THE FAILURE OF STEEL AND WROUGHT IRON WATER TANKS AND RESERVOIRS
Few fields of study are more fruitful of results and lead to more genuine progress than a study of the causes of failures. Such studies may be unpleasant and disagreeable, they may at times be even disheartening, but the man who would make substantial advances must heed the lessons which failures teach. It is true that valuable information can be obtained likewise from a study of materials which have given successful service.
During the past few years we learn of standpipes, water tanks and reservoirs collapsing in various cities in the United States, and it seems to increase year by year. Closely connected with the query as to the cause of failures is the ofttimes more important question, who is responsible for the failure? If the matter in hand is an experiment which we are making for our own information, the question of responsibility is small and is practically swallowed up in the cognate question of the cause of the failure. But if, on the other hand, the failure involves the loss of human life or the destruction of valuable property, the question of responsibility may be very grave. Failures may be ascribed to one of the following causes, viz.: bad material, bad workmanship, bad or faulty design, or to unfair treatment.
We will now go into detail with each of these possible causes of failure, hirst, bad material. This does not cover those cases where wrong kind of material was used or material not adapted to the work. If steel is used where wrought iron should have been employed, or the metal used was not the required thickness to withstand the strain and stress, failures due to these causes come under another category.
If the material for the standpipe or reservoir was purchased on the engineer’s specifications it is reasonable that it should he what the specifications called for. Here you cannot lay the blame to the manufacturing concerns that are furnishing the material for standpipes and reservoirs if specifications are not properly drawn.
Experience on this point has brought us face to face with explanations of the difficulty we are considering. First, and perhaps most important, is the price. It is frequently put up that waterworks departments will not pay the price to secure the material desired. Information is not usually given as to how high the wished-for price should run, needful to secure good material to go into steel and wrougtit iron tanks and reservoirs such as the manufacturers would like to furnish. Prices arc largely determined by competition, and in absence of something more than a verbal statement from the manufacturer, higher grade and heavier material would be furnished at a higher price. On the other hand, as time progressed, and large manufacturers began to study for themselves the behavior of materials in service, and as they’ began to employ their own experts, also testing devices, knowledge of the properties and characteristics of iron and steel began to widen. It is evident that the situation has changed and that where iron and steel is bought on definite specifications, the voice of the manufacturer as to quality is no longer potent, and the excuse that inferior material is suitable for the purpose is no longer entitled to consideration or weight.
Where a true selective economy is exercised in the purchase of such material, the choice is not determined by either quality or price, but by the relative industrial efficiency of the material under consideration. The opportunity to economize may therefore lie in the substitution of a high priced material for a cheap one, viz.: in a change from a coal which the fireman likes to one which gives more heat for a dollar.
The second cause for failures in steel and wrought iron tanks and reservoirs, is bad workmanship. The tendency to slight the job is universal. A rivet or bolt has been substituted for a smaller one, or cross-braces have been eliminated. thus causing increased strain on those which were formerly put in. The rivets are not properly drawn so as to make perfect, water-tight seams, and here is where, in nine cases out of ten, corrosion sets in, and the rivets are practically gone before any pitting or corrosion is detected on the surface of the iron and steel. There is little doubt that individual experience will furnish numerous instances of bad workmanship, and will recall many an engineer who did not hesitate to declare that bad workmanship was the principal cause of the failures in service.
Those of you who have frequently been brought into contact with bad workmanship have, no doubt, like myself, often wondered why work was so badly done. Not infrequently, in contemplating a failure in which bad workmanship played an important part, we have said to ourselves that there was absolutely no excuse; and yet, since workmanship is so important an element in our theme, it may not be amiss to go a little deeper into the matter. Everyone will recognize that lack of skill, inefficiency and laziness are important elements in bad workmanship.
The engineer who makes or decides upon the design of a structure carries a heavy load of responsibility. He is first in the Field, and practically tells all who follow what is to be done. It is he who must decide on the material, also the amount and sizes, and how it shall be erected; his work covers the entire structure from the foundation. The engineer who makes the designs works under a serious problem, computing the strain and stress to which the whole structure will be subjected, and the construction of the foundations, etc.
Unfair treatment, as already indicated, proves that there is a natural disposition on the part of us to relieve ourselves from blame and put the fault on someone else; but, if our observation is worth anything, there is no ground for parceling out desserts among those involved in failures, and the responsibility, therefore, is more fertile than this one of unfair treatment. This field is the especial paddock of the manufacturers or designer of the structure. If a steel or wrought iron tank or reservoir has been in service a number of years without inspection or painting, thus allowing pitting and corrosion to set in on the metal and on the scams and the rivets that hind the seams, the entire structure is weakened. Thus it is plainly evident that unfair usage is many times a legitimate explanation of failures.
