Determining the properties of members of historic structures

Question

From time to time I am tasked with additions or alterations to old buildings or structures. I would need to establish the "as built" structure and from that analysis see how the amended/new structure suits the existing structural components.

The challenge lies in the assumptions one is forced to make. Over the years things change, so do the material properties and standard member sizes. For example, I have come across a bombed bridge that had to be rehabilitated in Angola. The bridge was built by the Portuguese government some time before the Angolan war. The bridge was constructed from large steel I-beams as main bearers. We could not establish the type of steel used for the I-beams, and the sectional properties we established by measuring the beams with a tape measure.

How do engineers determine the material properties of members of old or historic structures? I am not aware of any document or database that captures all steel (or rebar) manufacturers of the world with their respective product lists and material properties over time, or what design standards were historically applicable at various times in various regions of the world.

Answer 1

I have never encountered a resource which compiled a worldwide database of historic structural data. But individual countries do have some resources.

In the limited work I have done with historic structures (in the UK) I have used these resources:

The Historical structural steelwork handbook contains material and section properties for UK and European cast iron, wrought iron and steel sections since the mid 19th century. It also includes references to historic building codes as well as loading and other design information.

and

TR70 Historical approaches to the design of concrete buildings and structures which contains historic reinforcement and concrete mix properties as well as historical design guidance.

However, additional research is usually required. National libraries often carry historic copies of building regulations and other design guidance. Institutions like the Institution of Structural Engineers and the Institution of Civil engineers in the UK often have historical construction information.

I would imagine that outside of Western Europe and North America finding similar guidance would be very challenging or impossible.

This paper may be helpful to get started doing historical research: Chrimes, M. Historical research: a guide for civil engineers, Civil Engineering, Proceedings of the Institution of Civil Engineers, Volume 159, Issue 1, February 2006, pp. 42–47.

Answer 2

My knowledge of historical work only extends as far as the borders of the United States, unfortunately.

However, for US work, the American Institute of Steel Construction (AISC) does publish a historical shapes database showing the different sizes used over the past 80 years or so and some of the manufacturers that made steel members. If you can take measurements of the beam's dimensions, it's possible that you can back-figure what size it's supposed to be. Given that, if the shape isn't made anymore, you might be able to determine around when it was made and correlate that to the common material specifications of the time.

Ultimately, if you don't have access to the design drawings or specifications used to construct the bridge and you absolutely have to know the material type, I don't see how you can get around taking a sample and having it tested in a lab. For something as critical to human safety as a bridge, I personally think it's a justifiable cost, especially considering some of the bridge failures that have occurred in the last decade or so.

Answer 3

When it comes to concrete and steel, there were few standards for materials and design prior to about 1920, and you may not find any prior to 1905, at least that is my experience. Reinforced concrete technology was largely proprietary prior to about 1910 - as such the first multi-story reinforced concrete building in the UK (Weaver's Warehouse 1897 to 1906 I believe) was built under license.

The Civil Engineer's Handbook in the UK can be a good guide for what was going on in the early days (UK and Commonwealth), and the Institution of Civil Engineers library has older copies of this. For example, the water to cement ratio was found in the 1923 version of the Handbook in the UK for the first time (known as the Abrams Ratio and was expressed as volume of cement to volume of water rather than today's w/c m/m ratio).

The design of the Weaver's Warehouse indicated that the use of reinforcement at that time resulted in loads being supported by an 'arching' effect rather than the 'ductility' design approach of present. This means that older structures could be at greater risk of brittle failure modes. I suspect that this is because there was little knowledge of embedment length, etc prior to the 1930's, and steel was often 'fixed' at the ends. The proprietary construction process of that project required that the slabs/beams be load tested as the structure was built.

From memory the first codes for steel and cement (eg BS 12) in the UK were released in the early 1900's, around about 1906 or so, I don't have the exact dates. Cements were far more coarsely ground in those days, and chemically very much different. There was little C3A and gypsum added, for example, prior to about 1920, as it was introduced at a later date when manufacturers found it could reduce the energy and cost of production. Furthermore, because little was known about the bonding of reinforcing steel and embedment length, it was often smooth, not deformed, prior to 1920.

On the other side of the Atlantic, the American Concrete Institute was founded around 1904, and the first building regulations were adopted around 1910 with a number of standards in place around 1912. The Portland Cement Assocation, with Duff Abrams later as its first Chief Engineer, was founded around 1917. Abrams was to originate the ACI concrete mix design method, the concept of water to cement ratio and reinforcing steel bonding capacity in the US.

What I have found in the US and Europe on structures dating back to about 1910 (but mostly during the 1920's) is that you can find just about anything, and testing is strongly advised. It is not unusual to find concrete typically designed and produced on a loose volume basis (ie by scoop or shovel 1:2:4 and no water to cement ratio). One project we looked at, an older bridge foundation and deck, had concrete mixture composition that varied from one end of the bridge to the other: the aggregates were a rounded river gravel on one end and the other end had a crushed granite, and they blended together as they reached the centre.

Another significant issue is the foundation design and construction. You can find anything. I have seen structures carrying load successfully for 50 years or more but with timber piles that were located only half under the cap (ie extending beyond the edges). So again there is no substitute for investigation and/or testing.

For concrete there is an international standard BS EN 13791 that includes statistical strength assessment for structures where no history of the structure is known. This can be of use for comparing the concrete to modern structural codes. But if there is to be a change of loading condition then there is no substitute for identifying cover depths and steel/section geometry using non-destructive tests. And of course reinforcing steel corrosion can be a significant factor.

For structural steel a critical issue is often the bolts/welds as well as the sections. The assessment is very much of an art here, and may require radiographic/ultrasonic/torque testing if the situation demands it. Very old steel structures are often riveted which introduces further complexity. As far as I am aware, there aren't any historical guides on bolts and rivets. Previous posts have given some references that could be of use for the sections. If in doubt about a steel structure, you can obtain coupons from the sections and have them tested for tensile capacity and ductility.