Bridge Disasters in America: The Cause and the Remedy

Two documents published some time since illustrate the preceding remarks. The first is the account of the tests of the iron taken from the Tariffville bridge after its failure, and the second is the specification for bridges on the Cincinnati Southern Railroad. The Tariffville bridge, though nominally a wooden one, like most structures of the kind relied entirely upon iron rods to keep the wood-work together. Although the rods were too small, and seriously defective in manufacture, the bridge ought not to have fallen from that cause. The ultimate strength of the iron was not what it should have been, but yet it was not low enough to explain the disaster; but when we look at the quality of the iron, we have the cause of the fall. The rods taken from the bridge show an ultimate tensile strength of 47,560 pounds per inch, but an elastic limit of only 19,000 pounds; while the strain which was at any time liable to come on them was 22,000 pounds per inch, or 3,000 pounds more than the elastic limit. The fracture of the tested rods, which, it is stated, broke with a single blow of the hammer very much in the manner of cast-iron, showed a very inferior quality of metal. The rods broke in the bridge exactly where we should look for the failure; viz., in the screw at the end. No ordinary inspection would have detected this weakness. No inspection did detect it, but a proper specification faithfully carried out would have prevented the disaster.

Look now at an extract from the specification for bridges upon the Cincinnati Southern Railway:—

"All parts of the bridges and trestleworks must be proportioned to sustain the passage of the following rolling-load at a speed of not less than 30 miles an hour: viz., two locomotives coupled, each weighing 36 tons on the drivers in a space of 12 feet, the total weight of each engine and tender loaded being 66 tons in a space of 50 feet, and followed by loaded cars weighing 20 tons each in a space of 22 feet. An addition of 25 per cent will be made to the strains produced by the rolling-load considered as static in all parts which are liable to be thrown suddenly under strain by the passage of a rapidly moving load. A similar addition of 50 per cent will be made to the strain on suspension links and riveted connections of stringers with floor-beams, and floor-beams with trusses. The iron-work shall be so proportioned that the weight of the structure, together with the above specified rolling-load, shall in no part cause a tensile strain of more than 10,000 pounds per square inch of sectional area. Iron used under tensile strain shall be tough, ductile, of uniform quality, and capable of sustaining not less than 50,000 pounds per square inch of sectional area without fracture, and 25,000 pounds per square inch without taking a permanent set. The reduction of area at the breaking-point shall average 25 per cent, and the elongation 15 per cent. When cold, the iron must bend, without sign of fracture, from 90 to 180 degrees."

A specification like this, properly carried out, would put an absolute stop to the building of such structures as the Tariffville Bridge, and would prevent a very large part of the catastrophes which so often shock the community, and shake the public faith in iron bridges. Reference has been made above to the proper loads to be placed upon wrought-iron when under a tensile strain. Similar loads have been determined for other materials, and for other kinds of strain.

The preceding remarks in regard to the loads for which bridges should be designed, and the safe weight to be put upon the material, are given to show how far the safety of a bridge is a matter of fact, and how far a matter of opinion. It will be seen that the limits within which we are at liberty to vary, are quite narrow, so that bridge-building may correctly be called a science; and there is no excuse for the person who guesses, either at the load which a bridge should be designed to bear, or at the size of the different members of the structure. Still less can we excuse the man who not only guesses, but who, in order to build cheaply, persistently guesses on the wrong side.

It will, of course, be understood, when it is said that bridge-building may be called a science, that it can only be so when in the hands of an engineer whose judgment has been matured by wide experience, and who understands that no mechanical philosophy can be applied to practice which is not subject to the contingencies of workmanship. There are many bridges which will stand the test of figures very well, and which are nevertheless very poor structures. The general plan of a bridge may be good, the computations all right, and yet it may break down under the first train that passes over it. There are many practical considerations that cannot be, at any rate have not yet been, reduced to figures. It is not enough that the strains upon each member of a bridge should be correctly estimated, and fall within the safe limits: the different members of the bridge must be so connected, and the mechanical details such, as to insure, under all conditions, the assumed action of the several parts. In fine, while we can say that a bridge that does not stand the test of arithmetic is a bad bridge, we cannot always say that a structure which does stand such a test is a good one.