For an example, a steel or wrought iron standpipe or reservoir had been in constant service for a number of years, without any inspection or repairs. The proper paint had not been used or applied in the proper manner. If there is not enough money expended for painting and repairs, the parts become weakened by pitting and corrosion to such an extent that it results in failure and collapsing. Thus it is entirely obvious that failure must be attributed to unfair treatment.
The writer was called on to make an inspection and report, a few months ago, on a steel standpipe for a waterworks department in a city of 20,000 inhabitants. After the water had been drawn from the standpipe, and an internal inspection had been made, it was found that the standpipe had not been painted for five years, allowing corrosion and pitting to set in quite extensively. Pit holes were found to a depth of 3-16 inches, in various places. An external inspection was made, and the seams were found to be leaking quite freely, especially around the rivets.
Here is where most of the waterworks department officials fail. They depend on the paint to fill the crevices of the leaky seams, and thus stop the leak. It may stop the leak for the time being, but these places still remain weak, and corrosion is still on its way. This particular reservoir or standpipe that was inspected, had over 600 rivets taken out of the seams and new rivets inserted, as in almost every instance the body of them was nearly gone, and only half or one-third of the head remained. The failure of the White Plains, N. Y., reservoir, on October 20, 1909), can be laid to bad seams, in which corrosion had been going on for years. The leaky seams were all chipped and recaulked.
Another point where a great many mistakes are made is in the selection of proper metal coating. You have, no doubt, observed where paint has been applied on metal the effect is to scale off after a certain length of time, due to the fact that the paint had not the linear expansion of the metal itself. Good metal coating, when properly applied, should expand and contract with the metal.
In preparing a steel or wrought iron tank for painting, the same should he thoroughly sandblasted, inside and out, so as to remove all corrosion and rust on the surfaces. It is not enough to use the best kind of a protective coating ; it must be properly applied. The first thing to be considered is the preparation of the surface. It makes no difference whether it is for a protective or a decorative effect. All loose scale and rust should be removed by the sand-blast process, and painting should be done at once. The metal coating applied on this particular standpipe was magnetic oxide (Fe3O3) 95 per cent. pure. It had the same linear expansion as the metal itself.
The water in the standpipe or reservoir should lie drawn off once a year, and the standpipe given one coat of magnetic oxide paint. While the life of a steel standpipe or reservoir is estimated to be about 50 years, wrought iron will last 100 years. A little time and money, properly expended in preserving these structures, together with the proper material, will undoubtedly increase the life 75 to too per cent., and will pay a high rate of interest on the investment. “What is worth doing is worth doing well.”
Report of Water District of Portland,
The trustees of the Portland, Me., water district, which includes the city of Portland and the Deering, South Portland, Westbrook, South Windham and Cape Elizabeth districts, in presenting their report for the fiscal year ending December 31, 1909, call attention to the fact that while it is the first year during which the water district has had exclusive control and manage incut of the waterworks, a substantial increase of net revenue over all preceding years is shown. The works were purchased in August, 1908, the entire expense of acquiring the property having been $152,643.88. The profits already earned have been sufficient to pay the whole cost of acquir ing the property and leave a margin of $43,500.57. During the past year the trustees refunded the temporary loan contracted for the purchase of the works, thereby placing the Portland water district on a sound financial basis. The source of supply is Schago Lake, seventeen miles from the business center of Portland; the waterworks were built by the Portland Water Company and the Standish Water and Construction Company. The system is pumping from the lake to reservoirs for distribution. The total length of the main lines and distribution pipes, in use January 1, 1910. was 195 miles 5,072 feet, equipped with 1.240 gates and supplying 955 hydrants and 12,062 service taps, of which 2,646 were metered The total income of the district for the year, from all sources, was $312,885.98, of which $303,-467.55 was from water rates. The operating expenses were $37,115.88 and interest charges of all kinds, $158,764.47, a total expense of $195,800.35, compared with $221,356.69 for the preceding year. The year’s net revenue or profits amounted to $117,005.63. Last year’s profits were $79,138.76, so that, for the two years the waterworks have been run as a district institution, they have made a profit of $196,144.39. ()f such a re-
sult the trustees and their efficient superintendent. Eben R. Dyer, have good cause to be proud, while the quality of the water supplied to the district and its abundant available quantity are such as to afford the best satisfaction to consum ers.
Burlington Filter Plant.
Burlington, N. expects to have its new filter plant in operation by July 15. It will supply 3,000,000 gallons daily, although the present consumption does not exceed 1,000,000 gallons. Four large filter basins are being installed, each with a capacity of 100,000 gallons. The sedimentation basins will hold 210,000 gallons, while the clear water basin will hold 200,000 gallons. The water will be pumped into the standpipe, whose capacity is 282.500 gallons.