Hacker News new | comments | show | ask | jobs | submit login
USC makes a robot that manufactures housing from a CAD file (contourcrafting.org)
18 points by alaskamiller on Sept 6, 2008 | hide | past | web | favorite | 6 comments

As a building architect, the system looks fascinating. However, the biggest problem I see is reinforcing - concrete performs very poorly in tension without steel (usually rebar) reinforcing, and in general is not allowed by most building codes. One solution would be to have the robot build a double wall that effectively acts as formwork for the real structural wall in the resultant cavity. The structural concrete would have to be poured after reinforcing is placed in the cavity, and structural continuity would need to be achieved by the rebar connecting to rebar in the foundation as well. The other approach that would work better for the robot would be a radical improvement in concrete technology that would boost the tensile strength without the need for rebar - that would be ideal.

I believe that you can avoid rebar when using CMU's (cinderblocks), provided you use a surface bonded adhesive. Possibly a similar coating could be applied to these walls, providing 100% waterproofing, mold resistance, and strength. Alternatively just make the wall wider at the base if doing a single story building - if the width is 10% of the height of the wall, it will be OK.

The robot could embed steel or kevlar tape pretensioned during the pour.

The main problem is the vertical reinforcing bars - unless, of course, the robot was redesigned to build the walls in vertical passes instead of horizontal. But then you have to cope with the horizontal reinforcing - I think it is a difficult problem which they haven't solved yet. The only clean solution I can think of is a new kind of concrete that has tensile strength - maybe there is a nanotechnology solution to this.

We see inventions like this every few weeks that have the potential to "completely revolutionize" a field (e.g., long-life batteries, 3D printers, etc.). The notable examples of ones that have come into fruition that I can think of are drinking straws (filtering on-the-spot) and the OLPC project.

It always frustrates me to see things like this so undervalued. To me, it seems something like this could have a societal impact on the same order-of-magnitude as the internet, yet the story only has 16 upvotes, and it will probably take over two decades, if not longer, for this manufacturing system to materialize in, for example, my neighborhood. Why is that? I realize there are certain costs to introducing it and it does have its risk (and some problems, as gibsonf1 mentioned), but wouldn't this be a marvelous investment not just financially, but in general for humankind? I suppose the point of my ramble is to raise the question: why is viral marketing so hard to perform in "real life"?

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact