When you mold any piece, you end up with a solid block at first. Then you must slowly make cut after cut to remove your prototype piece. You end up with two halves and a hollow space that, when the two halves are put together a portal opens and... well no, but you do get the 3D shape you intended to mold.
Round 1 of molding these robots (way back last fall) taught me quite a bit about molding complex bots with lots of small parts. Of the four designs, I had made molds for two of them. I quickly learned to be sure to have enough of an opening so that pouring resin into the mold was not a nightmare. The most important thing I learned was the need to add extensions to any small protrusions (through the use of diced up tooth picks) to give some extra space for air to be pushed. There is obviously air in the mold when you start pouring and, inevitably, that air needs somewhere to go. Adding small extensions to parts where this is probable (limbs, fingers claws, etc) gives that air a place of its own. Sure when you pour in the resin, some of it will fill those extensions and need to be cut away later. But better to have an extra bit of flak to cut off of a robot's claw then ending up with a short claw due to an air pocket.
Round 2 of molding this week yielded a couple of perfect pieces and a couple that, due to their complexity, weren't delivering consistent results. Rather than spend tons of time (and resin) modifying and modifying the mold after the fact, I'm simply going to re pour the molds on two of them. I've sketched out the optimal venting that needs to be in place. I'm confident the next molds for them will be easier to cut in half in addition to yielding perfect pieces every time.
Its a slight set back, but a necessary step in the learning process regardless. I'll be sure to factor in the molding process ahead of time when I start designs for Wave 2 of 9th circle Robotics after Wave 1's launch.
And despite what you might read in the press about our molding process, 9th Circle Robotics prototypes Gibson and Aldiss-327 DID NOT go rogue during an attempted molding and they most certainly did not kill 36 of the 42 members of the production floor staff. Further claims that they were only subdued after sustaining heavy damage when a functioning Underhill unit fired upon them with his cannon, resulting in the need to undergo a slight redesign before round 3 of molding, are ALSO ridiculous and unfounded. 9th Circle Robotics planned for a bit of redesigning between molding sessions. The shared legal team of 9th Circle Robotics and its parent company Mephisto Designs is quite confident they will be able to get the courts to see things their way and all charges will be dropped. The public has nothing to worry about, these robots will arrive at every doorstep in the country on schedule.