Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

RE: Gusset Plates in a 3rd Minnesota Bridge (Blatnik Br) are Found to be Faulty

[Subject Prev][Subject Next][Thread Prev][Thread Next]
What exactly do they do about fatigue in bridge design? Fatigue is
accumulative. So do they design for a fixed lifetime (number of cycles) and
monitor the bridge throughout its life. Or do they design for the endurance
limit of the material, assuming it has one. (Aluminium doesn't, steel alloys
generally do.)

My understanding is that for aircraft: components are replaced after
predetermined number of hours of service: or some other measure of life

Also Polymers have high creep rates, as do fibre reinforced polymer
composites like timber. Once again a structure has a predefined life: this
time based on maximum deflection at end of service life. So carbon fibre
patches have a finite life, and so do timber bridges. Concrete I also
believe has relatively high creep rates.

So it would appear that ultimately all bridges have to be removed, and
replaced. A difficult task when the community becomes dependent on the
bridge, and industries once distributed both sides of bridge, are now
concentrated on one side only.

Basically the new bridge has to be available the instant the existing bridge
is closed for demolition. The transportation system itself needs a higher
level of redundancy. The problem is keeping the redundancies as reserves for
emergencies and other special situations.

As with redundancies in structures: they cease to be really redundant
members if indeterminate analysis is performed and the structure becomes
dependent upon. Which is why qualitative aspects of robust design are
important, to determine under what situations the redundancies come into
play, and what situations they are ignored or otherwise considered to
provide additional benefit. Or have negative effects. Like truss designed as
pinned: but fabricated with continuous chords and welded or nail plated, and
has some moment of resistance at the nodes which has traditionally otherwise
been neglected.

Another thing is to avoid getting too analytical(?) precise(?) in design. I
was taught to use distortion energy theory for investigation of failures,
such as occur when testing prototypes. For design use more conservative
theories such as maximum shear stress theory: that way have some reserve up
your sleeve for unexpected variations. But the authors of the structural
codes think otherwise: a problem with the legal nature of the code. But
don't have to design to the code, only check compliance. So can still adopt
more conservative theories for design: but too many who work to letter of
the code.

Also from a mechanical engineering perspective I was taught to use FEM to
reduce the number of prototypes which need to be tested, and test the
prototypes to calibrate the model. Civil engineers seem to take FEM as
accurate: don't build prototypes or proof test materials as Telford did, and
they only build one-off. Roebling also tested his materials. As for Navier:
well he eventually got his bridges to stand up.

Today the quality of materials and of the mathematical theories all seem to
be taken for granted. In building and construction the engineers are
typically consultants in a design office: they do not have laboratories to
check the empirical formulae in codes of practice. If there is a typing
error in a code or paper they cannot check it.

It is also not necessary to have hundreds of engineers calculating the same
thing day after day all over the world. Let those that know and understand
do the rigorous analysis and testing and provide published pre-calculated
solutions, and handbooks of design capacities/resistances. That way there is
less variation in design, and more public/learned criticism of the published
solutions. Leaving it to individual designers and checkers generates too
much variation in interpretation of the intent of codes, and therefore too
much variation in what gets implemented in the field. Standard designs or
pre-engineered solutions provide worked examples for students to work
through, and the students provide a checking mechanism.

Standard designs also provide calibration points when design codes are being
changed. For either the standard designs are considered acceptable and no
change is desired, or the design code is being changed with the intent of
imposing change on the standard designs.

Put simply the design process itself is defective, and licensing engineers
won't fix it. Licensing engineers doesn't adequately reduce variation in the
process of design: it only assesses potential capability. A licensed
engineer may have the potential to get something correct if all other
resources are adequately supplied: but it doesn't mean they will. Innovative
design may pose a large number of issues: not all of which may be recognised
by the design engineer.

The whole lifecycle also needs to be properly managed, from conception,
through detail design, to implementation, operation, maintenance and
eventual disposal. The division of labour between design consultancies,
building contractors, fabricators and the likes, hinders some aspects of
developing the built environment.

Multinational businesses are today the most powerful political entities and
they have no geographical boundaries. They design, make and distribute
products. But there is not exactly a business marching across the planet
building railways and bridges, or roads and bridges. Taxes go into general
revenue and used for what ever government considers important this year. But
if road infrastructure was owned by a business and users paid directly for
such service provided: maybe there would be more pressure to maintain the
system. May be new technologies would develop faster. The other aspect is
standardisation. If own the infrastructure across the globe, there would be
a greater tendency to standardise components: thus bridges would be modular
variations of one another. The bridges would be easy and fast to replace:
the modules would be constantly in production. Ok! The repetition may be a
bit boring: but not all bridges are monumental landmarks. Most are simply
there for function. There is also a difference between initial supply and
maintenance level supply. Initial supply requires standardisation: so Ford
was correct when he said "dam the customer they can have any color they want
as long as it black.". Once the population is supplied with the basic
function however, it then becomes necessary to provide more customised
features: the production technology has to change.

