[reprap-dev] Zero Touch Display -> very simple 3d scanning

reprap-dev at bodgeitquick.co.uk reprap-dev at bodgeitquick.co.uk
Mon May 16 11:07:51 PDT 2011


Hi Reece,

    This is a very old yet familiar concept to me I was working on  
this in the early 80s. Full details can be seen in the patent GB  
2133537.

  Priority claimed: 16.12.1982 in United Kingdom - doc: 8235896
     Patent ceased on 08.12.1993

It has a simple drawing of the infrared beam array.. which is the best  
way to see how it works I think.

    It is based on the shadow of the object created by many beams  
becoming broken. Watching simple un-processed data from the receivers  
gives a visual effect similar to the hole graphic image in this clip.

    The position of objects or any 2 beams crossing point being  
calculated by simultaneous equation. The positional data was provided  
every 20ms using 64bit fixed point calculations on an 8bit 64180 CPU  
the whole system including CPU and Power supply fitted on a 32mm wide  
PCB fitted around the touch area.

    I used two finger touch and touch negation for interlock control  
on Power  stations it also allowed for 15% failure of any of the  
infrared beams.

    These touch screens were retro fitted onto 9" 19" and 21" monitors  
in Power station control rooms connected to an array of PDP11 via RS422





Quoting Reece Arnott <reece.arnott at gmail.com>:

> http://www.youtube.com/watch?v=wY-AIEfIIeY
> http://ecologylab.net/research/zerotouch/index.html
>
> This is a rectangular set of infrared LEDs and sensors that are set  
> up for giving back multi-touch information to turn any screen into a  
> multi-touch device but I'm more interested in the possibility that  
> it could be relatively easily set up to give visual 2d slice  
> information for an object.
>
> Imagine a relatively simple enclosure that just has this frame set  
> up so it can go up and down controlled by a stepper motor/threaded  
> rod attachment similar to the Makerbot or Reprap z axis. You simply  
> place you object to be scanned inside this and move the frame up in  
> known increments.
>
> At each step in the z direction you get a cross-section of the  
> object you are scanning. If you step in the same amount as you would  
> slice the z layers in the Reprap (it may even be able to be  
> integrated to be the same object)
> you can shortcut the process of "3d photocopying" and simply produce  
> a set of 2d polygons for each slice and then go straight to the  
> gcode to draw them. If you wanted/needed to produce a STL file of a  
> bunch of triangles then you could simply extrude these 2d polygons  
> into the z direction as needed (I already had the code to do this in  
> my Carapace Copier software for testing purposes but I took it out  
> when I was finished with it - I kept it in a large unused_code.txt  
> file though if anyone really wanted it :-)
>
> Its probably just a little bit late to be changing the focus of my  
> PhD as in my current plan I'm starting to writeup in the next couple  
> of months but I think it would be really cool if someone took a look  
> at this with the idea of "low cost 3d scanning + low cost 3d  
> printing = low cost 3d photocopying" in mind.
>
> -- 
> -------------------------
> Reece Arnott
> University of Otago
> Dunedin
> New Zealand
> -------------------------
>
>
> _______________________________________________
> reprap-dev mailing list
> reprap-dev at lists.reprap.org
> http://reprap.org/mailman/listinfo/reprap-dev
>






More information about the reprap-dev mailing list