Daily Challenge Day 33

As indicated last night I was looking for a web site, I had visited the website last year, and the service advertised shed design using a virtual wind tunnel. I wasn’t able to find the site, most of the sites I found, as expected, concerned computational fluid dynamics (CFD): I doubt the site I was looking for uses such technique. First because as far as I know CFD still takes a few hours and is not overly practical for general use. Secondly because CFD is an approximate numerical method, and therefore needs to be calibrated either against the wind loading code to provide compatible solutions or be calibrated against actual wind tunnel testing. I tried various search conditions and couldn’t find it using google search. I also searched my MS Access database of web links which contains about 80,000 bookmarks. Then I remembered I downloaded an example report from the site. Unusually finding that file on my archive drive was easy. Looking at the report it wasn’t actually clear who did the calculations. The report was an example for South Australia, and contained a certificate of an independent technical expert, which made it confusing. Other states may permit self-certification, South Australia does not. In South Australia the certifier is not permitted to be involved with the original design and specification. If you are the one determining the size of structural members then you cannot be the one certifying the specification or calculations. So assuming they had complied with such requirement looking for the certifying engineer by name didn’t seem much value. So I took a chance just looking for the report by filename and google search grabbed it straight away. Once I got the website I searched my database and found the previous saved bookmark.

So this is the current title of their website:

Sheds Design | $900 Flat fee | Any type | Any size – Get your big shed designed for just $900!

The previous title include price for $800. Compare this to costs for building design and drafting services, as made available by the following websites:

  1. New Wall Design
  2. Streamline: Drafting & Design Services

These are the few sites which display prices. Personally if I am searching the internet, I do not expect to have to phone up for prices and further information. If I wanted to talk to someone I would have looked in the yellow pages and wasted my time phoning around asking the same questions over and over again. Why spend minutes on a conversation, when I can scan writing in a few seconds? So if website doesn’t have prices, I assume I cannot afford and move on, and possibly ultimately decide to do it myself.

Anycase the general gist is that the near complete documentation requirements for a custom shed will be less than $2000 at this point in time. Near complete because there are no structural drawings included, unless the building designer specifies the structural requirements.

Further compare this to our fees for standard shed and/or canopy designs. Around 1996 we charged, around $500 to $600 for standard shed design: consisting of structural arrangement and connection details on A4 sheets bound with structural calculations. The shed designs were the maximum heights possible from a given section and other parameters. To get the maximum height as mentioned previously I wrote Quattro Pro spreadsheets to generate data files for our in-house plane frame analysis program. The spreadsheets, adjusted the wind loads based on variations with height. With the passage of time both drawings and calculations were made more detailed. We were able to give shed/canopy manufacturers sizes of main frame within 30 to 60 minutes of being requested, on condition that no columns were removed from the shed requiring carry beams to support the rafters. We pushed the price of the standard shed designs upto $990 by around the year 2000, and shifted to using MS Excel and replacing the plane frame analysis with Kleinlogel formulae and also getting MicroStran for 3D analysis.

After around 2003 the fees for custom shed design increased to over $1000. These custom designs typically included offices to warehouse buildings, therefore no longer BCA class 10 buildings. Various features such as bulkheads and awnings were added to street facing sides to reduce the large areas of sheet metal. The structural calculations were typically conducted using MicroStran and Multiframe, and my spreadsheets. We worked in conjunction with a building designer and steel erector, rather than for shed manufacturer. We also moved into producing workshop details for the sheds and canopies: the steel erector wasn’t happy that the manufacturers often failed to provide all parts. The last such shed designed, working in such team, was a shed with a span of 32m. Unfortunately due to changes in the economy and non-business issues, the person finding the work needed a rest, and the work dried up. Though there seems potential that it may start up again, though a bit difficult with our current location. Anycase, in the mean time we were working with carport manufacturers incorporating our 2D plane frame analysis program into carport design spreadsheet already set up for costing. That and designing glass and aluminium balustrades, sports nets, and anything else that turned up. As that is what we primarily do, small scale structures of just about any nature.

Anycase the thing to be noted is as I have always indicated, most things can be designed once and built many times. The original standard designs provided envelopes, defining suitability of a given structural section. Now computer software operated by sales people can determine the specification for a shed or canopy in less than an hour consultation. However, it should be noted that the software neither designs nor engineers. As I have indicated before engineering takes place at the frontiers of science and technology. There is no frontier involved here.

Whilst the software maybe doing structural calculations to make its decisions, it is neither engineering nor designing, it is simply selecting suitable components and specifying. The customer and salesperson are specifiers, and the software is an oracle assessing whether the wants and whims are physically possible and the requirements to make such possible. If there are no suitable components, then the software simply reports not possible or something similar. A designer on the other hand can make it possible. For example the cross bracing in the wall may simply use a single bolt to fasten to the column flange, the shed could be given parameters where such single bolt is not adequate: not because of the bolts but because the sheet metal fails in either bearing or tear out. The software has a single connection configuration, a designer can adapt or invent any connection detail to suit the purpose. The software is just a product configurator with limited custom configurations available.

