3D printing consultants, Hype and Hysterics

The article starts off with the basics concluding many people think 3D printing will disrupt 3D printing. Then the authors hits us with a growth curve chart of 3D printing. This is the basic setup of the professional amateur and ironically the same method that hypers use to persuade us in the reverse direction. Having the rhetorical style of a child or computer is not my concern so onto the main points:

 

Our heroes identify three hurdles

The chief constraints are economics, speed, and material science.

and just one paragraph later

it is likely to play an important role in most manufacturing operations over time. Companies that begin experimenting with the technology now will be positioned to utilize it successfully in the future. However, they should view additive manufacturing as part of a suite of advanced manufacturing tools that can improve performance, operational efficiency, quality, and the customer experience.

Which reads like a “if my first statement turns out to be wrong im not really wrong” Whatever the outcome, in 5 years these people will pat themselves on the back for their sagacity.

 

Economics. Additive-manufacturing materials are prohibitively expensive for most high-volume manufacturing applications, often more than offsetting any benefits that may be derived from any reduced labor that additive manufacturing confers. For example, thermoplastic materials used for traditional injection molding cost $2 to $3 per kilogram, whereas the corresponding photopolymers used in additive manufacturing cost anywhere from $100 to $300 per kilogram, according to a 2014 report by Wohlers Associates.

Industrial printers, which can cost hundreds of thousands of dollars each, add to the economic challenge and make the up-front investment for industrial applications substantial.

This is false equivalency or extreme stupidity. Injection molded plastics and photosensitive resins are so different in price because they are totally different materials used in totally different processes and applications. Those $300/kilo resins are used as a replacement for wax in the lost wax casting process. ABS is unsuitable for this application. These ‘photopolymers’ are ‘photopolymers’ for the exact reason that they are not extruded like a ‘thermoplastic’  The equivalence for 3D printing is this $2 to $3 kilo of ABS pellets vs $10 to 20 Kilo filaments and that is to an end consumer. A company serious about integrating 3D printing methods into their workflow will inevitably not being paying retail for their inputs.

As far as the cost of so-called industrial printers; the same can be said for any other industrial process. Industry and capital investment are bedfellows.

 

Material Availability and Performance. Today, 3-D objects can be printed from a wide range of polymers, paper, ceramics, composites, and alloys that include aluminum, nickel, chromium, and stainless steel. However, many specific alloys and compounds required for industrial applications are not yet available for additive manufacturing.

The durability and consistency of additively manufactured materials pose further concerns. Additively manufactured parts may not perform as well as those made with traditional methods. Titanium alloys used in additive manufacturing, for example, result in lower yields and tensile strengths than titanium alloys used in traditional methods.

This is bundling some truth with factual errors. It is true, 3D printing cannot print with all alloys and compounds. It is not true that 3D printed titanium alloys are weaker than traditional methods.

For example, one study compares the strength of a titanium alloy TiAI6V4

Ti6Al4V is a widely used biomaterial for many medical applications (Biomet, 2009), (Oshida, 2007), (Bronzino, 2006). Experimental results show that the properties of full-dense Ti64 processed on EBM fulfill the corresponding norm for medical implants (ISO, 2010), (ASTM, 2010) and are even superior to casted titanium alloys (Table 1).

 

Properties Norm (ISO Standards, 2010) Ti64 (EBM) Ti64 (cast) (ASTM, 2010)
Yield strength [Mpa] 760 849 825
Elongation [%] 10 15 10
Area reduction [%] 37 15-25
Young modulus [GPa] 125

So rather than 3D printed titanium alloys having lower yield and tensile strength than traditional methods such as casting, it is significantly stronger is some respects. Some 3D printed materials are weaker than their traditionally manufactured counterparts and there are many applications where you do not want to use a 3D printed part because it is too weak. None of this touches on the topic of integrating mesostructures or generative design into industrial components which would be a great boon to 3D printed parts mechanical performance in industry.

