Winning the competition

The spaghetti contest was after the standup architecture course, so thanks to this course we had already a very good geometry for this contest. But still this geometry was a theoretical one, building it is something different. We chose to build the geometry which would suit the best for making the bridge out of compression and tensile components which were optimized during the standup course to perform the best on both normal force and bending moments.

Finally the weight of the bridge was about 825gram and it carried about 54 kilogram, which was enough to win the competition. A funny fact is that the bridge didn't break on the spaghetti itself, the wooden strip actually broke. This wooden strip is given by the organization and holds the hook that carries the load.

But as you can see we're holding three cheques, the 1000 euro is for winning the strength competition. The 600 euro is for winning the beauty contest, we did that with another bridge which we built in one day:

The last 200 euro we one with a second price for making a specific assignment within 3 hours.

The assignment was making a tower of 50cm high as light as possible which should carry 4 kilograms, but it was prohibited to build in a square of 10 by 10 cm until 30cm high.

Finally our tower was the second lightest (2 gram difference) which could carry the 4 kilograms.

Improved Diana Results

Because I had some weird results with Diana I asked Andrew to look at the models. We discovered some odd mistakes in the Diana data file. These mistakes were actually very important for all results. So I had to adjust all Diana models I made and recalculate them. I managed to do that and I found quite reliable results. The conclusions out of these results are actually not very different than the earlier made conclusions, but now they are justified.

In short the conclusions from these tests are:

-Equal left/right division of tie rods improves the division of the stresses and therefore improves the performance.
-A wider curve than the standard spline gives more irregular results and it requires more material so its performance is not enough.
-Equal division on distance gives a more equal division of forces than an equal division on angle of tie rods.

The forthcoming days I will try to evaluate the models a bit better and formulate how the research should be continued to complete it. Next to that I will try to update the model.

Yesterdays Presentation

After the presentation I thought about Axels comment about losing the 3D aspect of the design. I mentioned that it was because of the constraints I made. The most important constraint is actually the contest context. This is defined by a point load in the middle of the span of the bridge. Such a load doesn’t really fit with a uniform element design. So you have to make a choice.
Constraints for a more uniform 3D design:
-Equally divided load over the whole span of the bridge.
-maximum of 4 to 6 elements attached in one node.
-lower maximum length of one element, for instance 5 centimetre.

I am more interested in the maximum structural performance of the bridge in the contest context so that is what I’m going to continue with.

Today I sharpened the model a little bit further, the upper part is more logical now.

The next days I’m going to try to find the perfect angle of the tie rods.

Practise with Grasshopper

Today I tried to make a normal bridge design with GrassHopper to improve my skills.

After a couple of hours I noticed it works al lot better for me than GC which was very hard to even make a start with a model. Eventually I managed to make a model where you can adjust the amount of tie rods and the thickness of all construction parts.

This model won't be useless, it is going to be my reference model to determine or another design is better. It could also be a good starting model to start improving on together with the input of the DIANA results about the 'block' and this standard bridge.

Constraints and research method

Contest constraints
First of all there is a resume of the contest constraints.

The bridge has to span 100cm horizontal and 42cm vertical. To make this span there is a construction space provided as shown in the figure. A vertical point load is applied in the middle of the bridge.

Material and physical constraints
A maximum of 600 gram spaghetti may be used.
Spaghetti strings have a maximum length of 20cm and a maximum of 3 strings may be glued parallel providing a string with a maximum length of also 20cm.
During the development of the process the constraints could change for several reasons, for example to make it more or less fragmented.

Research method

I’m going to start with a massive block and I would like to use the vectors of the stresses in this block for modelling the first pre-frame which is a new basis for the DIANA analysis. I check in Diana where problems or inefficiencies occur and I use that to remodel the frame into a more efficient one.

"Michell Structures"

The Generative Component practical last Thursday was very interesting, the possibilities for making complicated geometries seemed very usable for a model I would like to make for the geometry of the bridge.
But to make such a structure you need proper constraints and a theory to use as a base for its form.
Andrew and Eliza told me that in the first place all the forces in the bridge can be dissolved into the direction of the span of the bridge. So the spatial construction of the bridge is probably still going to be a quite 2D construction. But what the exact form of it is going to be is still quite hard to tell. Andrew put a reader online about special structures, in this by name the ‘Michell Structures’ are very interesting. These structures are defined by removing material until a specific wireframe remains. After this process he tries to find a mathematical solution that goes with each wireframe.
I’m going try to use this theory as a help to create the shape for the bridge.