Now more flexible software like Pro Engineer, Revit and Tekla Structure can create parametric models much faster than writing custom software: unfortunately such software is too expensive to put on every salesperson’s laptop. However, using parametric models as shown in this truss tutorial , a few changes to a few parameters and new shed plans can be generated quickly along with workshop details. I couldn’t find any parametric uses for portal frames, though it should be possible as in 1989 I saw demonstration of Pro Engineer and the sale rep, created a simply stick diagram of gable frame and demonstrated how span could be related to height: though had some problems setting constraints for the roof. However that was really a time issue and knowing the constraints required, not the limitations of the software.

Even without creating parametric families watching videos on youtube indicates that a portal frame can be created in less than an hour in either Revit and Tekla structure. The tutorials I saw were cumbersome, and contained errors, and went for about 1 hour: I imagine if a lot more confident and familiar with the task at hand then it could be done in less than 1 hour. Once the portal frame is defined, the rest of the building shouldn’t take much longer. With building information model sharing of data between design and structural analysis software, it is looking like a portal frame shed can be designed and detailed in less than a day: with the primary obstacle converting the digital solution to paper like documents: that is printing traditional like documents even if to pdf files. Traditional documents are not required because CNC machine code can be generated to directly control automated beam lines.

Which is where we need to consider another issue. In the UK, USA and India specialist shed fabricators use flat bar which are welded into tapered sections to minimise the use of steel in the building structure. Flat bar can be automatically punched or drilled. The flat bar is  then automatically welded together. On the other hand cold-formed steel is typically higher strength steel than flat bar, and can be automatically roll formed into a variety of structures. Unlike the USA and UK, at the present Australia seems to lack stock standard plain channels which nest with c-sections.

However the steel house frame industry does have a variety of nesting and compatible sections, all of which can be automatically generated as required from highly specialised rollforming equipment. Further the software they use can model the building in 3 dimensions and analyse such proposal, then generate  CNC instructions for the roll form, and print mark numbers on each component. They can design a whole variety of building structures and other structures using such software. The two main systems I know are:

  1. Vertex CAD/PDM
  2. Framecad

With one of the suppliers using such systems being Dynamic Steel Frame. With the capabilities of such systems I have seen demonstrated so far I would suggest that it wouldn’t take much effort to design rollform machinery compatible with boxed sections and designing carports, verandahs and sheds. The shed and canopy manufacturers are typically relative primitive, using the minimum industrial technology.

Whilst the major national suppliers of sheds and canopies are making business harder for the small local suppliers, I would say that they themselves will be in trouble as the software, they use if they have any, is no way comparable to the likes of the Revit type BIM’s and the Vertex like PDM’s. Similarly most have no integration of their design, materials resource planning (MRP). manufacturing resource planning (MRP II), and factory automation systems. In short design to fabrication is slow. Construction is also a bottleneck, but I doubt it can be resolved by 3D printers, as there are set up times required for the printing equipment.

However before any revolution in automated production. The first stage is likely a revolution in the design, with plan drafters  and owner builders increasingly able to specify their own building requirements which are fully compliant with building codes without need of the services of “architects” or “engineers” . I reiterate engineering takes place at the frontiers of science and technology and we have no such frontiers here.

If it can be manufactured, and mass produced it will be designed once using parametric models, and then made many times. Owners of buildings don’t always want to pay the prices of building manufacturers, they can cost the materials themselves, and identify the materials cost significantly less than the supply price. If the project looks easy to fabricate and build themselves, then they will do so. The parametric models, will eventually make the designs readily available to all.

If manufacturers with 3D printers can offer services over the internet, then roll formers can also offer custom manufacture of buildings. They just have to computerise their order system, and the control of their rollform equipment.

So if you are a small builder, and can find a drafter using either Revit or Tekla structure, make sure they are using the software efficiently, that is parametrically, so that they can rapidly adapt the first set of drawings for future projects. Also make sure provide them with a reasonable workload, else they won’t around for long. (eg. Don’t say will provide them with work for all your projects, and demand low priced distributed across such, and then only turn up with complex projects which cost far more than the average.)

Note that design and documentation is just a means to an end: it is a bottleneck to implementation, so every minute spent on design and documentation is a minute too long. On the other hand deliberately delaying implementation may be a good thing, once design and documentation time has been near eliminated.

