The Materials Discussion. . .

[garethace]

Just a random, blog-type, architectural construction/design rant.... :-) Maybe, just maybe we can instigate some combination of looking at pictures as pictures and looking at pictures that tell you something... so, metal framing building... off the cuff...

 

Anyone here ever use tubular sections? Apparently they are useful things in some circumstances, i.e. where you have possibility of lateral loads on things, or where things can twist/subjected to a lot of torsion etc, etc.

 

One such circumstance, would be at the gable of a large factory, where you need a huge big door. To learn some more feeling for this kind of phenomenon, in the structures of architecture.... I can point you here.

 

I know, on a bicycle, the manufacturers go to extreme lengths to avoid your bodies energy being drained away into just 'flexing the frame' as opposed to transmitting the energy onto the road/tire point of connection.

 

Making strange bends at critical places, and using very tiny tolerances between forks, chainstays, seatstays and wheels, using very stiff aluminium material tubular sections, which are machined to be thinner walled in the middle than at the ends... all this design, not to waste those precious cycling kilo-joules.

 

Bikes are an entirely different set of design circumstances, to building I know, but it just highlights how deeply that some engineers/designers have gone in relation to 'the act of making' something.

 

Just to get the discussion some way started, I decided to link this image, perhaps others have more? Here is an example of an image, that really shows, how much people 'think' about frame construction... to a real bike user/sport enthuasiast, this is so much more than a pretty picture.

 

http://www.specialized.com/SBCBkModel.jsp?spid=9693

 

I feel that with architectural visualisation, the image often 'means' so very little unfortunately, just a statement of fact actually and not something 'I want' to believe. BTW, for those of you, who really find that image confusing without wheels etc, attached, that black post sticking up to the top of the picture is where a saddle would go and ultimately the 'rider' of such a piece of equipment. That frame is for rough, off-road usage... unlike the lean, and minimalist ultra light permanent road surface cycling bikes, that I described above.

 

Every bit of machining, weld, struting or whatever on a bicycle frame, is more money the consumer has to pay up... And if you are standing on the shop floor with wallet/credit card in hand, you really are conscious of what you will pay for. Compared to when we design/visualise buildings some times, and are not as concerned with properly using the clients budget! :-) So.... jumping back from bicycles, and back to building specification and technology world once again, here is a quote from a Steel Systems Construction reference book:

 

To be effective in meeting these goals, the bracing system must provide stability to the whole C or Z section, not just one flange of it. Laterally urbraced girts and purlins tend to fail by the lateral-torsional buckling mode, well before their full flexural capacity is realised. When these sections finally lay flat, they are only as strong as their weak axis section properties allow. Profiled sheeting of standing seam roofing only provide 50% lateral bracing.

 

So bear that in mind, with reference to the earlier talk about a cyclist transferring his/her energies, not into bending the frame of the bike, but into the action of cycling, and moving forward. Aluminium is currently the preferred material for cycling frames, due to its stiffness, unlike Steel which flexed a lot and was comfortable to ride therefore, but ultimately a lot of what you were working for, was just 'absorbed' by the actual frame of the bike.

 

Cheers for reading.

 

Brian.

[garethace]

Now if, you take the above post in relation to bikes/buildings as being about materials and strength of structure... read the following piece from that bike frame web page also.

 

THE STORY

DEMO FSR frame. Super low center of gravity and standover with sub-seatstay actuated shock. Full cartridge bearing pivots, 9” of travel with full chainguide and front derailleur compatibility. 150mm wide 12 or 15mm thru-axle rear, ISCG mount, rear derailleyr protector. 26 and 24” wheel compati ble with two geometry settings, 16.7” chainstay length, forged and CNC’d OS HT, Super ORE TT/DT.

 

Bacause, if designing bicycles were to be compared to architecture... then the above, is very like the 'constraints' not posed by materials or structure.... but things like ordinances, site services, civil stuff basically... the bike frame has to satisfy the conditions where it can 'wear' many other standard components, that the frame will need to give it the functionality of cycling.

