Need a book? Engineering books recommendations...

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

Manufactured Building Products: Structural Assessment

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Yes Gordon: its public service charity work. They may want the answer
yesterday, but they don't get it, they have to wait, and it takes as long as
it takes. Further there is always someone else to push into the queue whose
problems are even more immediate. Some of the problems may only take a
couple of hours to solve, but with a constant stream of enquires and little
time to do the work, some clients still have to wait 12 months or more. They
have little choice in the matter. (I read this morning Australia has a
shortfall of 28,000 engineers.)

Well they did have a choice. They could have got it designed before they
submitted an application for development approval. Or even better got it
designed and approved before they built it. I estimate that roughly it takes
about 5 times longer to prove something is barely adequate than it does to
design and specify. That is pushing the limits of what the code allows, but
then don't need contingencies for last minute changes during construction.
And for the existing it's not just a matter of applying for development
approval or demolish, if it needs to be strengthened that also has to be
done within a time frame or the demolition goes ahead. The council (AHJ)
doesn't really want to enforce a demolition order. So providing a rational
justification all parties are willing to accept is preferable, and limit
state, probabilistic design methods and performance based design codes all
help make that viable. It just takes a long time to do, but where waiting is
viable people reluctantly wait.

After all if they go elsewhere, they will get an extortionate fee, a
simplistic design not an assessment, will still have to wait, and they will
also have to pay to strengthen a structure which has often been around for 6
years or more before discovered, and often the owner is unaware that they
own property with illegal construction. Further we now have new laws which
uncover this construction at point of sale, and development approval has to
be granted before a sale can go ahead.

Locally our building surveyor has a reputation as being the most pedantic
and uncompromising building official in the state. So it seems people avoid
seeking development approval, to avoid delays in getting approval. The real
cause of the delays is anyone can submit an application, usually do, and the
application is deficient. Submit decent application get approval quick.
Further more little chance of deviating from the deemed-to-satisfy
provisions of the code, and using commentary and theoretical basis unless
get it right.

The other point is if you use it then it gets adopted. Just about everything
I have used from commentaries has migrated into the body of the code. The
local building officials are more directly in contact with those that write
the national codes, so any criticism of the code in the process of a
project, will also migrate to the code writers. Determination of resistance
may have scientific basis, but acceptable level of performance is a
subjective and political issue. The codes cause the building officials as
much trouble as the designers.

Most clients are just happy to finally get some closure on the problem, even
if they still have to make some modifications. Others are not happy about
the long wait. My view, is they are usually the ones who built something in
my neighbourhood without design and without approval, or otherwise thought
they could prepare their own application. For the illegal construction the
demolition order equals the design wind event: sorry if you designed it,
then it would have survived. For the DIY's applications if they had got
professional designers in the first place it would have been approved
already. It is these people with their emergencies, pushing in the front of
the queue, who cause delays for everybody else: who are otherwise attempting
to do the right thing.

The other issue is, when at first it appears how can anyone answer the
question in a few minutes. It usually turns out they have years of
experience, and that experience is important. It seems some engineers have
minds like computers and are wiped blank at the end of every day. They do
not understand the relationship between physical parameters; they simply
plug numbers into code formulae. So that every project is one-off. But if
loads are standardised, and available sections are standardised, and the
structural form is standardised and the application constant, then the
limitations can be determined and simple lookup tables and design curves can
be developed, or software. The problem is it may take 1 year or more to turn
1 week into 1 minute.

As for the older engineers. Besides having some 40 years experience, they
probably also designed the existing structures they are looking at, so they
can make quick decisions. And the profession does itself no favours by
constantly arguing about their importance to life safety, and that
compliance with new codes is about life safety. Compliance with codes does
not protect life, no matter what the design load is there is always the
probability of it being exceeded, low though it maybe. To convince the
public it is save to live in an engineered building in the path of a
hurricane is dangerous. For people to get complacent about environmental
hazards because a building survived the last event is even more dangerous,
especially if the structure is getting weaker with every event.

People really need to know that we are not designing buildings to be
serviceable "safe" shelters during the extreme event, and further that the
buildings will experience significant damage and require repair at even
lower loads. Under Australian codes there is no mandated performance level
for serviceability. Serviceability is a qualitative and subjective matter
for designers and owners. Setting a serviceability level can provide greater
resistance than required for ultimate strength requirements, it can also
cost more. On the other hand: had a few clients turn up declaring they don't
want a steel framed floor because too bouncy? Span tables for timber framing
are based on checks for serviceability. Many engineers seem to ignore
guidelines for floor vibration, because such handbooks are not mandatory. It
is thus possible for a structure to be code compliant but low quality as
perceived by the end-users.

