0:00 All right. Hello everyone. I’m Luca. I’m a post-doctoral researcher at the Technical University of Denmark in Copenhagen. And today I’m going to talk about the stressbased design for 3D concrete printed u structural elements, material efficient structural elements. And I’ I’m trained as an architect but since the beginning almost of my studies I’ve been involved in research about 3D printing and in the design and the making of structures with 3D printing before during and now after my PhD.

The research I’m going to show today is mostly part of my PhD done at the University of Southern Denmark in collaboration with the Hyperion Robotics, Verber and Danish Fibers. And all my work starts from the notion that concrete is the most used material on earth after water. And this material creates large portions of the global emissions which are accounted to 7 to 8 8% and in this sense what I’m looking at is that the potential of design in improving to improve this issue and reduce reduce the the the emissions created by by the concrete industry and if you look here you we can see that large portions of these emissions are due to to the material itself.

1:49 So there’s a a ton of work that needs to be done on the material but at the same time there is quite some work that can be done to improve how we design so and how we build. So that’s where I’m placing my work and like like it was said before very well there are very ambitious and virtuous uses of 3D concrete printing. However in the industry what is is being looked at is size and efficiency in production.

So building bigger and building faster. However, I don’t I don’t think that’s what is really interesting about 3D printing. What is really interesting is looking at how we can improve and and reduce emissions in in the process. So u the big question is how do we design for 3D concrete printing to enhance the efficiency and sustainability of concrete structures and the the first challenge is that conventional construction processes are divided in separate boxes in geometry material and engineering and in the manufacturing part.

3:07 However, we need to start thinking more in a more intertwined way. So, we need to see fabrication, material and geometry as one process and 3D concrete printing needs that even more because all all the parameters that defines a ge the geometry are influenced and will influence both the fabrication and how we use the material. So the first thing I’ve worked on is understand the understanding the material and u the fabrication process.

So, I’ve I’ve worked on developing a a fabrication setup or better putting different pieces together. And I’ve been working with an industrial robot very similar to the one we have here and a feeding pump. And the material we’ve been using in our research is a mortar more than a a cement which is characterized by small small aggregates and at the same time we used fibers to avoid cracking and we also integrated the conventional rebars in the process.

3:36 So the first experiments we did some preliminary experiments were on understanding the early age properties of our material. So how does the material behave as soon as we deposited? How how much weight can we can we put on it. But we also looked at u how the material behave after printing and how it behaves in the different directions according to to the layer direction. And at the same time we did the same for the rebars and how the the behavior intention worked in the according to the direction of the print and the position of the rebar.

5:16 U this informed our modeling and the design preview setup. So we we wanted to have an accurate prediction of what was going to be printed and we we analyzed we used the printing speed as a as a parameter together with the layer height and we checked how how much the width of our print would change at different speeds and height. And as you can see there are some of these combinations that led to failure or to very very ugly prints.

5:57 And we also did the same for for the cross-sections. So we we saw that we had a range in which we could print a printing window. And we had different parameters and widths we could we could we could achieve by controlling accurately this the the printing speed and in relation to the height. So this became a tool a digital tool that we used to control and and optimize our our prints.

6:26 So we we tested it with some purely geometrical experiment and by changing the speed we could achieve a variable width in a in our print. And we tested this in a more structural application where we where we try to vide the speed in a column to see how we could achieve some thicker areas for for better stiffness. The the big case studies case study I’ve worked on is on horizontal structural elements on on beams and then on a slab and this is because 43% in a in a middle size concrete building 43% of the concrete is being used on horizontal structures and back in the 60s here is a bunch of examples from Puja Nervy we have been building buildings integrating structural performance to optimize the design of slabs of rib slabs.

However, this through by because of increasing costs in in manufacturing and in in the workforce, this stopped until new technologies of design and of fabrication allowed to revamp this this approach and especially in academia there’s been a lot of work of with for to use different fabrication approaches to optimize slab elements. Our approach, my approach was to try to work with the on a porous beam. So can we can the strength to weight ratio of a beam be enhanced by introducing porocity in its longitudinal section to reduce the the use of material.

8:33 And by paracity I mean literally creating voids in in my elements. And the first step was to compare a simple grid grid porous beam with a with a full beam. So these these are one to one elements we designed and and printed. They are 3 mters long and and they were reinforced with the with with the same amount the same volume of reverbs. And to to print the the grid beam, we had to come up with a with a way of optimizing again the tool path and the amount of material we we were employing.

9:16 And we did that by by creating a a woven layer structure where each second layer was printed sort of perpendicularly to the previous one and we had to work on the the speed of the robot to reduce the material in the intersection intersection. So here you can see how the the color varies because we are we are changing the speed to to adapt to to to the to the material we had below.