But to make the geometry there are more constraints needed. With Andrew we also talked about the composition of several complex geometries. The best base for this is mostly still the triangle. So this is a clear constraint to begin with. Beside that the spaghetti has also a maximum length of about 20 centimetres.

The forthcoming days I’m going to practise with GC and I’m going to try to extract some data out of Diana that is useful to make a first shape.

Todays presentation

At the presentation I’ve got various comments.
To make sure the construction is better in 3D it should be constructed with a specified element and a specified situation which makes it inevitable to use a normal 2D construction.
The situation is going to be the spaghetti bridge contest situation.
To specify the element I’m going to determine a spaghetti beam with specified dimensions.
Out of this element the whole construction should be built.

Using one element

I would like to prove that 3D structures are better for bridges.
But how should I test this hypothesis? And what is better?
As mentioned earlier there are various forces applied on a bridge construction.
Very often these forces are treated separately with separate constructions.
To make the bridge rigid en stabile some tie rods are applied or other stabilising elements. The bridge construction should in my opinion be one construction which deals with all these forces.
But not all forces are the same, for pressure and tension there are often different elements used. And if one uses different elements one is actually separating the construction again. So how can we make it a whole spatial construction?
Well you could use one element of a specified material and specified dimensions which deals with both tension and compression. In this situation the composition of these elements is the key to solving all problems: Strength, rigidity and making the construction stabile. Making the composition the solution of the problem is what I call ‘better’ in this case.

But how could we test this? I think the spaghetti bridge contest is a good context to test this with. It has two clear problems:
-A point load in the middle.
-A height difference between the supports.
The bridge has to carry its own weight as much as possible; this demands a perfect optimization between the number of elements used and the load that the total construction can bear.

In brief:
Which construction out of one type of element is the best for carrying as much weight as possible proportionally to its own weight?

Possible research questions

This question is about the current spaghetti contest. This year the bridge has to span one meter but it also has to go 42cm upwards. To test the bridge there is a point load applied in the middle of the bridge. Again the bridge has to carry its own weight as many times as possible.
For more info (in Dutch): http://spaghetti.tudelft.nl/index.php
So the question is:
What is the best construction for this spaghetti bridge?
Preconditions next to the normal contest rules could be a restriction in two types of construction elements. A hollow tube constructed out of spaghetti for compression, and a massive spaghetti rod for tension.

Bridges have to carry several loads, its own weight, dynamic forces of vehicles, and some other smaller loads like for example wind. In the current bridge designs you can see that these forces are being dealt with separately making it a very 2D construction as mentioned earlier. And if not, the ‘integrated’ solution is always a very heavy but cheap concrete construction, which adds nothing to the landscape besides its function where it could be a real sculpture of structural design.
The question here is:
How do you design a lightweight integrated bridge construction?

Personal background and topics of interest

After completing secondary education I started with the Bachelor program of Architecture in Delft. Before I started with this I was already interested in and curious about how the structure of al kinds of objects performed and how they could be improved. As well because of this interest I’ve chosen to start with the Building Technology master. The first semester of building technology was a bit disjointed but I certainly liked the product design course. For now I would like to expand my design capabilities with this course about parametric design. And hopefully use this as a tool for my next designs.

The past years I’ve been competing in several contests where one had to make a bridge out of spaghetti which had to carry its own weight as much times as possible.
Another competition which I’ve attended was about making a portable pedestrian bridge which also was capable of making different spans to make it possible to use in every situation on Dutch train stations.
In both contests I noticed that the designs of al bridges were designed by putting some sort of truss or 2D arch construction. Subsequently calculating the stresses in this construction and when this was optimal the construction was copied and a floor was fitted in between these two trusses and maybe some simple straight on linkages were added and that’s it.
I think this can be done in another more efficient ‘3D’ way, with more efficient I mean stronger, lighter and better integrated functions.