We often hear it argued that a bridge must be safe, since it has been submitted to a heavy load, and did not break down. Such a test means absolutely nothing. It does not even show that the bridge will bear the same load again, much less does it show that it has the proper margin for safety. It simply shows that it did not break down at that time. Every rotten, worn-out, and defective bridge that ever fell has been submitted to exactly that test. More than this, it has repeatedly happened that a heavy train has passed over a bridge in apparent safety, while a much lighter one passing directly afterwards has gone through. In almost all such cases, the structure has been weak and defective; and finally some heavy load passes over, and cripples the bridge, so that the next load produces a disaster. For the test of a bridge to be in any way satisfactory, we must know just what effect such test has had upon the structure. We do not find this out by simply standing near, and noting that the bridge did not break down. We must satisfy ourselves beyond all question that no part has been overstrained.

A short time ago the builders of a wretchedly cheap and unsafe highway bridge, in order to quiet a fear which had arisen that the structure was not altogether sound, tested a span 122 feet long with a load of 58,000 pounds; and inasmuch as the bridge did not break down under this load, which was less than a quarter part of what it was warranted to carry safely, the county commissioners considered the result eminently satisfactory, and remarked that the test was made merely to satisfy the public that the bridge was abundantly safe for all practical uses. The public would, no doubt, have been satisfied that the Ashtabula bridge was abundantly safe for all practical uses had it stood on that bridge in the morning and seen a heavy freight-train go over it, and yet that very bridge broke down directly afterwards under a passenger-train.

Now, according to the common notion, that was a good bridge in the morning, and a very bad bridge, or rather, no bridge at all, in the evening. The question for the public is, When did it cease to be a good bridge, and begin to be a bad one? A test like the one referred to above can do no more than illustrate the ignorance or lack of honesty of those who make it, or those who are satisfied with it. Such a test might come within a dozen pounds of breaking the bridge down, and no one be the wiser. The entire absurdity of such testing has recently been illustrated in the most decided manner. The very same company that built the bridge above referred to, made also another one on exactly the same plan, and of almost precisely the same size, and tested it when done by placing almost exactly the same load upon it. The bridge did not break down; and the county commissioners, for whom the work was done, were satisfied that it was "abundantly safe for all practical uses," accepted it, paid for it; and in less than ten years it broke down under a single team and a little snow, weighing in all not over one-tenth part of the load the bridge was warranted to carry, and not over one-half the load with which it had been previously tested. If this bridge had been "tested" by five minutes of honest arithmetic, it would have been promptly condemned the very day it was finished.

In view of the preceding, what shall we say of a bridge company that deliberately builds a bridge in the middle of a large town, where it will be subjected to heavy teaming, and, owing to its peculiar location, to heavy crowds, and warrants to the town that it shall safely hold a ton per running-foot, when the very simplest computation shows beyond chance of dispute that such a load will strain the iron to 40,000 pounds per square inch? We are to say, either that such a company is so ignorant that it does not know the difference between a good bridge and a bad one, or else so wicked as to knowingly subject the public to a wretchedly unsafe bridge. The case referred to is not an imaginary one, but existed recently in the main street of a large New-England town. The joints in that bridge, which could safely hold but 20,000 pounds, were required to hold 60,000 pounds under the load which the builders had warranted the bridge to carry safely. The case was so bad, that, after a lengthy controversy, the town officers had a thorough expert examination of the bridge, which promptly condemned it as in imminent danger of falling, and as having a factor of safety of only 1-15/100, which is practically no factor at all. Notwithstanding all this, and in the face of the report, the president of the bridge company came out with a letter in the papers, in which he pronounced the bridge "perfectly safe." Thus we actually have the president of a bridge company in this country stating openly that a factor of safety of 1-15/100 makes a bridge perfectly safe, or, in other words, that a bridge can safely bear the load that will break it down, for he very wisely made not the slightest attempt to disprove any of the conclusions of the commission; and this company has built hundreds of highway bridges all over the United States, and is building them to-day wherever it can find town or county officers ignorant enough to buy them.

