Need a book? Engineering books recommendations...

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

Design Aids? ( RE: Steel slides for quick reference)

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Christopher Wright said:

<quote>
Don't mistake reference materials for calculation aids. ...

Anyone who's interested in graphic aids should research the lost art  
of nomography--how to make alignment charts to solve equations You  
can still find articles on how to design them <http://mathforum.org/ 
library/drmath/view/63338.html>. ...

</quote>

Sometimes the distinction between calculation aids and reference materials
becomes blurred. Section property tables save having to calculate section
properties. Design capacity tables save having to calculate member
resistances. A member capacity curve even more useful. An action-effect
curve overlaid on several member capacity curves, the most useful. (eg.
bending moment versus span.)

Nomographs must be making a come back. The Structural Engineer (IStructE UK)
recently published an article on the lost art.

I use the computer for just about everything, but I prefer calculations for
ranges and limits, rather than single point solutions. I then print out
tables and charts, which I can use for rapid sizing whilst drafting or
talking to clients.

In SA anyone can prepare and submit documents for development approval, and
usually do. When they receive the 4 page request for further information
that is usually when engineers are called in. Item 1 is usually no
prescriptive solution available, and item 2 provide structural calculations.
The rest of the items are usually an indication that an architect should be
employed. The cold-formed shed industry in particular does this. They have
standard calculations for one or two shed designs: or rather gable frames.
If the customers shed is within the envelope of the standard gable frame,
then approval is usually granted. It seems no one has ever thought to draw
all the envelopes on a single chart. The result is the industry guesses, and
most of the time the customers frame is outside the available frame
envelopes, or they cut the columns out. Consulting engineers are then called
in and placed under pressure to try and make the guess work. A few successes
and they think you can make anything work. So it is necessary to put a
constraint on their guessing. (the standard calculations are also just
random)

My first attempt was to provide a simple graph of height versus span, for
each cold-formed section already assessed, and based on a rough rule: namely
that the height and span of the gable frame could change on condition the
perimeter remained constant. I presented the graph to one manufacturer and
found they had a totally strange way of reading graphs and scrapped that
idea. Using Kleinlogel formula in Excel I was able to move away from rough
estimates. So for a single frame span I produced tables of heights and
sections required: my height increments were 300mm. The manufacturers
started arguing about using smaller size section because they were only
100mm over. So I plotted maximum moment against height, then over laid the
horizontal bands of sectional moment capacity, and showed that they had to
use the larger section. I then went to the next stage, instead of manually
changing height and span in Excel, I programmed iterative loops in VBA, and
produced height versus span limit charts for all 18 cold-formed c-sections
available: effectively iso-moment curves. I plotted the chart in AutoCAD so
that they could draw the shed to scale, and the band the eaves point lies in
gives the required c-section. This the sales people can partly use. Such
charts are beneficial for product development and production planning.

Overlay historical sales data of shed dimensions, and cut out unnecessary
variety in section sizes. On site the chart identifies permissible variation
in the height and span without causing a change of section size.

Codes of practice define loads, and methods of determining resistance,
standards limit available materials and section sizes, and standard practice
further limits the structural forms used. In the building industry a lot of
the structural calculations are an unnecessary waste of time, paper and
storage space. For a lot of situations the only thing that changes between
one project and the next is the span of the beam: and often not by very
much. It is more efficient to determine the maximum span of the available
section for the application, than to keep re-doing the calculation. The
problem is presentation of the required evidence-of-suitability to the
regulators.

For timber framing this is resolved by the national residential timber
framing code AS1684: a pre-engineered (calculated) solution to the loading
code AS1170 and timber structures code AS1720. There is also a publication
on the use of steel in homes with span tables and the likes. There is free
software to AS1684 available from most timber suppliers, but the printed
span tables are faster to use. A sliding chart or wheel would be even
faster. Though that would create a point solution: tunnel vision.

I don't see why pre-engineered solutions have to be limited to residential
construction: hospitals, schools, shopping centres, factories, warehouses
and the likes. (Though I have a preference for factory ships and
manufacturing goods in transit. Humans are meant to be mobile not anchored
to the ground.)

The only thing that makes AS1684 residential is the floor loadings, so it is
used in single storey applications beyond housing. There is more scope for
pre-engineered solutions, and such solutions with published working would
set a reference standard as to what is required and expected of engineers.
It would help improve quality of design where a lack of attention is
otherwise paid to connections, and lateral buckling. Assuming the solutions
are produced, reviewed and approved by the most competent to do so. As
published solutions, they can be more severely criticised and amended in
response. Performance based codes do not remove prescriptive solutions; they
increase the number of solutions available.

Admittedly Australia doesn't have the range of materials and sections that
other countries have. But that is somewhat the point of pre-engineered
solutions, do we have too few sections or elsewhere too many.

To improve quality and productivity in fabrication and site erection, need
to first know what product has to be produced. Consultants can vary sections
to suit the codes as much as they want, but it may be far more economical to
simply adopt a much larger section and optimise fabrication for that. Cannot
know the answer if have tunnel vision looking at single point solutions,
keeping regulators happy.

What impact would it have on the world, if someone decided to throw steel
framed hospitals of an assembly line and supply any where in the world: a
fully operational hospital with personnel? To keep the architects and
sociologists happy the buildings don't all have to be the same: but many of
them can be. A hospital is a machine it has a function to perform, some
variation of form is permissible but not if it impacts on the quality of its
primary function. Should we take a hospital to the sick, or force the sick
to get to the hospital? The problem with asking an architect to solve the
problem is that they will provide a building: not design a solution.