9:47 So here you can see one of the first printed specimens that was tested structurally in threepoint bending and to our surprise the you can see in yellow and brown that is the porous beam. To our surprise, that already behaved better than the than the full beam. And looking at the purely at the strength to weight ratio, we already achieved a 58% improvement. And we kept even though it’s it doesn’t give us a complete sense of of how much we are actually optimizing.

10:28 We we use this parameter as a as a compar comparison element but at the same time we also run a full life cycle assessment and u we saw that 70% of the average environmental impact was still due to the material as we saw in the in the previous in the in the diagram earlier. But we could also see that 45% of the of there was a reduction of 45% in the environmental impacts in the porous grid beam.

But we wanted to look at how we could integrate finite element analysis to to improve the way we we designed our beam and the way we printed our beam. So using a principal stress line field we extracted some principal stress lines and and analyzing those we could see we could generate compression pure compression theoretical pure compression and and pure tension lines that we used to print as our tool path but also to insert pre-bent rebars in the in the tension area.

11:42 And here you can see how we went from a linear finite element to getting and selecting principal stress lines. How those became layers and how then we adapted the speed again looking at the finite element analysis and at the fabrication constraints. And here again this is a a printed specimen. We tested a few of those and we kept one for the for the pictures and here we achieved 120% stren increase in strength to weight ratio compared to the to the full printed beam.

12:23 And then the last step I have a for the last step I have a short video. And and in this case we we on top of applying this porocity in the longitudinal section we also given the case of a of of a point load in the center we also optimize the the shape. So the external profile of of our bean and then the the rest of the process is is similar to the previous elements.

12:59 And here you can see how our digital workflow went from from designing the profile of the elements to selecting the stress lines and and using those to create a tool path and then reaching the the the point where we fabricated. And and this allowed allowed us to avoid any moment where we had to discretize our geometry. So we didn’t go through meshes but we went directly from from our curves from our stress lines to to the tool path and the fabrication process.

13:40 And here again the the result looking at only at the strength to weight ratio we achieved the 200% increase compared to the first full beam. And then the next step was trying to apply and to challenge our workflow and apply to to to a slab element. So in the same fashion we extracted our our stress lines in this case moment lines. We selected them looking at volume to deflection.

14:12 So we we did a range of options and we we looked at what what was printable and what gave us the best balance between volume and and deflection. And then we designed the depth of our ribed slab. This slab is 3.5 m in length and 1.5 1.6 6 m in in width and it’s supported in in two points and the load case includes a distributed load as well as a as a as a point load in the in the center and so we we describe the height using height maps as well as we then work using to reduce the shear in the in especially to reduce the shear in the port.

15:04 We also increased the width of our layers and this translated directly into motion speeds for our robot so that we could achieve the pre-planned heights and widths and also here we integrated different sets of of rebars and the process of including rebars cannot be shown because it was done manually while printing most of the times But what we also had to do here was to program the robot to print non-planer because we our ribs are not are not the thickness of our slab is not is not always equal but it’s it’s varying.

15:49 So we had to to work on our on our tool path so that we could achieve this by having variable layer height and to as Alexandra was showing earlier to optimize the placement placement of the material by being always perpendicular to the to the previous layer. And again this a couple of these elements were tested here following euro codes and we could see there was very there was no visible deflection in with the distributed load and then we we could achieve a a breaking point only by by loading it with our hydraulic press in in in its center.

16:32 And this is the the element that we that we printed. And so as a an outlook to to my research, what what I understood what we understood is that using tool path based design allows a direct transition to optimize the the manufacturing process. And at the same time we could see how weight reduction equals in a way carbon optimization and and again this is this can be done through shape optimization but also through a tool path optimization.

17:14 And so optimizing the the fabrication process itself and and in this sense also having a high resolution control even in such a large printing process can can help controlling the the the printing and improving the printing and and its its performance and what we also saw and what I also feel like it’s very crucial for these processes is to streamline more and more design to engineering ing and fabrication I would add to this workflows to to be able to to work between these three processes in a in a more integrated way.

And and then what what’s next in this sense is to to bring this is to bring this from from lab to construction. And of course that that entails a lot of things. But firstly is to try to understand how to make the process more consistent. And the second thing is to look at at codes and both how we can comply to those codes but also how we can challenge and make these codes comply to to what we are doing and to have ways of validating what we are doing that can can be embedded in in new codes and of course looking at sustainable material alternatives it’s it’s it’s key to to make extrusion 3D printing better.

Concrete printing in general and concrete construction more sustainable. And then of course looking at improving design engineering and manufacturing processes from an environmental an economic but also a a social point of view. Thank you very much. To learn more about the CDM computational design symposium series, to see the archives of previous presentations, and to learn about future events, visit CDFAM.com.