It might be supposed, that, under the above condemnation, the authorities controlling the bridge would have taken some steps to prevent the coming disaster. They did, however, nothing of the kind, but allowed the public to travel over it for more than a year, at the most fearful risk, until public indignation became so strong that a special town-meeting was called, and a committee appointed to remove the old bridge, and to build a new one.

One of the worst cases of utterly dishonest bridge-building that we have had of late years in Massachusetts, was that of the iron highway bridge across the Merrimac River at Groveland, a few miles below Haverhill, one span of which broke down in January, 1881. This bridge was built in 1871-1872, and consisted of 6 spans, each about 125 feet long. The whole cost of the structure was $80,000, and the contract price for the iron-work was $28,000. The company which made that bridge, agreed in their contract to give the county a structure that should carry safely 3,000 pounds per running-foot besides its own weight; but they built a bridge, which, if they knew enough to compute its strength at all, they knew perfectly well could not safely carry over one-quarter part of that load. In fact, the weight of the bridge alone is more than it ever ought to have borne. The company warranted each span of that bridge to carry safely a net or moving load of 165 tons, and it broke down under a single team and a small amount of snow. The company warranted that bridge to carry safely a load which would strain the iron to 50,000 pounds per inch, when it knew perfectly well that 15,000 pounds per inch was the most that could safely be borne.

There are several concerns in the United States which make a specialty of highway bridges, and which, taking advantage of the ignorance of public officials, are flooding the country with bridges no better than that at Groveland. On an average, at least twenty of these miserable traps tumble down every year, and nothing is done to bring the guilty parties to punishment. Dishonest builders cheat ignorant officials, and the public suffers the damage and pays the bills. Is human life worth enough to pay for having these structures inspected, and, if found unsafe, strengthened or removed? Can we do any thing to prevent towns and counties from being imposed upon by dishonest builders? We certainly can, if those who control these matters care enough about it to do it. There are two ways of buying a bridge,—a good way and a bad one; and these two ways are so plain that no one can misunderstand. To buy a bad bridge, just as soon as your town or county votes money for a new bridge, certain agents—and they are as numerous as the agents for sewing-machines or lightning-rods—will call on, or write to, the town or county officers, and will offer to build any thing under heavens you want of any size, shape, or material, and for almost any price. They will produce testimonials from all the town and county officers in the country for the excellence of their bridges, and would not hesitate to give reference, even, for their moral character, if you should ask it. If they find that you don't know any thing about bridges, they will, to save you the trouble, furnish a printed specification; which document will commit you to pay the money, but will not commit the bridge company to any thing at all. When the bridge is put up, you never will know whether the iron is good or bad, nor whether the dimensions and proportions are such as to be safe or not. You will know that you have paid your money away, but you never will know what you have got for it until some day when your bridge gets a crowd upon it, and breaks down, and you have the damage to pay. This mode of buying a bridge is very common. To buy a good bridge, first determine precisely what you want; and if you don't know any thing in regard to bridge-building yourself, employ an engineer who does, to make a specification stating exactly what you want, and what you mean to have. Then advertise for bridge-builders to send in plans and proposals. Let the contractors understand that all plans and computations are to be submitted to your engineer, that all materials and workmanship will be submitted to your inspectors, and that the whole structure is to be made subject to the supervision of a competent engineer, and accepted by him for you. You will find at once, that, under such conditions, all travelling agents and builders of cheap bridges will avoid you as a thief does the light of day. You will have genuine proposals from responsible companies, and their bids should be submitted to your engineer. When you have made your choice, let the contract be written by your lawyer, and have the plans and specifications attached. Employ a competent engineer to inspect the work as it goes on; and when it is done, you will have a bridge which will be warranted absolutely sound by the best authority. This mode of buying a bridge is very uncommon.

