3-D Printing a Jet Engine Component

General Electric posted a cool video of  jet engine fuel nozzles being fabricated–in one piece–with a 3-D printing process. Extensive data collection during the process is done for quality control purposes (they use the term “big data,” of which I am not overly fond.)

GE says:

Welders have monitored weld pools for centuries with shaded glasses, listening to the “bacon sizzle” of the molten metal, and later using infrared sensors, cameras, and pyrometers. GE is collecting all this data, as well as information from sensors checking the mechanical stability of the 3-D printing machines and the laser beams, and feeding it into algorithms that reduce terabytes of raw data to megabytes of useful information.

It seems that certain skills, such as understanding what is happening to molten metal via direct sensory perception, are becoming less important in this manufacturing process…other skills, surely, are becoming more important. It would be both interesting and worthwhile for someone to perform a multi-decade analysis of the actual skill mix required to produce a particular product.  For example, how does the set of skills that built the J-47 jet engine in the early 1950s compare with the set of skills for building the engines being produced today? Millions of words and trillions of pixels have been devoted..by academics, journalists, consultants, educators, and even the occasional practitioner…to talking about “jobs of the future,” but a high proportion of this writing and talking is of the hand-waving variety. It would be nice to see some serious historical (and quantitative) comparative research.

More on 3-D printing in today’s WSJ. Note that the Ford and Mattel examples are for 3-D printing of prototypes, not of actual customer products.


4 thoughts on “3-D Printing a Jet Engine Component”

  1. “… talking about “jobs of the future,” but a high proportion of this writing and talking is of the hand-waving variety …”

    One thing I am sure of: We do not know and cannot know what productive work people will be doing in the future.

    If you took a survey of knowledgeable people in 1813 and asked what people would be doing in 1853 and 1913, they’d be mostly wrong. The rate of change today is very rapid, far more rapid. What people will be doing even in twenty years is impossible to know.

    The idea of learning about the work practices of the past is a good one. Some academics do it, but most of the knowledge is lost. This is especially so in fields that relied on a lot of unwritten knowledge. A century ago, the entire country was built to accommodate the use of horses for traction and transport and possibly millions of people were engaged to one degree or other with the equine economy. Almost all of that know-how is gone.

  2. This is referred to as tribal knowledge. It’s a concept that has gotten a lot of attention in the aerospace industry lately. After Boeing outsourced much of the production of the Dreamliner, the delays and setbacks were blamed on the fact that they didn’t utilize all that unwritten experience embedded in the culture of the company.

    These big corporations don’t really like it because there is no easy way to systemize, quantity, process, or six-sigma-fy it.

    Now there is a big effort to develop tribal leadership, create cultures, and facilitate knowledge transfers. Unfortunately, the kind of wisdom they are trying to capture isn’t easily conjured and doesn’t seem to travel well.

  3. The worst cost/effort way to transfer that knowledge is to have an apprentice sitting there, observing and documenting the thing, sometimes more than one apprentice. It is so ungodly expensive, so much so that we reflexively do not consider this solution. That is a shame because as a society we really need to regain a sense tat there are things we are too poor to afford to do. We’re more people to understand that, we would have mastered a key prerequisite for employing the army of surplus labor at a living wage (a term of art that desperately needs its own tribal knowledge upgrade).

Comments are closed.