Tell someone that the design wind speed has a 5% probability of being
exceeded during the 50 year life of the building; they will think the risk
is too high even though they also think the wind speed is too high. Yet
still the housing industry complained about the speed and the codes
increased the risk to approximately 10%, or was that drop the life
expectancy of the building to 25 years and maintain the risk at 5%. Given
the rough basis of our codes here: is 5th percentile resistance being
greater than 95th percentile load, it looks like the life expectancy has
been reduced. (code now (2002) phrased in annual probabilities not the
original 5% for 50 year life expectancy introduced in 1989.) Just politics
and economics. A 30 year mortgage on a building with a 25 year life
expectancy doesn't sound good: but that doesn't mean the building will only
last 25 years it could last longer. But the longer it lasts the more likely
it is to experience loads in excess of the design load.

The majority of our buildings exist already. And each and every building is
a risky one-off real world experiment. As with any other product once
released to the environment users will find a multitude of applications
beyond the intentions of the designers. It could be contended that each and
every one of those alternative applications requires an engineering
assessment for suitability. It would make life really cumbersome if needed
to produce formal calculations before made any decision and took any action.
Sometimes the need for calculations is unrealistic. Don't stand on that desk
until complete some calculations and get code approval.

Here the building code states: evidence-of-suitability. There is no mention
of requiring calculations, yet certifiers always asking for such. The
requirement of certification is an independent technical check of the
proposal: they are not doing that if they request calculations. On the other
hand the certifier is not permitted to say what is required. So for a DIY
application, if simply looking at the proposal suggests it is no good, then
no point in certifier doing calculations, requesting submission of
calculations will get the proposal designed.

But there are also issues of intellectual property rights. The AHJ whilst it
may gain experience from prior submissions as to what is and is not
adequate, it is unfair to use such experience to approve applications for a
structural product produced by copy-cat manufacturer/builders.

Designers on the other hand can use their experience of designing other
versions of a product, when the product has no protection. That is the
designer can distribute the cost of research design and development across
multiple manufacturers, rather than hit the first manufacturer with a huge
fee. On the other hand those civil engineers who set the market rates failed
to recognise the value of selling hundreds to thousands of building products
per year: the manufacturers get rich the consultants struggle to make a
living. (Hence shortage of engineers)

But then engineers could always take the risk of setting up the
manufacturing business. After all have more knowledge of the products than
any of the manufacturers have. Which is the other issue not constrained to
one building product.

It is also not necessary to do detailed calculations on a project by project
basis. Can make quick conservative assessments at the time of the problem,
and do the more detailed assessment later, to determine how close to the
edge your educated guess was. It is an economic battlefield, the enemy is
fast approaching and you have to bridge the ravine now, not tomorrow: life
is in danger.

It is therefore preferable that engineers are proactive, and put the
investigative work in before the clients come along with their problems.
That means clients pay more for you knowing the answer rather than for you
having the capability to find an answer eventually. Most of the technology
exists already, the design models in the codes are derived from existing
technologies, the codes are pushed to their limits when applied to new
technologies, and eventually the codes have to be revised, based on the
performance of newly existing.

Really do need to recognise the difference between building products, and
one-off custom designs. The approach to dealing with repetitive building
products needs to be different to that for one-off custom design.

For common place building structures really need to be doing the research,
design and development between projects not on the project. At which point
regulations become a hindrance, requiring submission of what is otherwise
pointless scrap paper. This can be avoided with independent technical
certification; unfortunately the builder/manufacturers do not do what is
necessary for the consultants to remain independent. The engineer looses
independence and becomes designer, and another engineer is required for
certification. The latter being left to the AHJ, who will request submitting
the calculations. It could all be a lot more efficient. Which is why we have
the light timber framing code AS1684: the structural calculations done
already. The problem is people paying attention to the feasibility of a
proposal existing off paper, and the practicality of fabricating and
constructing, and otherwise knowing when a proposal is beyond the scope of
the framing code. And when an element is beyond the scope of the code
(AS1684) it becomes a timing problem: a 1 minute problem turned into 1 or
more hours of calculation. But once again develop the design-process
infrastructure, and get the time frame down to 1 minute. It necessary to
engineer our own activities as well as that of our clients.

