You can hand-program a CNC lathe—but not a 3D printer

Getty Images

It seems to me that we take manufacturing software systems for granted. When I first stood in front of a CNC machine tool, computer-assisted programming didn’t exist. Scratch that. It probably existed, but our small shop sure didn’t have it.

We used our brains, a Teletype Model 33, and reels of paper tape to make our equipment behave as desired. For that matter, it was another five years or so until we got our first computer, an IBM PC clone that the shop owner’s kid installed in an effort to eliminate the coffee cans filled with paper job tickets. The shop’s first CAM system didn’t come along until a few years after that.

What do Teletypes and ancient computers have to do with additive manufacturing? Nothing much, actually, except for one thing: Unlike subtractive manufacturing—i.e., CNC lathes and machining centers—3D printers can’t be programmed by hand. I suppose it’s theoretically possible, but imagine the millions upon millions of code lines that would be required. For an LPBF (laser powder bed printer), for example, it would be the equivalent of milling out a deep pocket with an end mill a few thousandths of an inch in diameter with which you take passes a few thousands of an inch deep. Forget it.

No, there’s a symbiotic relationship formed by a 3D printer, its software systems, and build processing. What’s more, creation of the crazily complex part designs made possible with 3D printing relies on equally complex generative design and topology optimization software. There can’t be one without the other, in either example.

Industry 4.0 and the digital thread will only tighten this relationship. I’m no manufacturing prophet, but consider metrology provider Hexagon’s recent acquisition of Volume Graphics, a leader in industrial computed tomography (CT) scanning. How long will it take to close the loop on the 3D printing process by feeding post-build inspection data back into the build software, iteratively improving part quality with each subsequent build? The concept isn’t all that different than the in-machine probing systems currently used on CNC machine tools, albeit in a more automated, all-encompassing fashion.

Until that glorious day, additive equipment manufacturers will continue to improve their melt-pool monitoring and build-simulation capabilities, perhaps to the point that the CT scanning scenario just described will no longer be necessary. Perhaps the additive process will become so tightly controlled and predictable that “final inspection” of 3D-printed parts will no longer be necessary.

Speaking for myself, I was never able to achieve that level of part quality all those years ago, but who knows? Whatever else happens, the software that is already so integral to the 3D printing process will only become more so in the months and years to come.