 

Here is my rant about civil + computer aided design tools.

 

Gee, in my recent threads here, I talked somewhat about the 'reality' of working in the field, as opposed to piping some views, that I read in books or were interested in.... So just to prove, that I haven't lost all committment to digital end of things, I decided to start this particular thread.

 

Basically, it is a series of 'what-ifs', what if the design software companies were to 'mainstream' this whiz bang new feature or that, how might it affect existing CAD documents and archives, especially considering my believe, that computers have the ability to destroy an awful more than they have to create.

 

 

Reactor Technology.

 

civil design takes over. The evolution emerging here is a dynamic environment

in which models of proposed designs are built with intelligent objects controlled by rules and styles. Examples are a roadway that must adhere to minimum and maximum grades, or parcels that must maintain minimum frontage and areas. The software aids in the design, and the entire model reacts to changes in real time, hence the term reactor technology. As design changes are made, they ripple through all aspects of the project and update working drawings automatically.

 

Land XML

 

The idea of sharing civil data between applications has been a dream for many years, and LandXML is increasingly becoming the answer to that need. LandXML files provide a format for storing points, surfaces, alignments, parcels, pipe runs, and roadway models. Any civil/survey software developer can now develop the code to read and write these files, thereby easily sharing these types of data. I recently took a roadway alignment developed in one application, exported it to LandXML, and then imported with another application to produce an uploadable file for a specific data collector. The roadway appeared on the data collector screen, ready to take in the field for stakeout. The developers of the two applications, as well as the manufacturer of the data collector, may never have anticipated this specific

chain of events to occur between their products, but LandXML opened the door for seamless communication anyway.

 

Bearing in mind too, having worked on a load of civil stuff as an intregal part of several projects now, i.e. the first stage 'to develop' any large scale development, is to organise where/how you are going to run utility type of stuff, months if not years even before the money, investment and interests comes around to actually do the building stuff on top of the grade level. So going through so much pre-planning and careful thought with civil stuff, you really don't want your civil digital documents, to suddenly get a live of their own, backups, or no backups,... it is always easier to have one consistent set, throughout, and avoid confusion of versioning etc, etc. Just bear that point, in mind, in relation to the above two quotes.

 

In a somewhat similar way, to how a bicycle frame designer would hope that their chosen software design platform, would guide rather than obstruct them from ending up with something, that 'fits' with all other necessary cycling components like derailleurs, gearing, drivetrain etc, etc.

 

Brian.

[IC]

More, more!

[garethace]

Patience, feed me... :-)

 

Anyhow, thanx moderator for cleaning up that other spare thread I left, nice one.

 

garethace.

[Crazy Homeless Guy]

well i appreciate the thought that one day a BIM work environment will actually work the way it is supposed to. everything is intricately linked together, everything updates dynamically, and some 'standardized' products have preinputed data by the manufacturer that details/limits/expands/ect.. their use, within the capabilities of a prebuilt object that is essentially off the shelf, therefore mass produced, therefore cost effective.

 

it is both intriguing and scary at the same time.

 

roads are roads. roads are built so another object can function on them. which came first the road or the car. the car, then the modern road was adjusted to fit it. and in turn, new cars are designed to function on the existing conditions of roads. they both continue to build off of each other, but the technology of the car will always control what technology the road has. both electronically and physical dimensions.

 

bikes are designed to efficiently transfer the energy a human produces into forward motion. the bike is fine tuned for an ideal condition, then mass produced for use by people of different body types an builds. the bike is not tuned for every person who will be riding the bike, so the truly ideal condition of the proper transfer of energy in that bike frame rarely exists. the bike frame is like the road in that it is mass produced to fit ideal conditions. the body is not like the car in that it is highly variable. the design of the body is not an exact science, the design of the car is.

 

buildings are designed to fit their individual habitants, whether it be a house, a lawyers office, a hospital, or a Wal-Mart. in this sense it is like different types of bike frames for different purposes. you have your off road, your street, your bmx, ect.. each bike type fits the use the person wants it for, but each bike type is mass produced. the same is not true of buildings, nor do i believe an environment where it is would be a good thing. cost effective yes, but how would it be good? would we wind up with anything more than your standard gas station, your standard box Wal-Mart, your standard duck for each type of building needed.