For those engineers who specialise in quality custom buildings, then
basically they do not have the infrastructure for providing rapid solutions
to immediate problems, and often they also do fairly conservative design
rather than push the codes to their limits.

Which is good for those of us assessing variations in usage: we can push the
codes to the limits and find reserve capacity and make modifications
practical. The impractical being demolish and build something else. For the
most part it is a matter of accepting lower levels of serviceability
performance. The buildings have quality with respect to compliance to the
adopted specifications, but have lower levels of performance. Some may
perceive the buildings lack quality when compared against higher performance
buildings, but that's another issue. {NB: I don't advocate pushing codes to
limits for design only assessment of existing. Manufactured products or
generic structural forms are existing, much of traditional can be maintained
by greater engineering effort in use of codes. If few willing to put effort
in then new codes increase performance of new buildings. So it becomes a
matter of whether want dollars in the materials and building performance or
dollars in engineer's fees to reduce performance to minimum permitted.}

Whilst some engineers may charge low fees and provide quick answers to AHJ's
with out real justification. There are others who have no choice but to
charge the low fee, and also no choice but to provide real justification to
the AHJ, anything less won't get approval. But it does provide bargaining
power to raise fees: yes, that may work in the neighbouring county but it
won't work here. And if they try it, they return to get our answer.

Working with manufacturer/builders really is a specialised area. From my
perspective the civil engineers in the industry have caused many of the
problems, they promote the concept of minimum weight structure. So the
competition between manufacturers is to use the smallest section size
possible. There is little promotion of higher performance products, and the
only interest is complying with regulations, no effort to exceed the
mandated performance criteria, and no concept of whole product: goods and
service supplied. {NB: And no insurance company smashing houses to pieces as
they do cars each year.}

Attempts to modify product design to reduce on site labour costs are of no
interest to manufacturer/supplier: they are held to ransom by installers and
erectors. No matter what the product is: carports, sheds, balustrades, sail
shades, water tanks, solar hot water systems, and next I guess solar power
panels. Though the suppliers of solar hot water systems must have got
themselves sorted out. {We get a lot of our work because we are the only
consultant with our office in the area, all others on councils short lists
are in the city some 25km away and are typically not interested in small
troublesome projects. The exception being when the city consultants work for
builders located nearby their offices.}

Though if it is possible to get to work closely with a manufacturer and
installer, then you can start to bring about some changes, get up front
design carried out. But it does take time to bring the manufacturer/supplier
around to seeing the benefits of planning and design, and it does require
effort on the part of engineers/designers to push the message through. It
requires being proactive, rather than waiting for client to come along with
another problem. 

A controlled and regulated environment, in which only "perfect" products are
released to the environment, by qualified people, is mythical. Part of our
task is riding the wave of chaos and bringing order. It is a headache, and
philosophising the reality of the situation is how I attempt to see things
from my client's perspective when I criticise just about everything they do.
Say it often enough in different ways, they eventually start to understand.

Working for manufacturers is not all that different than working with the
same architect on many different jobs. You start to resolve and avoid common
problems: but it takes time.

In any case I think it is inappropriate to automatically assume a quick
answer today is the result of poor engineering. Without the efforts of
Coulomb, Navier, Bernoulli, Telford, Brunel, we wouldn't have the structures
of today. How many modern engineers are able to derive new engineering
theories from just a brief background in mathematics? How many engineers
willing to take an educated risk and introduce new technologies? How many
real imaginative ingenious contrivers of civilisation are there? How many
real engineers? Very few I hazard.

If modern civilisation was dependent on the modern engineer, then
civilisation would not exist: it is too great a risk to take the first step.
We have the benefit of history to build upon and avoid common errors of the
past. But we also have the benefit of existing theories to move forward to
prescriptive solutions for common applications.

Performance based codes are not there to eliminate prescriptions but to
permit a greater range of prescriptions: better suited to the application.
They do not infer a need to keep churning out calculations to reach the same
conclusion reached last week. Each set of calculations builds experience,
but don't have to wait for next project to assess the consequences of
varying one or more parameters, and generating tables or curves. Even if
only saves 50% of what have to do, still have more time for the other 50%,
and with time as move from project to project, the percentage which is past
experience grows.

Some structures may always require individual calculations, but experience
determines whether the first pass through an iterative process fails or
passes.

Thus do need to select projects carefully.


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 
******* ****** ****** ****** ******* ****** ****** ********