The only obstacle I see at the moment is that the available software is too expensive and channeling an appropriate amount of work through the software is too hit and miss. Secondly custom shed manufacturers with software such as Revit, Tekla Structure, and Graitec structure, do not appear to be employing the right people to get the most out of the software. If you have such software and are asking me about my scripts to automate Acad LT, from Excel, then really ought to send your employees on proper training courses or give them access to the software and opportunity to get the most from the software. Shouldn’t need my scripts, should be able to achieve something far superior in far less time than writing such scripts.

My estimate would need between 3 months and 1 year part-time to develop the necessary resources to get the most out of such software relative to the needs of an individual business. The development work probably needs to be done after hours on weekdays, or on weekends, so that doesn’t disrupt ordinary work flows.

It is highly unlikely that anyone can develop custom software from scratch, with any where near the flexibility of the high end BIM and structural analysis software,and have it automated to suite salespeople and the public as operators, for a fee less than $50,000. Sure can do something for less than $5,000 as the zero fee spreadsheets on ExcelCalcs and here demonstrate. But even then to customise to suit the needs of specific business looking at some where near a week: and at federal minimum rates that’s a fee of at least $672.70. But to get such work done need people with know how and resources, so it will cost more than that. So only if there is a high demand for simple tools, which almost immediately meet the needs of the end-user, without need for training ,can the fee be dropped less than that of the federal minimum for labour.

From my education in industrial engineering, my task is to enable and empower and make myself redundant: but be creative enough that I am immediately re-employed.

Every time that a structure’s design is standardised, or made into an adaptive parametric design using computers, then that is one more class of structures which do not need so called “engineers”. That then reduces the work for the “engineers”, and ultimately reduces their numbers, and also their availability for future work which may turn up.

However parametric design does have its limitations. Whilst it can modify an object with respect to its dimension and geometry, it has limited capacity for including or excluding an object. For example can adjust the dimensions of a truss, but increasing the number of diagonals to limit the segment length between nodes is not so easy.  Sure it can be programmed but such is moving beyond the parametric capabilities built into the graphical editor of the BIM software.

Still the people with the parametric software, can build the parametric models and then promote the speed with which they can then provide documentation for such buildings and structures. They can promote via the internet, but then someone has to be able to find the web site.

As indicated at the beginning I couldn’t find a site I knew existed, but then look at the site title. Last year the site title had $800 in it, now it has $900 in it. It comes up top of list for: “shed design 900”. But why would I be looking for that, I want to know the price of the shed, if I knew already I wouldn’t be looking.

Put simply fabricators rank higher in google search than the designers. Here, this site would appear to be ranking high for students, graduates and experienced professionals looking for free design tools. The site ranks number 1, in terms of both of the business names, so people who know us can find us. But the traditional 80% of our market, the once only, private individuals, who only provided 20% of our income, not sure they can find us. They are important, though not very helpful with respect to our survival, hence we moved onto the larger projects to get the income needed to survive.

But with changes in the economy, the marketplace, and the new technologies available it is now looking like private individuals can be assisted in shorter times frames, at the prices they can afford. Which in turn may increase the volume of private individuals pursuing such services beyond impositions from local council: like get development approval or pull that structure down.

Which highlights yet another issue. Sales of houses, highlighted people who had bought houses at a time when no legal requirement for real estate agent to check building approvals. This is turn led to requirement to seek development approvals for such, before could sell the house. Is the decline is such work and indicator that all such work has been cleared, and most houses now have up to date approvals? Or is it that there is a decline is housing sales, with people living in such houses not being in a position to sell and move?

Where is the structural work? Do I care? Should I just put the structural/mechanical design work aside and focus on industrial and manufacturing engineering? I doubt that will help as manufacturing seems to be declining or possibly more importantly failing to get started. Agriculture and mining what do these industries need? What does the world need?

As far as I know the world needs safe drinking water, and two international decades have failed to provide. The world also needs housing, schools and hospitals: and I’m relatively certain these can all be thrown off an assembly line faster than Toyota builds cars.

They just have to be designed so that they can be made, and it is preferable that the designs are parametric computer models so that they can be readily adapted.

Which reminds me the other day I was going to discuss the potential of Alltraders software and its potential for online services to the public which offer far greater scope than steel beam calculator, which I was also going to discuss in greater detail.

Some of the problems with some of the shed/canopy quoting and specification software is that it is constrained to using a manufacturer’s product. That is the users are locked into getting materials from a single supplier. Yet the software is largely just doing generic design. Here is a system I believe is not restricted to any manufacturer, and can be customised to suit the individual business: SZI Technology. But still the focus is manufacturers rather than the public. As far as I know the majority of manufacturers are small and employ less than 10 people, and most of them are likely immediate family and relatives.

The technology really needs to reduce in price and become more user friendly, and potentially operable from a mobile smart phone.

So I will be watching these spaces, and seeing how things develop.