 

Speed. Industrial 3-D printers are much slower than the traditional high-volume manufacturing machinery. One example is plastic injection molding. According to our analysis, a traditional plastic-injection-molding system can produce nearly 26,000 parts per workday. By contrast, an additive-manufacturing laser-sintering machine can produce only 111 comparable parts per workday. In some cases, such as with aerospace engine parts, it can take two days to print a single object.

Perhaps this is the most opaque topic in consideration of 3D printing. It is interesting to note that both advocates and detractors are wholly polarized on this topic. Using the authors numbers, injection molding is 234.23 times faster than 3d printing. I won’t dispute this if we compare the time it takes to make one unit on an injection molded system and one unit on a 3D printer. But this completely misses the point of even talking about speed, the entire economic structure for the past several hundred years has been pushing towards a manufacturing scheme to leverage cheap labor combined with machine systems. But if I want to make a product and have it in my client’s hand tomorrow, 3D printing is certainly much faster since you bypass pretty much the entire supply chain and go from computer/raw data > local printer > client. The issue with speed is framing. If we talk about manufacturing a product as only the physical process of forming it, injection molding becomes really fast. If we talk about going from nothing to product, 3D printing is faster. So in many ways talking about specific numbers is a way to avoid the actual issue of speed.

Sometimes 3D printing will be faster, sometimes injection molding will be faster. Making a 1:1 comparison and ignoring all other factors is spectacularly feckless.

The rest of the article goes on to pitch their consulting business. So I suppose their attempts at a critical look at 3D printing were designed for posturing since they are so factually weak. However, if we take such middling positions we will be left watching as greater men, companies, and societies reap the first harvest of 3D printing.

One statement is great, one is pathetic

1986: The internet is going to change our lives

2016: The internet is going to change our lives

 

3D Printed Chess Set

I’ve never been very good at gift giving. Lacking Confucian ostentation and meticulous planning, gifts to my friends and family have always ranged from forgettable to bad. I have fond memories of playing Chess with my dad on a cheap old wooden set. When I saw the Duchamp chess set I was reminded of those times and felt a strange compunction to create something as a memorial to my time playing chess with him, and that old set.

chess-test

I’m not sure if I will be able to finish it before this Christmas but that has dissuaded me from starting now.

 

Immediately I knew I wanted something that would be a precious work by itself. Probably using a heavy mixed polymer like those tungsten and stainless steel mixtures but from my experience these have bad layer bonding and are weak to the effects of time. Quality was actually my secondary reason for going with SLA.

 

B9

 

The knight seemed like the obvious first piece to print since it is the most challenging. Thinking about it now though, the slender neck and the fatter head might be a issue. The printing was issue free and took roughly 5 hours. However, because this is DLP, I can fit about 2-3 pieces on one plate or if I print at 70 microns, 6-8 pieces. So roughly figure 2 hours per print at 30 microns, times the number total number of pieces, is 64 hours of printing time at 30 microns. Or around 10 hours of total printing time at 70 microns. I’ll decide after I get the first casting back.

 

chess-test1

It turned out well. I’m unsure how the lowish resolution of the model will turn out once it is casted. Looking at it in resin, it give a feeling of hand beaten copper. The larger base that you see on the bottom of the model is foundation support added during the slicing to help it stick to the plate. This might have been a stupid thing as it will require me to shave it off by hand along the transversal. For the next parts I will defiantly switch to point supports.

As for materials choice, Silver for white is a defiant as the characteristic shine matches white perfectly and platinum is too luxurious. I’m leaning towards brass for black but will have a talk with the caster and discuss some alloys that might be a better fit.

The next question will be if I print all the pieces or print one of each and create silicone molds and use wax injection for the molds then finally wax for the casting. The upside of going mold to wax to cast is less work for me. I don’t think, for this volume, that it will actually be faster because in the time it would take to make one mold I could have already printed half the total set. The upside of printing is that I control the process and it will probably be cheaper… I better fire up the printer.