Many careful experiments made by Sir W. Fairbairn and others have led to the conclusion that a load may be applied to a wrought iron bar, removed and reimposed thousands of times without the slightest injury to the bar, so long as the stress per square inch does not exceed the elastic limit of the material.

Directly this limit is exceeded, the first application of the load produces a permanent set; each repeated application increases that set, until at last rupture takes place.

The failure of iron under repeated loads or blows of this kind is known as the fatigue of iron.

It will be useful to notice one or two other points connected with the elastic limit. In wrought iron, steel, and indeed in most other building materials, the temporary elongations produced before the limit of elasticity is reached are proportional to the loads which produce those elongations.

Thus, in the bar above referred to, if a load of 1 ton produce an elongation of 1/12000 in the length, 2 tons will produce 2/12000 , 3 tons 3/12000 and so on, until 12 tons produce an elongation of 12/12000 = 1/1000 the length.

At this point, however, the permanent set occurs, and beyond it the elongations are not in proportion to the load, but increase more rapidly than the loads increase. Thus 13 tons will produce more than 13/12000 elongation, and so on.

In cast iron, however, the temporary elongations caused, even by small loads, are from the first irregular, not in proportion to those loads, and an appreciable set is noticed at a very early stage.

False Permanent Set

In some cases, after imposing upon a bar a load far within the elastic limit, a permanent set seems at first to have been caused, but upon leaving the bar unloaded for a short time this set disappears, and the bar slowly returns to its original length.

Set Caused By Continued Load

It has been found that a load within the elastic limit, which will not cause a permanent set if imposed and quickly taken off, will nevertheless cause a set if it be allowed to remain for a considerable time.

To put it in another way, the elastic limit is lower for a continued stress than for a temporary one.

Elastic Limit Raised By Different Processes

It has been shown that the processes of hammering, rolling, and drawing iron or steel, when cold, into bars or wire, increase the tenacity and the elasticity of the material.

Elastic Limit Raised By Stretching

Again, it has been shown that -when a bar of iron has been subjected to a load less than the elastic limit, and continued for several hours, so that a permanent set ensues, the elastic limit of the bar thus altered is considerably raised. For example, General Uchatius tested a bar of soft steel, and found the following results: - l

Limiting Stress. Tons per square inch.

Ultimate Elongation per cent.



Bar of soft steel ...




Same loaded for 24 hours so as to elongate 3.3 per cent ••••••




Same oil-hardened.....




Other Definitions Of The Limit Of Elasticity

It should here be mentioned that Mr. Eaton Hodgkinson's experiments led him to the conclusion that the very smallest load produces a permanent set. His conclusions have been questioned by more recent investigators, but even supposing they are correct, they do not affect the engineer. The permanent sets, if any, produced by loads less than the limit of elasticity are so small that they cannot be measured by an ordinary instrument - in fact, they are inappreciable.

When such loads are constantly repeated, though they may produce an inappreciable set as regards the original length of the bar, yet it is net an increasing set, does not lead to rupture, and may therefore practically be ignored.

When, however, the load is greater than the limit of elasticity, an increasing set takes place upon each application, which eventually leads to rupture.