Use of Computers PDF
Use of Computers PDF
T
he aim of this article is to consider
computer use in the design of con- on practical design methods and but that the engineer needs to be sure that
crete structures and how this is approximate solutions – too much the package being used takes into account
changing the design process. It is often emphasis on highly theoretical, and all the parameters required in the code.
argued that this change is not for the necessarily computer-driven, solutions
• a belief that software calculations must Column fixity
good of design. Problems may arise
when trying to produce reinforced con- be correct and that hand calculations Some slab packages assume pins at col-
crete designs from computer output. The can only be crude approximations umn locations; this is normally accept-
article contains a checklist of items of • increasing complexity of modern codes able for slab flexural design. However, if
which the engineer should be aware deterring engineers from hand calcula- bending moments are developed, they
before relying on the output received. tions as they believe they can no longer will affect both punching shear design
Simple methods for checking the validity provide the complete answer. and column design. If the column-pinned
of computer output are highlighted and Over-reliance on computer output is assumption is made, a separate frame
the need for a body to check proprietary not only a concrete issue and has been analysis may be required to calculate col-
software is discussed. addressed in other publications. The umn moments.
Institution of Structural Engineers pub-
The current situation lished guidelines on the use of computers Boundary conditions
Today’s young engineers often appear last year(1). This report was, in part, a Owing to the inherent continuity of rein-
unwilling to undertake analyses by hand. response to concerns raised by the forced concrete construction, elemental
Whether an approximate calculation to Standing Committee on Structural Safety design packages often require restraint
confirm the viability of a concept or the (SCOSS)(2). conditions to be specified at the boundary
final design analysis, the preferred Design consultants have acknowl- edges. The forces generated then need to
approach is to use a computer. In itself edged the problem. For example, Buro be transferred to the restraining member.
this is not necessarily a bad thing and Happold has a policy where a hand calcu- The element design package will not
computers can be valuable in understand- lation must be made before the computer check the validity of these boundary con-
ing behaviour. For example, by varying analyses is started and the computer ditions and the engineer must do so. For
different parameters, the sensitivity of a model is accepted only if it corresponds example, a beam framing into a thin wall
solution to a number of variables can be to the hand calculation within defined may be closer to pinned than fixed. Many
quickly assessed. However, if the reliance limits. Arup has introduced a one-day frame programs overestimate the mom-
is such that the engineer loses the confi- course, Design without the computer, ents transferred between flat slabs and
dence to carry out simpler methods of where the objective is to teach young boundary columns. Inadequate consider-
analysis, the ability to carry out an inde- engineers how approximate answers can ation of these factors could lead to crack-
pendent check of their model is compro- be quickly calculated to check more ing in the wall, under-design of the beam
mised. In addition, in the search for an detailed analyses. Both companies have and, in the case of flat slabs, cracking and
exact answer, the nature of reinforced internal handbooks giving approximate additional deflection.
concrete construction is forgotten. The solutions and schematic element sizes.
Torsion
material is not elastic, stresses are induced
Reinforced concrete design When a structure is not dependent on tor-
through many mechanisms other than
load (shrinkage, thermal effects, creep) All these points are equally applicable to sional resistance for equilibrium, most
and the process of sensible rationalisation other structural materials. However, there codes say that torsion can be ignored.
and detailing often negates the benefits of are a number of issues specific to rein- However, if torsional stiffness is present
any perceived additional accuracy. forced concrete. It is suggested that the in a computer model, the equilibrium
There are numerous factors behind this engineer should understand how the pro- found will rely on torsion and the tor-
situation, some of which are listed below. posed computer program deals with each sional stresses developed should be
It is important to realise that all parts of the of these issues before analysis is under- designed for. Some packages deal with
industry have played their part: taken. torsions in the post-processing of results,
• the ready availability of software in the and some assume that the torsional resist-
Shear strength ance of all elements is zero. Others will
office, in universities, and free from the
internet Shear design is basically empirical and leave it to the engineer to take into
• lack of education in approximate meth- requires additional rules to define the account. Again, the engineer must under-
ods, some universities placing too boundaries under which code values are stand the assumption implicit in their
much emphasis on obtaining the exact applicable. These vary from code to code design and the computer package being
answer and can be interpreted differently by soft- used.