4 ms·

Thanks for the technical perspective. > The signatures were likely added to the inside mould by engraving or CNC cutting them at the same time the texture was

Thanks for the technical perspective. > The signatures were likely added to the inside mould by engraving or CNC cutting them at the same time the texture was added to the outside mould. I'm also guessing that given the detail of each signature, this would have been digitally handled in RS-274 or a plotter language? The level of detail is so spot on that it's tempting to overlook the circa 1980s timeframe and what consumer manufacturing could do cost-effectively back in the day. Another observation is that none of the signatures overlap with the internal support ribs...almost as if the image was transferred in sections. I wonder if the original design team signed a silhouette of the rib pattern, or if the signatures were post-processed to eliminate collisions.

birdman3131 8y ago
Most likely standard gcode. We still run CNC machines from that era. At least one of our machines dates to 1979. While it would take a lot of tape there is nothing that I can see that makes those hard to do. (Ours no longer use tape but a drip feed from a parallel port.)
philpem 8y ago
They would have had a technical drawing of the top lid, which (if it was a mould drawing) would have shown the various ribs. Put a sheet of Vellum over the top, sign it, then use a graphics tablet (Summagraphics were around back then) to digitise it into a CAD workstation. Heck, if you're cheap, use two pots to make a two-piece digitiser arm and scan it that way (Mike Cook wrote an article in BBC Micro User back in the 80s about this). Measure the angles then do some geometry to figure out the Cartesian (X/Y) position of the arm. Create a point cloud, join the dots. Once you have the outline, creating GCode would be a relatively simple task (for a computer). Also I don't think the ribs are for support, though they will certainly do that. To me it looks like they're to allow the plastic to flow into the mould faster. Hunt down the "Engineer Guy" video on injection moulding if you're interested in the details. It's a fascinating process. Simple concept, but the devil is very much in the details.

birdman3131 8y ago

Most likely standard gcode. We still run CNC machines from that era. At least one of our machines dates to 1979. While it would take a lot of tape there is nothing that I can see that makes those hard to do. (Ours no longer use tape but a drip feed from a parallel port.)

philpem 8y ago

They would have had a technical drawing of the top lid, which (if it was a mould drawing) would have shown the various ribs. Put a sheet of Vellum over the top, sign it, then use a graphics tablet (Summagraphics were around back then) to digitise it into a CAD workstation. Heck, if you're cheap, use two pots to make a two-piece digitiser arm and scan it that way (Mike Cook wrote an article in BBC Micro User back in the 80s about this). Measure the angles then do some geometry to figure out the Cartesian (X/Y) position of the arm. Create a point cloud, join the dots. Once you have the outline, creating GCode would be a relatively simple task (for a computer). Also I don't think the ribs are for support, though they will certainly do that. To me it looks like they're to allow the plastic to flow into the mould faster. Hunt down the "Engineer Guy" video on injection moulding if you're interested in the details. It's a fascinating process. Simple concept, but the devil is very much in the details.