Soil springs and other fables
In our regular interaction with our good friends in the structural engineering world, we are often asked to provide springs. Obviously, we're not talking about literal springs here - like those in your car's suspension. Rather, we are referring to a numerical value that generally describes a Hookean relationship between a force applied to the ground surface and the resulting displacement. These springs are often used in structural analysis packages as a simplified means to incorporate the effects of foundation stiffness on a structural system.
In an ideal world, there would be no need to provide spring values. Rather, all soil-structure interaction problems would be modelled holistically, using numerical methods that explicitly capture the continuum aspects of geo-materials, through the use of appropriate constitutive models.
In fact, this is a reality now. There are numerous commercial software packages that can solve complex boundary value problems in continuum mechanics. These packages often come loaded with many types of elements for modelling structural systems (e.g. beams, plates, shells, solids, membranes, etc.) along with numerous constitutive models to represent the behaviour of geo-materials. The software packages often provide pre- and post-processing facilities to develop the model and analyse the results. And at the heart of the software is the means in which the governing system of PDEs are discretised and solved over the analysis domain. The most well-known solution methods typically being either the finite element or finite difference approach.
However, in our experience this approach is still not the norm for many design consultancies, because:
It can be difficult to find engineering professionals that have the necessary depth of understanding of the theoretical and practical aspects of reliably carrying out advanced analysis involving structural and geotechnical systems as a whole. Let alone the need to have another few in any one organisation to enable an adequate level peer review to be carried out. Wisely, a number of consultancies continue to use their tried and tested design processes - at least until such time that the practical use of advanced analyses matures within the field.
There is still a relatively high price point to acquire licenses for geo-structural software packages that can perform these types of advanced analyses. This is particularly the case when full 3D modelling is required. For many small-to-medium sized consultancies, the business case to invest in this software just does not stack up, yet.
Many geo-structural software packages do not come natively equipped with tools that are relied upon by structural engineers to improve productivity and quality control, e.g. automatically performing design checks of structural elements against a specified design standard.
Often times, geotechnical design considerations are contractually separated from the structural engineer's responsibilities. So the opportunity for a collaborative design effort is reduced. Geotechnical design information may be provided in the form of a geotechincal design report that is submitted to the structural engineer at a fixed point in time. Beyond this point, the geotechnical design information can sometimes enter into stasis.
So the use of equivalent springs to represent soil behaviour in structural design remains. However, springs are no panacea and sometimes they can lead you down the rabbit hole. From our experience, there are a number of subtleties and edge-cases that one needs to be wary of when using springs. Questions that we are often asked, include:
Is the use of springs at least a reasonable approximation for the structure being considered?
Is it possible to ignore the foundation effect all together?
How might construction sequencing affect spring values?
What about design in heaving soil mass or where there are buoyancy forces. Can i have negative springs?!
Can I use the same spring for different geometries or arrangements of foundations/footings?
What are the simplifications of real soil behaviour that springs do not capture?
Do my spring values vary with time? What about strain level?
The list goes on...
In light of this we have developed and delivered on a number of occasions to our partners, a brief technical presentation (30-45 mins) on these (and other items) relating to the use of springs in design practice. If you would like to organise a presentation for your consultancy please contact us and we would be happy to introduce ourselves to your team.