 

there obviously is a market for this. but is it good?

 

i don't really think this is exactly what you are suggesting though. i think you are suggesting some level where we use computers for a productive design environment that allows us to explore options that are available to us in a dynamic way that show what the exact effects and consequences are of each and every piece that is applied to a building.

 

this reminds me of a combination of the thread about gehry's software system from the other day combined with the thread from a while ago that discussed where the future of the arch viz industry was going. the later thread about the arch industry talked heavily about the integration of all facets of arch design being developed to work in collaboration with each other. the thread about gehry's architecture discussed the building system software that he used to execute his designs, and why he needed to use that software over other resources available to heavily control everything from precision manufacturing to cost control.

 

i think we are seeing software cooperate in a more friendly manner with each other, but it has been a long drawn out process. look what happened when with Sun and Microsoft. Microsoft used Sun's Java, and modified it, adding commands of their own, effectively making Java a non cross platform coding language.

 

look what happens between different web browsers. some browsers support different technology, sometimes proprietary, forcing you to use their browser in order to view the pages.

 

to effectively make this happen, a system of regulation will need to be developed like the W3C for the world wide web. they are the authoritative group that controls the standard HTML commands so that companies like Microsoft do not completely take over the code, and make it their own. their are similar groups for JPEG, MPEG, ect.. it would need to be a grouped controlled by several entities that have equal authority in making a decision, otherwise you wind up with one company that is controlling to much, and people state that they are a monopoly over the industry.

 

anyway, i may be completely off track with what you were suggesting with this thread.

[garethace]

Here is basically what I am talking about...

 

Economy of Scale: Software Development. . . Is it important to know about something to know how to write digital tools for it? Agree or disagree? I mean, anything from games to how to design a ship for the high seas.

 

I recently read a post by Vincent (another forum, chess AI algorithm expert from Holland) here, about early writers of chess playing software not actually knowing that much about playing chess. I mean, I guess, the same could be said about John Carmack (writer of code for Quake, Doom etc) and games. I dunno. But recently I wrote a blog rant comparing design of architecture, it's software, with that of a closely related cousin, software to design bicycle frames. So calbiker (a biker), Vince, and some others might find this piece of interest. Software 'to design' I think should be closely related to the sort of thing it is meant to design, yet that goes totally counter to the point of commercial software in the first place, which is to provide a generalised solution, to as wide as possible a base of customers. Hence success of Microsoft model say, in operating systems sphere. But is this lack of specialisation of softwares, to save costs on development, getting a bit much?

 

Anyhow if you have 2 hours to spend... listen to some of this....

 

http://wpcarey.asu.edu/seid/cabit/events.cfm

 

Down at the bottom the video streamed of the debate is excellent... here are the rough notes I have already scribbled down... will put into some kind of format soon. It compares Sun's J2EE and Microsoft's .NET.

 

Web services. An awful lot of 'stuff' out there that was installed in the 1950s. Data residing in all kinds of places. CxO level people making short term decisions on IT infrastructure, because they do not last much longer than 4 years.... IT just got out of control. Today, bandwidth has gotten cheaper... a lot of backbone, even on college campuses etc, storage has got cheaper... a lot of bandwidth there... coming in through all different services and lines... I want to put this data into the package that I want, rather than the package that peoplesoft want to sell... because there is no such thing as a standardised workflow engine... Ask any company/CEO... to explain to me the process of how your business runs... he/she could not do it.

 

In the typical organisation today, you will find very good systems which were built in isolation from each other, and very good at doing their own particular specialisation.... one example that comes to my mind, is Land XML, in city corporations, where the CAD guys were great at doing CAD and GIS guys were great at doing GIS, yet focussed more upon doing their own stuff very well rather than communicating with each other.... incompatible software. Absolutely no easy way to share and exchange data between different departments... Businesses and organisations grew on a department by department manner... add to that, the business is becoming more global and businesses are expected to talk to each other more.... and the situation is totally untenable.

