Theoretically, in terms of potential, additive technologies can be perfect for the production of architectural models, successfully replacing the traditional, manual methods of making them. In fact, this topic is very complex, and the key challenge facing architects is the right way to design models and their optimization for 3D printing. Among companies providing services based on 3D printing, the scenario often repeats when they receive files that are completely unprintable, while the people or companies ordering the printouts have neither the knowledge of how to fix it, nor the money to pay someone for it, who knows how to do it?
This article was inspired by a discussion on some reprap forum, where one of the users asked for help in preparing an architectural model for 3D printing on a cheap Chinese 3D printer. Quite quickly it turned out that the essence of the problem was not the program settings or the device he he intended to 3D print the model, but the 3D model of the building itself, which was completely unsuitable for this. It was a classic case where practically everything was done wrong – a kind of quintessence of the inept design of architectural mock-ups for 3D printing.
This article is a contribution to a very wide issue, which is the proper design of architectural models in terms of additive technologies – it is not a tutorial or a list of good practices. Its task is to make architects (or their potential contractors) aware of what challenges they will have to face when thinking about using 3D printing in the production of mock-ups, what to pay special attention to and what mistakes to avoid? It was created on the basis of hundreds of conversations that I had with architects, architecture students and representatives of service and commercial companies who provided 3D printing services of mock-ups or sold 3D printers to architects. Here are the 7 most common mistakes we encounter during this type of conversation…
1. The architects that literally know nothing about 3D printing
In fact, it is not a mistake, but a problem… I am not going to comment on the quality of teaching of future architects at universities, because I do not have the slightest idea how does it look like out there? But I can say one thing: over the last seven years, I have had the opportunity to talk to nearly a hundred people specializing in architecture in the context of 3D printing. None of them had a clue what 3D printing is, how does a 3D printer work and what are the technological limitations of each additive methods? Every conversation or meeting was in fact a shortened training on the basics of 3D printing, during which I made my interlocutors aware of what to pay attention to when designing a model, or what and why it will never work on a 3D printer?
It can be said that the source of all problems in 3D printing in architecture is the state of knowledge of the architects who commission such prints. Lack of awareness of what additive technologies are and what their advantages and disadvantages are compared to other, classical manufacturing techniques, means that in the end a digital model is created that is not suitable for printing or requires enormous work to bring it to the level of feasibility. My other conclusion is that architects want to commission a 3D printout of a mock-up for not entirely rational and well-thought-out reasons – they do it very often because:
- “3D printing is the future”
- “others are already doing it”
- “we would like to try something different”
- “I saw something like that at the fair or YouTube and it looked great”
- “they say it’s much cheaper…?”
2. Everything in one file
A classic situation: we get a 3D model of a multi-floor building (let’s assume a four-floor tenement house), which has everything: perfectly reproduced exterior facade, perfectly reproduced each floor, perfectly reproduced interior equipment (furniture, fittings, staircases – sometimes in rooms or in corridors you can even meet people). And it’s all locked in one .STL file.
This type of design can theoretically be printed only in powder technology (CJP, MJF or SLS), but only on the condition of maintaining the appropriate scale and the resulting size of the smallest details (see points 4 and 5). In practice, it usually turns out to be impossible – in turn, much cheaper and more popular methods such as 3D printing from thermoplastics in a line (FDM / FFF) or from light-curing resins (SLA, DLP, LCD) are not suitable for this at all.
In order to be able to correctly print this type of mock-up, each floor would have to be exported as a separate file, as well as each of the interior or exterior façade elements (e.g. gutters and window sills). A legitimate question is what about the panes in the windows…? If they are in the project, why 3D print the interior of the building, if nothing can be seen through them? If they are to be transparent – can they be obtained in separate files and do they really have to be made on 3D printers?
To sum up – when we receive one such complex 3D file, in 99.99% of cases it means the end of talks about 3D printing. From the technological point of view, it will be impossible to perform properly – from the design point of view it will de facto mean creating the project from scratch.
