One of the challenges in 3D printing is making it cheaper and more affordable. There are a few ways to do that. Mostly it boils down to reducing cost, increasing efficiency and increasing output. One of the major drivers of increasing the output is the speed of the printing process. In other words, how fast can the 3D printer produce the products. But how do you do that?
One of the intriguing facts is that all printers do not have – or only extremely limited – closed loop control. In essence, the 3D printers cannot self correct. It makes the whole printing process somewhat of a black art. Just ask any 3D printer machine operator and he will tell you. The best results are produced when the machine is properly cleaned, machine and powder are in a humidity / temperature controlled environment and the builds / trays are packed in a certain way. Then still a stray human hair inside the powder can ruin a whole build.
Then I read this article on Wired today. It is a 3D printer build in the research lab of Pieter Sijpkes. He built an ice printer. The idea and execution is fascinating, but then my eye fell on this: “After every five layers have been deposited, a laser-displacement system measures the geometry of the top layer and adjusts the valve-control data to correct for any errors.”.
I thought “Yes, that sounds like closed loop control!”. I am sure he needed it because the robot arm is probably not terribly precise otherwise, but I love it anyway. I wish that manufacturers would pay more attention to closed loop control in the design of their machines. I am confident it increases the reliability of the 3D printers tremendously and makes it possible to print faster. Both will lower the price of 3D printing – while reducing the frustration of machine operators when they have yet another failed build.
I love to hear opinions on why closed loop control is so hard for 3D printers and any proposals or ideas on how it could be implemented!