 

Members of these organisations now wish to work remotely, on an airplane, work from home then sync up. Sun Microsystems doesn't want to give everyone an office at Sun at more... they will enable you as an employee to work for home, become very mobile or give you an office if you really want it.

 

But in the field it is really very difficult to find ways to get these old distinct legacy systems to talk to each other. Yet there is a need for a number of systems to interact together at any one point in time, even if it is just for a short time. Suppliers, competitors, employees, governments... all of these systems need to talk to each other.

 

It becomes a huge mess.... changing the way we build software. build software that inherently able to talk to each other. vast proliferation of bandwidth... vast proliferation of different new devices on the market now which people have.... what do you think of as your computer today? Is it the PDA, phone, your car, your desktop, laptop, what... or all of these? ..... is it fun programming in all the numbers for your mobile phone having lost one? We have so much better bandwidth nowadays.... what we have is a ridiculous proliferation of bandwidth through broadband, wifi, cell phone networks and so on... which talk to a ever wider variety of devices coming out all the time.

 

How can you you lever this, to take advantage of all of the information that drives your business?

 

XML exposes the functionality of your information to various parties at agreed places/times, levels of authorisation... that is a web service... not only does it do that, but it changes how we design/build our programs. The old way,... proceedural way of look at designing software.

 

Now we just get, 'Components' that 'know' how to place documents in a supply chain management system, or to place documents in a ERP system, or any other kind of task you can think of. It is now just about how these actual 'components' talk to one another. They need to 'talk' in a standard way to each other.... integrating all new/legacy systems in a way that has never been possible before.

 

You can build or customise your new email framework, or infrastructure... in a way that suits you. Because each I/o functionality, each security functionality, or any other feature that you can imagine becomes virtualised... components are all exposed through web services, and put together....

 

Sample situation... I want to use something just for today... but I don't want to pay for it very often.... but today I need it... I only pay for it today. Utility computing... Not a permanent license.. MS Office... think about all goodies you get that you never actually use in there!!!!! Imagine you could just assemble your own software? Extend and leverage my existing access to my own data and information flows, through web services?

 

Therefore it is simple,... If your current vendors don't work, get rid of em... like Scott McNealy, talked about car dealerships and buying automobiles in USA way back in the old days, customers could literally march with their feet.

 

'Snap in' as opposed to 'lock in' environment. Noone wants to be locked in forever to a single vendor. You want to make your valuable data more available to the partners, customers, employees etc, through the filter of web services... i.e. data is more context aware. You only 'see' the services that you want to see, that you get access to... Web services on demand... the portals that I want to subscribe to... what things do I need to do my work... so even on a hand held device a trader on the floor at Nasdaq can assemble his/her software and services rather like you would build your own house.

 

I suppose the CAD manager here at my workplace, already does this to a certain except already with the way he manages, or customises AutoCAD to be used, the way he wishes it to be used in house... Similar managers probably do this with other programs like having specific excel macros or word document templates... guidelines or rules for employees and partners to adher to.

 

Case Study.... a company that rents cars, booking an airline reservation... old compaq servers.. cobro environment legacy... (a real challenge).... so wrap it in a wrapper and 'put it out there'... it ends up saving 4 dollar per car rental, which represented a profit for the company.

 

Other example... monolithic online banking platform, wrap it in web services, and extend it to do other services, that had not been available to the online banking customer, like offer morgages etc, which it couldn't do before.

 

Other example.... Cobalt programs used by oil companies... big nasty monolithic system... savage loyalties for IBM to change or update how the software works.

 

In this instance, the oil company just needed to modify their reporting system from the software... to satisfy some new standards or environmental regulations... they ended up doing that through web services rather than going through IBM.

 

This is really called the 'Virtualisation of Services'... it can help your company or organisation to achieve the right application stack, right combination of applications. And yet it somehow manages to leverages all of your existing assets.

 

True componentisation. Re-utilisation.