Holdfast Training Services Trial Blog

I never thought I’d say this, but I really do long for Eurocodes.  Whilst the plethora of EC books for design exercises in Phase 1 may have seemed confusing, whilst being in the USACE office I now realise just how thorough and user friendly they all were.  I can almost picture (longingly) the flow charts we used for steel design.

Instead, I am presently hamstrung by a Steel  Construction and Design Manual (SCDM) that is 3 inches thick and literally made of cigarette paper.  Not only is there no easy flow to the manual and its contents, but it sits in isolation with no sponsored design examples. Instead there are a plethora of Design Guides that are not up to date, therefore do not reference the SCDM accurately.  In some case, such as one of my projects (a supporting structure design for overhead cranes), US structural engineers use a Canadian design guide that is more thorough…and to add injury to insult, it uses metric units rather than imperial! Added to that is the choice of which design philosophy to use, and the one you have decided to use may not be the one that the Design Guide explains and/or uses in its design examples.  That said, though my design calculations seem to take an inordinate amount of time for me to consistently detangle my knickers, it has forced me to go back to first principles!

Two design philosophies exist in the USA: Allowable Strength Design (ASD) and Load and Resistance Factor Design (LRFD).  Both are strength based philosophies, though ASD was historically stress based. My first question was: What on earth is the difference and which one do I use?

1.  When considering the steel yield vs displacement graph, the combined force levels (i.e. load, moment  shear) for ASD design are kept below Fy, by taking the nominal strength Fu and dividing it by a factor of safety (aka permissible stress design).  For LRFD, the combined force levels are kept below a ‘computed’ member load capacity which is a product of Fu multiplied by a resistance factor (aka limit state design ~ Eurocodes!).

2. ASD treats live and dead loads equally, thus uses one FOS for both live and dead loads – this accounts for uncertainty in load and capacity.  Consequently it is simpler to use and more conservative.  LRFD however recognises the inconsistencies of dead and live loads thus allocates a higher FOS for live loads than dead loads as dead loads are believed to be more accurately calculated….  LRFD also recognises inconsistencies in material properties and construction tolerances.

So, though LRFD is proven to be a more ‘efficient’ approach in that it harnesses more of the strength capacity of a member, legally we can use either method for steel frame design.  ASD has historically been significantly quicker than LRFD for preliminary design although recent editions of SCDM have become more thorough for ASD.  I honestly believe that having uses ASD, it appears quicker than LRFD (not that I have done the latter yet) but it is still far slower than EC….largely due to the complexity of the Design ManuaI (and because imperial units are driving me insane!)  And that is the reason that every engineer has told me why old-school US structural engineers are steering well clear of adopting LRFD (not the units piece).

The same goes for timber and masonry design; concrete is the only one where LRFD is mandated (by the ACI).  But to add even more confusion, USACE has mandated LRFD for certain design work such as hydraulic steel structures. Then, I have also found, that product catalogues will vary in what methodology they have used for their allowable loads – this makes things incredibly tedious.  All in all a disjointed approach across all  the institutions!