The Ashtabula bridge, it is stated in the report of the committee of the Ohio Legislature appointed to investigate that disaster, had factors,—we can hardly call them factors of safety,—in some parts as low as 1-6/10 and 1-2/10, such factors referring to the breaking-weight; and even these factors were obtained by assuming the load as at rest, and making no allowance for the jar and shock from a railroad train in motion. Well may the commissioners say, as they do at the end of their report, "The bridge was liable to go down at any time during the last ten years under the loads that might at any time be brought upon it in the ordinary course of the company's business, and it is most remarkable that it did not sooner occur."

One point always brought forward when an iron bridge breaks down, is the supposed deterioration of iron under repeated straining; and we are gravely told that after a while all iron loses its fibre, and becomes crystalline. This is one of the "mysteries" which some persons conjure up at tolerably regular intervals to cover their ignorance. It is perfectly well known by engineers the world over, that with good iron properly used, nothing of the kind ever takes place. This matter used to be a favorite bone of contention among engineers, but it has long since been laid upon the shelf. No engineer at the present day ever thinks of it. We have only to allow the proper margin for safety, as our first-class builders all do, and this antiquated objection at once vanishes. The examples of the long duration of iron in large bridges are numerous and conclusive. The Niagara-Falls railroad suspension bridge was carefully inspected after twenty-five years of continued use under frequent and heavy trains, and not only was it impossible to detect by the severest tests any defect in the wire of the cables, but a piece of it, being thrown upon the floor, curled up, showing the old "kink" which the iron had when it was first made, and wound on the reel. The Menai suspension bridge, in which 1,000 tons of iron have hung suspended across an opening of 600 feet for sixty years, shows no depreciation that the most rigid inspection could detect. Iron rods, recently taken from an old bridge in this country, have been carefully tested after sixty years of use, and found to have lost nothing, either of the original breaking-strength, or of the original elasticity.

The question is frequently asked, Does not extreme cold weaken iron bridges? To this, it may be replied, that no iron bridge, made by a reliable company, has ever shown the slightest indication of any thing of the kind, though they have been used for many years in Russia, Norway, Sweden, and Canada, and nothing that we know in regard to iron gives us any reason to suppose that any thing of the kind ever will happen. But here, again, every thing turns upon the quality of the iron. Iron containing phosphorus is "cold-short," or brittle when cold, and will break quicker under repeated and sudden shocks in cold weather than when it is warm. With good iron, properly used, we need have no fear on this point. The securing such iron is a matter to which the utmost attention is paid by our first-class bridge-building firms, but it is a matter to which no attention is paid by the builders of cheap bridges. We might suppose that a person, in putting an insufficient amount of iron into a bridge, would be careful to get the best quality; but exactly the reverse seems to be the case, on the ground, perhaps, that the less of a bad thing we have, the better.

Many persons, in building wooden bridges, take no pains to get iron rods which are suitable for such work, but purchase what is easiest to be had in the market, and in many cases never find that the iron was bad until a bridge tumbles down. There are, without the slightest question, hundreds of bridges now in use in this country, which, as far as mere proportions and dimensions go, would appear to be entirely safe, but which, on account of the quality of the iron with which they are made, are entirely unsafe; and there always will be, as long as public officials purchase iron which they know nothing about, to put into bridges. When a bridge is finished, the ordinary examinations never detect the quality of the iron; so that the wise remarks of many inspectors, or the opinions of those in charge of these structures, as to the exact condition of a bridge, are of little or no value.

We often hear iron bridges condemned, while wooden ones, so called, are supposed to be free from defects. It does not seem to occur to persons holding such ideas, that wooden bridges rely just as much upon the strength of the iron rods that tie the timbers together, as upon the timber. As a matter of fact, where one iron bridge falls, a dozen wooden ones do the same thing. One very decided advantage which an iron bridge has over a wooden one, is that we can make sure of good iron in the beginning, and that we can also be sure that it does not decay; while, however good our timber may be in the beginning, we never can be entirely sure of its condition afterwards. There are wooden bridges now standing in this country, all the way from sixty to eighty years old, which are apparently as good as ever; while there are others, not ten years old, which are so rotten as to be unfit for use. It will not do to assume, that, because no defects are very evident in a wooden bridge, therefore it has none. When a wooden bridge, originally made of only fair material, has been in use under railroad trains for twenty-five or thirty years, and in a position where timber would naturally decay, we are bound to suspect that bridge. To assume such a bridge to be all right until we can prove it to be all wrong, is not safe. To assume a bridge to be all wrong until we can prove it to be all right, is a safe method, though not a popular one. Any person who has had occasion to remove old wooden bridges, will recall how often they look very much worse than was anticipated.