Also what is the difference between a car and a house? Or any building and
the transport available: trucks, ships, aircraft, and pipelines? How can
they complement or substitute for one another? Toyota has only challenged
the car industry so far: but a car is the first dwelling most people in the
industrial world get. With computer networks, laptop computers, mobile
phones, why not turn a multistorey car park into an office? Better still
park at you favourite scenic look out.

The point is that a building is a product like any other, and something
anchored to the earths surface is not necessarily the highest quality
solution. Why resist a hurricane when can move out of its path, before the
season arrives? Why have a mortgage on a block of land and a house in a town
where the major employer has just shut down, and you work elsewhere and
rent, and cannot find a buyer for the house? Why move around with your job
and live in houses of differing quality? Other than egos of architects and
owners why have buildings with 100 year life spans: when can have buildings
with 5 year live spans better suited to our current needs, but last much
longer? Government can put a highway through our property anyway, so why own
a block of land, if it cannot produce food? Why waste materials building a
car and a house if they can be integrated some how?

Some buildings maybe innovative and require structural engineering to make
them possible. Most buildings however are variations of a common theme: and
engineering is largely required to improve ability to supply. Apparently 1
billion people in need of housing: or is that a prison cell that robs them
of mobility and places them under control of central authority? What ever
the perception they are looking for a healthier lifestyle.

Apparently Australia has a shortage of engineers. To me that is not a call
to train more graduates, but a call for the engineers to stop wasting time
calculating what we already know, and to get imaginations and start doing
some engineering. To push the work down the line, from engineer to
technologist, from technologist to engineering officer/associate, and
engineering officer/associate to technician. We can train design technicians
in one year to do the actual job, why wait 4 years for someone who has
potential. Take the experienced top 20% of each lower layer and push them up
to the next. As they move up they learn the detail behind the design aids
they have been using.

Codes of practice maybe complicated but they don't have to be used as is.
Analysis maybe complex and need to be adjusted to suit the situation, but
what is its purpose: to calculate a value or understand the influence of
variables? If we do the calculations once then we can move onto other
issues.

For example a design chart may indicate that for a given application the
moment capacity difference between one section and another is too great and
there is scope to introduce an intermediate section. Cold-forming provides
the opportunity to optimise a section to suit a single application: such as
girt or lintel. Most cold-formed sections offered as girts are not optimised
as such: they are just sections dumped into the market. By designing
structural sections specifically for an application they can be assembled
faster with less error.

Really architects and engineers need to get back to being builders, who use
theory as an aid for decision making. Instead of criticising modern
builders: displace them.

Once a steel fabricator integrates CNC machine tools into a computer
automated factory, what use do they have for consulting architects and
engineers? Get a single building design, feed it into the computers, press a
button and produce multiple copies, and market to the world. Steel is
already being fabricated and transported by air for one-off structures. The
economy for quantity production should be even greater: one copy for each
country then change the design; would probably suffice. As for differences
in national standards find: the limitations, and optimise for a global
market. Hurricane resistant dwellings may turn out to be the optimum
solution. If no one investigates it, then no one will know.

When under pressure to provide answers in 5 minutes, then every design aid
imaginable becomes indispensable. Last weeks solution may be over sized, but
the increased detail this week doesn't mean there is any value reducing
section size. Changing the section size can result in a whole range of
additional components to be detailed and fabricated. Knowing the limitations
and issues beyond loads and resistances is important.

Sorry! Just some points to consider by those who have previously complained
about increasing complexity of codes, low fees, and unreasonable timeframes.
For those who have occasionally mentioned competition from retirees, heres a
clue: they have been there! done that! and know the answers we have yet to
work out. They also have 40 years or more experience to say: ok so far!

Design tools are simply a matter of timing. Do you do wait for a project, or
do you complete the calculations ahead of the need? Need to be able to
address both one-off custom features, and yet generate ranges of solutions
for common situations. I find Excel most flexible for this purpose rather
than off-the-shelf engineering software.

I believe in the next few years the time frame from concept to
implementation is going to collapse significantly, the problem is the
documentation requirements for the regulating authorities. We are still
producing custom project specific one-off calculations: and getting buried
in paper. Yet I don't believe there is any regulation that requires such.

In SA we don't have licensed engineers with seals and such. But scope exists
to get another consultant to review the specifications as an independent
technical expert and certify them. If they need our calculations to check,
and do not have the tools to get the answers in the short time frames we
can: then they are not the experts we need. If left to council, their
consultants will request calculations. If we go to the technical expert
first we can force a proper independent check: get them to produce their
calculations and audit trail. We can also use a private certifier for the
whole of building rules consent.

What is needed is a system of checks and balances to ensure that calculation
results which span across projects are properly applied to specific projects
and custom features are properly addressed.

Pre-engineered solutions also represent calibration points which authors of
loading codes have to justify changing.

The whole design/approval process could be streamlined, made simpler and
faster, and yet subject to greater public scrutiny. Bringing about
refinement in the definition of what is an acceptable level of performance.
For the issue is not if it will fail, but when it will fail and how?

Performance curves and system curves help assess permissible variation and
develop quality robust design-solutions. Point-value calculations good for
the regulator, maybe, but: worthless for design.

So as anyone else charted their area of activity: or even considered it?


Regards
Conrad Harrison
B.Tech (mfg & mech), MIIE, gradTIEAust
mailto:sch.tectonic(--nospam--at)bigpond.com
Adelaide
South Australia
 



******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
* 
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*
*   http://www.seaint.org/sealist1.asp
*
*   Questions to seaint-ad(--nospam--at)seaint.org. 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: http://www.seaint.org 
******* ****** ****** ****** ******* ****** ****** ********