3. Models are sets of surfaces
Another common case is when you upload models that consist of surfaces. I have personally had the opportunity to participate in four or five of the following talks:
- the basis of the order was a conceptual model consisting of several dozen simple solids imitating individual buildings, arranged on a board with an area of 2-6 square meters
- most of the solids were successfully placed on a 3D printer table with an area of 20 x 20 cm, and the rest could be divided and combined after 3D printing
- the material did not play a role
- models infill was “optional”
- the deadline was crazy.
During the initial talks, it seemed that it would be difficult to get a more banal order – the only problem that would seem to be solved was the rescaling of the machine park to several dozen 3D printers. Unfortunately, before we had the opportunity to go about solving it, another one appeared with the receipt of files for 3D printing – the solids were not solids, but sets of surfaces, single faces arranged in their shape. In addition, the surfaces had virtual thickness, tending to -∞.
As in the second case, it meant the end of the job. While we could undertake the proper mapping of the solids by drawing them from scratch, the principals were either not willing to pay extra for it, or it turned out that there was no time for it.
The same problem also applies to more complex models, such as the tenement house described in point two. It happens that the model consists only of external walls on which the entire facade is designed, while the body of the building is empty inside – or rather: it does not exist; it was not designed by anyone.
3D printing is based on solids. Even if we want to print only the walls of the building, they cannot be surfaces, but solids of a certain thickness / width (e.g. 1 mm).
4. Incorrect scale
Let’s get back to the model of the tenement house from the second point – the designer prepared it in such a scale that the body of the building would fit on the FDM / FFF 3D printer table with an area of 20 x 20 cm. The problem is that, as a result, the gutters have a diameter of 0.05 mm, while the 3D printer head, from which the plastic comes out, has a diameter of 0.4 mm, what means it is impossible to 3D print them at all. We also completely ignore the fact that it is supposed to be created vertically, while an incomparably better effect would be achieved if it was printed “flat” and glued to the finished building (see point 6).
A common mistake is to evenly scale buildings “down” so that they fit on a 3D printer, forgetting that apart from the model body itself, the dimensions of individual components are also reduced. As a result, their correct 3D printing becomes impossible.
How to solve this problem?
- maintain a reasonable scale and larger size of the mock-up
- scale the model’s body, but keep selected elements in a larger size (scale) only so that they can be reproduced on a 3D printer
- 3D print the solid itself, and make all small elements that would not come out on a 3D printer in a different technology and fit them to the ready mock-up (see point 7).
5. Too much details
The error derived from the points 2 and 4 – the designer knowing what are the limitations of a given technology – in our case of 3D printing, should know if and when a given detail will come out correctly, and when not? If the model consists of floors that are closed with walls and a roof without the possibility of seeing the interior of the building, the last thing that is needed in the project are tables and chairs, a fully equipped toilet and kitchen and door handles on the interior doors. Apart from the fact that such details are unnecessary in themselves, it is impossible to print them on a small scale on 3D printers (unless separately, but see point no.2).
6. One model = one print
A classic architectural model consists of at least several hundred small details, folded and glued together. In the case of additive technologies, architects go from the extreme to the extreme and recognize that the 3D printer should print the entire mock-up, preparing one comprehensive 3D model. This leads us to the above-mentioned problems, when it turns out that in order to make a mock-up correctly, a number of elements and details must be distinguished from it, for which there is usually no time or budget.
Meanwhile, at the very beginning, it is enough to think carefully about the project and prepare it taking into account all technological conditions to print the mock-up quickly and cheaply, and the end result would be really high.
7. No other manufacturing techniques
The last problem is the widespread lack of awareness of what 3D printing is – where are its advantages, and where is it unlikely to work? A 3D printer is a tool for relatively quick production of plastic parts, light-curing resins that imitate them, or metal powders. It is not perfect – it has as many weaknesses as a number of other devices and machines working in other technologies. If you want to print large, flat surfaces – e.g. walls, a 3D printer of the FDM / FFF type will be perfect. If you want to print small and precise details – facade finishing elements, windows, window sills, furniture or people – a SLA / DLP / LCD 3D printer will be perfect for this.
However, there are things that are best cut out of styrodur, wood or paper, or molded by hand from modeling materials or cast from resins.