabby treatment of [by] the Architect in question. I strike some men of the same class here occasionally, though it has not so far resulted in any damage to our interests here. I have adopted a rather militant policy in Melbourne, and whenever a favourable opportunity offered, have decried the unscientific character of expanded metal construction so far with excellent effect." JM added he knew Col. Templeton, the company's general manager, and Mr Stock their actuary. "They leave their architects very much alone but Stock is favourable to Monier."

Mention of competition from expanded metal continues through 1907. By 1908 at least, RCMPC adopted the practice, when preparing a bid, of estimating the price of a competing bid in expanded metal. Costs for the Collingwood portion of the reinforced concrete slab covering over the Reilly Street drain were tabulated as:

"our type" £2615-17-00
Expanded Metal Design £2924-08-00
Tolhurst's Design £3722-04-00

E. H. Tolhurst was the Collingwood City Engineer. Reilly Street is now Alexandra Parade.

In tendering for a second New Zealand Mercantile Wool Store, RCMPC estimated that a competing design using steel columns and girders, and floors reinforced with expanded metal would come in at £1803, compared with "our design £1811 … for an honest 2 cwt per square foot with factor of safety of 4 while theirs is not". They calculated that for the same design load, the opposition would have a factor of safety less than 2.5.

On 29 May 1909, The Age reported "A Street Sensation". A 30-foot section of the Reilly St drain cover in Clifton Hill had collapsed and a dray, with its horse and driver had fallen into the resulting hole. (The horse was not freed until the following day.) As any form of reinforced concrete tended to be associated with the term 'Monier', RCMPC went into damage control mode. Lynch went to investigate and reported that the portion that collapsed was a 3-inch thick plate reinforced with expanded metal, supported by ribs each containing 2 Kahn bars. The concrete was "very poor and irregular, say 1:3:6" and the expanded metal was rusty. P T Fairway reported that the metal seemed to have fractured with little extension. There had been little adhesion [bond] as the metal had ripped out of the adjoining concrete. He concluded that the reinforcement appeared to be insufficient.

3. 4.

Collapse of a drain cover at Barkly St, reinforced with expanded metal. (Not designed by RCMPC.) University of Melbourne Archives, Reinforced Concrete & Monier Pipe Co. Collection.
3. General view. GPNB/1162.
4. Detail, viewed from the opposite bank. GPNB/1163.
Note. The index of negatives in the RCMPC Collection listed the above undated images as "Barkly Street Drain Collapse". So far we have found no reference in the RCMPC archives to Barkly St - only newspaper reports and correspondence referring to a Reilly Street collapse.

Monash wrote to the Editor of Cazaly's Contract Reporter to tell him that RCMPC was not involved and "following so closely on the collapse of the Mittagong Tank, NSW, this is another instance of the danger of entrusting this class of construction to non-specialist firms. This only bears out the experience of Europe and America". He also wrote to Catani:

Mr Healy, of your office, rang me up today to ask whether we had anything to do with the Reilly Street Drain Cover which collapsed last week. To prevent any misapprehension, I am writing you to explain that we had nothing whatever to do with this work. It was carried out in Kahn bars and Expanded Metal.

In my opinion the failure was due to a combination of bad design, defective materials and defective workmanship. The design was bad in providing less than half the quantity of steel reinforcement which I consider the case properly required, and in making no proper provision for shearing stresses, and also the concrete was specified to be weaker than I consider suitable for reinforced work. The materials were defective in respect to the choice of a class of so called reinforcement of very low ductility, as shown by the fact that all the steel broke off short with a crystalline fracture. As to workmanship, the person building the work seemed wholly ignorant of the necessity of mixing and consolidating the concrete in a manner to secure adhesion between the steel and the concrete, a fact which was practically wholly neglected. I enclose herewith a couple of photographs showing the failure. A section of at least 300 square feet of drain collapsed, which was supposed to carry safely an added load of 120 lbs. per square foot:- that is about 15 tons, with a factor of safety of 4. It, in fact, failed completely under the load of a horse, dray and earth, weighing at most 3 tons.

I think this accident bears out what I have always contended, that it is not safe to trust Reinforced Concrete to inexperienced hands.

On 30 May 1910 a Percy L Perry wrote to Monash (who happened to be overseas) reminding him that the previous February he had called at the Melbourne office "on behalf of Mr L F Atkins, the Manager of the Monier Co. of the West" regarding Diamond Mesh Lattice Reinforcement. Atkins has asked Perry to forward a sample of Atkins's patented DMLR and to "ask you whether you can make him any offer". The sample was in transit, accompanied by the results of tests done by the Government of Western Australia which "are most decidedly in favour of the Diamond Mesh". The cost of manufacture in WA was about 2/- per super yard, being 1/- for wire and 1/- for making. The RCMPC file contains two pages of test results and sketches of plate failure crack patterns. About a week later, Fairway replied, as Acting Superintendent: "Immediately we received your sample, the [DMLR] was thoroughly gone into and it was found that no economies could be effected. We are quite unable to make any offer."

Thereafter, if our research notes are a true indication, there is little mention of expanded metal in the RCMPC records, although clients and competitors did specify it from time to time. Monash's earlier comment about the unscientific nature of expanded metal reinforcement certainly applies to the diagonal form in which the metal does not lie in the direction of greatest stress at the centre of the floor panel, and the load transmitted through the metal is obliged to follow a zig-zag path. The tendency of the 'zig-zags' to straighten out under load must have tended to disrupt the surrounding concrete. The Kahn version of expanded metal (illustration above) does have the advantage that the main bars are straight and continuous. Its stiffness in the transverse direction is suspect, but this could have been overcome in slabs by adding a second layer with its main bars running at right-angles to those of the first. Judging from JM's comments on the Reilly St collapse, the process of expanding the sheet appears to have reduced the ductility of the metal, which is a serious disadvantage, resulting in a more sudden mode of failure.

Note.  JM conflated two arguments in his original letter to Taylor. The first was that unqualified people were using expanded metal without knowing that the thickness of the concrete and the amount of reinforcement had to be calculated according to the intended load. This argument could, of course, apply to any system of reinforced concrete. The second was that the mechanical efficiency of expanded metal was lower than that of a mesh of straight rods laid parallel to the edges of the slab. This was valid. Because the basic idea of reinforcing concrete with metal could not be patented, patents had to be based on some novel way of arranging or forming the reinforcement. This was the main reason for the use of expanded metal in floors etc. Its use in heavy structural work died out after the patents expired. However, during the period of our study, it was quite widely used. When an adequate amount was employed it gave satisfactory service and it did offer economic competition to other proprietary systems. (Today, reinforcing meshes, made from two layers of orthogonal rods welded at their junction points, offer the labour-saving advantages of mesh together with a rational orientation of steel.)

5. Two parallel rolled steel joists support a temporary timber floor. Diamond expanded metal reinforcement is laid and a portion of the concrete is shown spread on the timber planks to form the future floor slab. 6. Schematic drawing showing a rectangular grid of small diameter reinforcing rods  projecting from a concrete slab.

5. Expanded metal in floor slab supported on rolled steel joists. The formwork (shuttering) to support the wet concrete can be seen below the grid. The fact that this could be supported from the RSJs, and did not require propping from the floor below, offered savings in time and money. (Christophe p.412.)
6. Monier reinforcement. (To larger scale.) Grid of individual rods. (Christophe p.18.)