21 October 2021
Ten things to know about engineering timber frames
Chithurst Buddhist Monastery, West Sussex. Photo: Duncan McNicol.
Steve Corbett lists the ways in which designers and engineers can work together to create successful projects.
In the Middle Ages, cathedral builders acted as their own engineers, and the master mason and master carpenter would be wholly responsible for the execution of the build. They learned from their mistakes and we have the benefit of learning from their success – the failures didn’t survive.
The lantern tower at Ely Cathedral, completed in 1334, is one of the greatest engineering feats of its time.
The structural design principles for such historic timber frames clearly pre-date modern building regulations. A tried and tested design can be proved on the basis of precedent, and a Certificate of Approval from a specialist engineer is perfectly acceptable. This method is widely used in the design of such structures, such as this crown post roof for a Buddhist monastery.
Tradition aside, the past 20 years has seen a welcome return of timber being used in modern building design and this is accelerating. Engineered timber allows large spans, and an infinite variety of forms, for both public buildings and domestic architecture.
It is promising to see that there are increasing numbers of structural engineers and architects who specialise in working with timber and encourage its use.
This article explores the ways in which timber designers and engineers can work together by outlining the main factors in that relationship. Most successful projects are forged by teamwork and collaboration.
1. The fundamentals
There are three principles that inform a successful project: design, structure and buildability. If the architect and engineer speak for the first two, the timber designer can offer other all-important elements:
- value for money
- eliminating redundancy
- making it buildable.
Now that most designers use 3D modelling it is easy to forget that building in this way has limitations not always evident on a screen. It is then that the timber frame designer can help.
It is not uncommon for the carpentry team to make suggestions that depart from the specification of the project engineer. However, engineers (or more correctly their professional indemnity insurers) understandably don’t like sharing design responsibility. It is therefore important to draw a line between the two disciplines and agree who does what. Typically, the carpentry team would be subcontracted to design the timber element of the building and engage their own engineer to justify it.
Structural timber frame design is usually a straightforward process – it is the interface with the rest of the building that causes the most discussion, for example the specification for the building envelope. Although not in the remit of the structural team, it affects the design and performance of the structure, and engineers must allow for the dynamic loads. In addition to structural calculations, engineers need to check serviceability – the effects of creep, deflection and movement must be accounted for if they create the possibility of damage to surface materials or finishes.
In terms of specialist timber engineering, a two-phase approach works well. The first stage begins early in the process, when the design intent is known but no details are developed. An early meeting should establish the general structural principles, along with buildability and affordability.
‘Intuitive engineering’ means you have an outline sketch, an idea of the form, scale and content of the structure, and a conversation that concludes, ‘Yes, that works’. The team can then move forward, get the consultants properly appointed and begin the second stage. If the conversation concludes that it does not work, it is time to rethink.
It is important to forge a relationship with the consultant engineer at an early stage. The engineer might come up with specifications that demand a certain performance from the material, but the carpenters’ experience may tell them it would be unachievable. In this case, a series of conversations can quickly narrow down the options and come up with a workable specification. Solid timber is anisotropic in its nature, which is to say that it behaves differently in its three planes: radially, tangentially and along the grain. An engineer may choose the option not to use solid timber but prefer to specify engineered timber because this is more predictable. However, nothing should be ruled out when selecting the right material.
When supplying a batch of solid timber, particularly of unusual specification, any good timber merchant should always ask what the material is to be used for. The end use of the material is critical if the right species, grade and quality is to be offered, and feedback from the producer is invaluable. It is tempting for the architect or engineer to specify a high grade of timber, just because such grades are listed. However, it might not be suitable, or even available, and it takes experience from the carpenters to manage the procurement. Timber selection is a certain skill that can make or break the success of a project. It may be considered a poor use of resources to specify the highest grade of timber where a lower grade would do perfectly well.
There is more to grading and specifying timber than the table of mechanical properties used by the structural engineer, and the carpentry team can inform this process. They know that timber conversion has a big effect on strength, durability and movement, and can predict the likely shrinkage and movement of each component. The rate of growth can be specified, but in the mainstream this rarely happens. Rate of growth is influenced by the geographical origin of the timber, but also by forestry practice, length of the felling cycle and local terroir. Softwood with narrow growth rings is stronger than softwood of the same species with wide apart rings, but, counter-intuitively, the opposite can be expected for structural grade temperate hardwoods, such as oak.
As original and exciting structures are developed, the input from the structural engineer is critical. With long spans, or unusual shapes and new engineered wood products, there is a great demand for ingenious solutions. Some projects might require a purpose-made connector, for example, and it is particularly satisfying if the connection is invisible, although there is still a place for the display of visible fixings so that the load paths and the building’s function are easily interpreted. As another example, while checking for adequacy for a 30-minute fire rating, the simple device of upsizing the sections will allow for charring and ensure the residual section is structurally adequate.
9. The golden triangle
This is a term used for three aspects of any project that need scrutiny: cost, time and quality. Two of these can usually be achieved, but sometimes the third will suffer. It might be quick and cheap, but of low quality. It might be a sensible programme and specification, but at higher cost. To balance all three is a challenge, and the structural engineer has a key role. Keeping abreast of new engineered products and methods of construction, combined with a bit of lateral thinking, can make a big contribution. Main contractors can help by encouraging teamwork and avoiding delays. The client can also help by being clear about what they want, decisive when necessary and patient. Good buildings need good clients.
10. Case study: St Edward’s School
This project is an excellent example of engineer and designer collaboration: a new three-storey academic centre, dressed in stone, surrounded by the Victorian architecture of an independent school in Oxford. Designed by TSH Architects, with main contractor Gilbert-Ash, the top floor is the new reading room. The structure comprises a series of 19 identical portal frames and a diagrid roof plane with diagonal members. The roof pitch is steep (50°), to mirror the surrounding architecture, and the walls and roof are lined with veneered plywood panels. A distinctive feature is that the glulam columns and beams are all in oak as opposed to the more widely specified whitewood. Much attention was given to the lighting design, supplemented by a clerestory of roof lights.
One challenge was the requirement that no fixings should be visible from inside the building and another was the narrow section (120mm), limiting the size of connectors to maintain permissible edge distance. To achieve a moment connection at the column base was straightforward, using bonded rods, but connections between columns and beams, and at the apex, were another matter.
A joint was devised using a solid billet of milled steel, with a tight 1mm tolerance, to act in shear, creating stiffness at the ‘knuckle’, likewise at the ridge. Consultant engineers at Andrew Waring Associates proved that a single M20 bolt on each side was adequate, provided that the threaded rod was secured with a 20mm-thick plate washer.
Another innovation was the use of tapered dovetail housings for connecting the primary roof members, instead of a steel connection plate, saving the cost of more than 250 connectors. The narrow section limited the housing depth, but this was deemed adequate, making use of the fact that these were in oak and not a softer spruce.
About the author
Steve Corbett, Senior Design Manager, Green Oak Carpentry Company.
This is an extract from the Timber 2021 Industry Yearbook. Download the full article, with further reading and images, here
24 November 2021
Southside Hereford: University Design Challenge 2022
25 November 2021
Global timber conference marks beginning of new era for the industry
24 November 2021
Mass Timber Office Forum
24 November 2021