If this post is inappropriate for the forum, please remove it. I would completely understand.
This post is to share a small university project and to ask for donations through a Donorbox link.
We are currently preparing the fabrication and assembly of a gridshell structure approximately measuring 3 m (H) × 3 m (W) × 6 m (D). The installation is planned to be exhibited in Tokyo for about one month next July.
The fabrication and assembly costs are estimated to be around 7-900,000 JPY (6000$), and at the moment, we do not have sufficient funds to cover the full amount. I will contribute by myself, but it will still not be enough.
If anyone finds value in this project and would consider supporting it, even with a small personal contribution, I would be deeply grateful. You can donate through the link below.
(For companies or larger contributions, please contact me directly. I’ll work with the university’s accounting team to find an appropriate way.)
Here is the rendering of the completed design.
Let me describe the background and project history.
This project began in 2020 as research and development on form‑finding for tension–compression mixed shell structures. This problem is far more challenging than it may sound, but we eventually developed a stable computational approach capable of handling mixed stress distributions correctly. However, the method initially applied only to continuum shells. If such a solution (a surface) were to be realized as an architectural structure, it would typically take the form of a reinforced concrete shell. Reinforced concrete shells, however, are considered somewhat off‑trend today, while gridshells have been gaining attention.
For this reason, we explored ways to reinterpret a continuum form‑finding solution as a gridshell. Gridshells are typically covered with flat glass panels, and, at the same time, the structural grid ideally aligns with the principal (or conjugate) stress directions to minimize or eliminate bending moments. We identified an alignment condition that allows planar panelization (a conjugate net) to coincide with the principal or conjugate stress directions. In this way, a continuum shell can be converted into a gridshell whose planar panelization simultaneously serves as a bending‑free structural grid. With several key advancements (such as a 60× speed‑up in computation time, removal of GPU dependency, and support for more complex topologies), we are now testing a new (and hopefully final) feature that ensures all beams remain torsion‑free.
This research aims to contribute to the field of Architectural Geometry, whose foundations were established by Helmut Pottmann. As part of this research activity, we have published three papers at SIGGRAPH and SIGGRAPH Asia—the top conferences in computer graphics (four in total if we include our 2015 paper on pure‑compression shells).
In parallel with the development of the method and the codes, we have been creating smaller-scale models made of concrete, paper, etc. The scale is getting bigger year by year.
It started with a 1:500 tiny 3d printed model five years ago.
Then a 1:50 model made of concrete. (2024)
And here is a 1:10 model made of paper sheets and masking tape (2025). This model cost around $2,500, but we were able to cover the full amount through a research grant.
The prototype we plan to create next year will be a 1:5 scale model. This picture shows prototypes assembled for a feasibility test (prototyped with 3 mm MDF; the final assembly will use 5 mm MDF panels). I am currently applying for several research grants, but I am not very optimistic about the outcomes. Therefore, I would be grateful if anyone who finds value in this project would consider offering support.
Regards