There is one defect in railway bridges which has often led to the most fearful disasters, and which, without the slightest question, can be almost entirely, if not entirely, removed, and at a moderate cost. At least half the most disastrous failures of railroad bridges in the United States have been due to a defective system of flooring. With a very large proportion of our bridges, the failure of a rail, the breaking of an axle, or any thing which shall throw the train from the track, is almost sure to be followed by the breaking down of the bridge. The cross-ties are in many cases very short, and the floor is proportioned for a train on and not off the rails. When an engine on such a floor leaves the track, it plunges off the ends of the cross-ties into the open space between the stringers and the chords, and generally wrecks the bridge. To prevent this, the cross-ties should be long and well supported, and placed so close that a derailed engine cannot cut through them. The track should also be provided with guard-timbers well fastened, and the width between the trusses should be so great that the wheels of a derailed train will be stopped by the guard-rail before the side of the widest car can strike the truss.

The importance of a substantial floor system has been very fully recognized by the railroad commissioners of Massachusetts, who have recently issued a very suggestive circular, accompanied by numerous examples of track construction for railway bridges. If this circular receives proper attention, it is sure to produce good results.

Another point which has often been neglected, is making sufficient provision to resist the force of the wind. A tornado, such as is not uncommon in this country, will exert a force of 40 pounds per square foot, which upon the side of a wooden bridge, say of 200 feet span, and 25 feet high, and boarded up as many bridges are, would amount to a lateral thrust of no less than 100 tons; and this load would be applied in the worst possible manner, i.e., in a series of shocks. There have been many cases in this country where bridges have been blown down; and a case recently occurred where an iron railroad bridge of 180 feet span, and 30 feet high, and presenting apparently almost no surface to the wind, was blown so much out of line that the track had to be shifted. The recent terrible disaster at the Firth of Tay was, no doubt, due to this cause.

At the time of the Tariffville catastrophe, it was gravely stated at the coroner's inquest, and by railroad officers who claimed to know about such things, that the disaster was caused by the tremendous weight of two locomotives which were coupled together, and it was stated that one engine would have passed in safety; and directly afterwards the superintendent of a prominent railroad in New England issued an order forbidding two engines connected to pass over any iron bridges. It is all very well for a company to issue such an order, so far as it may give the public to understand that it is determined to use every precaution against disaster; but such an order may have the effect of creating a distrust which really ought not to exist. If a railway bridge is not entirely safe for two engines, it is certainly entirely unsafe for one engine and the train following; the only saving in weight by taking off one engine being the difference between the weight of that engine and the weight of the cars that would occupy the same room. For example, a bridge of 200 feet span will weigh 1,500 pounds per lineal foot. An engine and its tender will weigh 60 tons in a length of 50 feet, and a loaded freight-train may easily weigh 2/3 of a ton per lineal foot. The total weight of the span, with two engines, and the rest of the bridge covered with loaded freight-cars, would thus be 320 tons. If we take off one engine, and fill its place with cars, we take off 60 tons, and put in its place 33 tons; i.e., we remove 27 tons, or just about 1/12 of the working-load. Taking off a large part of the working-load, however, is taking off a very small part of the breaking-load; with a factor of safety of six, for example, taking off 1/12 of the working-load is taking off less than 1/70 of the breaking-load. An order, therefore, like that above, can only be of use when the working-load and the breaking-load are so nearly alike that the actual load is a dangerous one: that is when the bridge is unfit for any traffic whatever; so that, if such an order was really needed, it would, in itself, be, in the eyes of an engineer, a condemnation of the bridge.