Thus a business that treats bridges as a repetitive product will approach
the supply of bridges differently than those who consider them as one-off
permanent structures. The problem is creating an enterprise which has an
objective beyond monetary profits. A vision or ideal to say: bridge the
world. Centralised planning of systems doesn't have to come from government,
it can come from business. Government is after all just another business
enterprise with special privilege: just another waring tribe in the battle
for scarce resources. Private bureaucracies are just as inefficient as
government bureaucracies: so privatisation is not a solution. If private
enterprises were more efficient, then they would be supplying what
government provides already. That is the real key: designing a product, an
appropriate mix of goods and services, which surpasses what any one else is
providing including government. When we step outside our front doors we are
already paying for the footpath and roads: indirectly we may also be paying
for the air we breath. At present there is no financial constraint which
stops us stepping on the footpath or using the road.

Constraining the use of footpaths and roads maybe a bit extreme, but
constraining the use of bridges may be acceptable. As I said bridges
segregate functions, a hospital may end up only on one side of a bridge. By
constraining use of a bridge, functions are forced to become uniformly
spread on both sides of the bridge. The use of the bridge is limited to
connecting services which are naturally constrained to one or the other side
of the bridge. Bridges are a technological privilege: such privilege seems
to get lost in time. 

A lot of the privilege comes from the benefit of historical precedence.
Ancestors who built systems for the benefit of their young. Systems taken
for granted and poorly maintained. Several generations who have wasted
energy in useless activity at their local keep fit club. Energy that could
have been used swinging axes and picks: or walking to work. All our
technologies are a privilege and should not be taken for granted. They are
also part of a complex artificial system we have built.

With rises in laptop computers, mobile phones, digital computers, video
conferencing, and the internet: are the loads on bridges going to increase
or decrease? Are we going to become more mobile, and increase travel? Is the
increased travel going to involve smaller more fuel efficient cars, or
larger mobile-homes? Or is travel going to decrease and we spend more time
working from home?

The operating environment in which a structure or any product operates
cannot be ignored. There is a two way interaction: the product changes the
environment which leads to new and alternative needs which in turn results
in the product being changed to adapt to the new environment: and so on the
cycle repeats.

That is what quality robust design and adaptive systems are about. The world
is dynamic, variation is never ending, and static solutions are of little

If communities and their governments cannot fund maintenance of bridges, and
private business cannot develop models to fund maintenance of bridges: then
may be the community does not want the bridges. (A business can operate at
zero profit if owners are also operators paid wages.)

Carry out risk analysis. Remove a bridge from the transport network and
check the impact on community and business. The military may also have an
interest. Make that should have an interest, and they should have the
systems to rapidly replace bridges critical to the defence of a nation.

Getting killed when a bridge collapses is a maybe. Putting a toll gate
across a bridge and restricting access, for say a month, is a definite and
immediate impact on the community. Loosing use of the bridge will have more
influence on the community, than maybe getting killed. Simply being in a car
and on the road is a hazard with a high risk of death: people accept the
risk everyday. Inconvenience is what they complain about and take action to

So funding should be more readily available when the impact of loosing the
bridge is better appreciated by the populace.

And you cannot stop adapting. Whatever you do, will be short term, go out
off favour and something else will need to take its place. A failure to keep
up the effort, a lack of people with drive to do so, is largely why
infrastructure has not been maintained and is deteriorating. Succession
planning is important to replace the person with drive, the
leader/visionary, to keep things going. Well replace anyone and anything for
that matter, in our technologically dependent society.

It is not just the bridge infrastructure that has not been maintained, there
as also been a failure to maintain the personnel and enterprises capable of
doing so. The construction industry really does need better management, more
vision and better foresight.

Knowledge should not be restricted to a limited few. The original
universities provided a universal education and the graduates went out into
the world as scholars who taught and passed knowledge on. Professional
constraint of knowledge is unhelpful: people need knowledge to aid their
understanding of decisions made, which impact their lives. Improved
education, information and communication systems are also needed. Access to
the internet or a public library, has both negative and positive impacts on
the ability to sell an idea. Not the least of which an idea can grow
extremely complex very quickly, and information overload can stifle the
ability to make a decision. It can also create fear and unnecessary anxiety
about things so remote and unlikely to happen it's not worth worrying about:
or for events for which there is little we can do. Then have to learn how to
make decisions and take action when the information is unclear: or the
benefits and associated detriments are difficult to choose between. The
community usually only wants the benefits and won't accept any of the
associated detriments.

Design really is complex. Systems don't stand still, static design-solutions
will become obsolete. I repeat we really need to be more capable of
adapting: with far faster response times.

The steel bridges are there, and their available resistance maybe a matter
of science: but their required resistance is a subjective opinion: and that
is a matter of politics and economics.

The politics which put the design-loads in the codes of practice can equally
well reduce the design-loads. If the design-loads are decreased then have a
different class or grade of structure: an alternative product with differing
uses. That differing use may impose traffic controls, which limit the flow
of traffic over the bridge. Traffic lights or toll gates can be put near a
bridge which restrict traffic flow: so that there is less traffic on the
bridge at any point in time. That will cause inconvenience and possibly give
rise to demands to build a new alternative bridge. Or otherwise change the
location of businesses previously dependent on the bridge. And with
different grades of bridges demanding differing levels of restricted access,
the community can decide what it can afford and what restrictions it can

Hundreds of possibilities if design first, and look at the codes of practice
last. Just need Columbus to stand the egg on its end by crushing the end.
You didn't say you could do that! I didn't say you couldn't either!

Conrad Harrison
B.Tech (mfg & mech), MIIE, gradTIEAust
South Australia

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*   Questions to seaint-ad(--nospam--at) Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: 
******* ****** ****** ****** ******* ****** ****** ********