Holdfast Training Services Trial Blog

Last week I was lucky enough to be sidled onto a 2-day course on Progressive Collapse Mitigation run for USACE structural engineers, organised by American Society of Civil Engineers and delivered by a Californian structural engineer – Jesse Karns. The US defence has spent extensive amounts of money in R&D on this topic, and has encompassed its findings and design approaches for all federal buildings in 2009’s UFC 4-023-03  (Jesse was a major contributor in all these areas).  Whilst the UFC is not code, it is a guideline requirement for all US federal construction greater than or equal to three ‘occupied’ storeys.

The US approach had admittedly been shaped by British Standards, so much so that the 2005 version of the UFC was almost a verbatim copy of the BS.  However, since then, the US government has invested a lot more money into R&D in this area…admittedly due to force protection risk assessments, combined with increased R&D into seismic issues.  I was tickled by a comment in the essay distributed by our very own Atkins secondee:  “The UK is currently seen as the centre of engineering excellence around the globe”…really?.  Well, after this course and a limited time in the US, I beg to differ.  The quote needs some justification, for example –  the UK certainly cannot profess to be a centre of excellence on levee design and flood mitigation when Holland sits under sea level with its livelihood relying on flood defences, and the US has over 100,000 miles of levees.  Nor can it profess to be on a world stage with regards to seismic design when there is limited risk to design for.  Off the back of this, I though I’d briefly blog on the US’s present approach to progressive collapse – where they clearly believe they are the centre of excellence, or is this just engineering arrogance…a bit like why it took 1995’s Oklahoma City bombing and the 2001 World Trade Centre Collapse for the US to really rethink the how rigorous its codes were, despite UK’s lessons from the Fallon Point disaster back in 1968. I digress…

The 2009 version is viewed as a pioneering document that leaves BS and Eurocodes wallowing in design assumptions that are based on weak and outdated research (albeit much of the design methodology has the same basis).  I though I’d note the following comments that were in the handout alongside the odd wry smile in my direction …’the mechanics of the methodology are much better defined than British Standards’, ‘the past UFC was based on British Standards and were not too bad for RC (some flaws nonetheless), and really bad for steel’.

So, here’s the lowdown, with some initial figures showing the R&D testing set-up for column removal:

Test model setup for column removal as a result of blast action

Actual testing post blast (note splayed column)

Design Requirement.  As said, the UFC applies to buildings with three of more ‘occupied’ storeys.  The design requirements are dictated by an occupancy classification, from 1 to 4 which is relative to consequent impact on loss of life.  ‘1’ being structures such as storage or agricultural facilities, and ‘4’ being hospitals, emergency shelters, aircraft control towers.  Three design requirements exist:

• Enhanced Local Resistance (ELR).  In outline, this is where the shear and flexural capacity of perimeter walls and columns are bolstered for additional protection.  It is done to increase the capacity of corner and penultimate columns to resist potential increased lateral forces from tie forces (see below), decrease the possibility that two columns will be removed in an initiating event, and forces a ductile flexural response by limiting shear failure.
• Tie Forces.  This British philosophy prescribes tensile forces that hold primary members together.  Ties are provided at across each floor level:  horizontally (internal and peripheral)  and vertically (at columns and load-bearing walls).  The theory is based on catenary actions, where internal tie forces are related to beam vertical beam displacement, load and original span length.

• Alternate Load Path.  Software design that allows for a localised failure but requires that alternate load paths be available to distribute loads to other undamaged parts of the structure.

The occupancy classification dictates which ones to use e.g. Class 1 – no specific design requirement, Class 4 – Tie Forces, ELR for all ground floor columns or walls, Alternate Load path for specified column and load bearing wall removal

A notable eyebrow twitch came in the results of connection testing.  Structural engineers on the west coast of the USA will always design beam-column connections, with detailed drawings.  Everywhere east of the Rockies, we just stipulate moments, forces and directions and let the fabricator ‘crack-on’.  Why the difference? Well, primarily because of the West Coast’s prevalence of seismic events and the knowledge of how buildings collapse…it has been found in structural studies post 1994’s Northridge earthquake that connection failures have consistently been the cause for progressive collapse.  The below graph from tests between 2004-07 identifies the displacement of a central connection directly above a column that has been removed.

Conscious that we want to harness a beam’s flexural and plastic capacity when considering progressive collapse resistance, allowing it to go into the plastic realm between Fy and Fu (and not beyond), the specific design of connections by an engineer should be essential when considering progressive collapse……