E2C2

E2C2

energy / environment strategies

E2C2 RSS Feed
 
 
 
 

https://www.euro-online.org/enog/inoc2007/Papers/mac www.euro-online.org download

Posts tagged additive manufacturing

A Short History of Making

the promise of additive manufacturing

In the beginning, the consumer was also the producer. People made simple tools and weapons and used them to survive. Gradually, consumer and producer grew apart. Nowadays, many of the objects we use are made half a world away. But, this could change, with the advent of a disruptive technology. Consumers and producers may once again be in close proximity.

Many of you will recognize the pictured item as created with a 3D printer. This technology is also called additive manufacturing, since very thin layers of material are added successively to create a three dimensional shape.  Additive manufacturing promises to transform the way many things are made and in so doing change sizable portions of the economy.

Briefly, this is what additive manufacturing promises:

  • Accelerates product development by rapid prototyping
  • Renders small run and custom parts feasible by virtue of no tooling
  • Makes heretofore impossibly complex shapes and assemblies buildable
  • Shortens supply chains by placing manufacturing closer to the consumer

While not all of these objectives have been fully achieved, 3D printing has become increasingly cheap and ubiquitous. Read other blog posts here to review my experience with additive manufacturing. As with most new technologies, the development path is not entirely smooth, but the upside is huge. The diagram below sketches a short history – and likely – future of making. For a larger printable version of the diagram click the pdf link.

pdf of  A Short History (and future) of Making

Economics of Additive Manufacturing

I’ve written previously about the advantages of digital fabrication and additive manufacturing (3D printing) specifically. But how do the economics for 3 D printing stack up in the real world against conventional manufacturing techniques such as injection molding.

Summarizing 3D printing selling points:

  1. Ability to print shapes and assemblies likely impossible with other methods – such as intricately formed or nested shapes and structures.
  2. Ease of creating custom shaped objects to meet individual parameters – such as ergonomically tailored sports or medical devices.
  3. Ability to manufacture low volume, but high value objects cost effectively by eliminating expensive tooling and molding.

3D printing offers some clear advantages in the instances one and two above. For instance one, when there are no other physically feasible manufacturing options for a particular form, 3D printing is the only choice. Printing simply does what cannot be accomplished by other means.

Instance number two is somewhat similar. A custom formed object/device fits an individual perfectly and will be produced only once or at most a few times for that person. A conventionally injection molded piece might be too expensive, given the high costs of molds. 3D printing has a clear advantage in both these cases. But what about case three – an anticipated low volume run or a situation where it doesn’t make sense to invest in sizable opening inventories?

Let’s look at an example. In two previous posts I described digital fabrication of a moderately complex lamp of my design, using 3D printed plastic and laser cut acrylic parts. The lamp is made of a central hub which holds the electrical socket and has twisted fins that extend to attach laser cut pieces comprising the lamp shade. (see the white finned hub component in the picture below). What would a comparatively low volume conventional, injection molded hub piece cost, versus the 3D printed version?

Fortunately, it’s now possible to get online quotes for both injection molded and 3D printed parts. I used www.IcoMold.com to estimate injection molded hub pieces and www.Shapeways.com for 3D printing estimates.

3D printing vs. injection molding

Total manufacturing costs, for selected quantities from 1 – 75 units are shown above. Costs exclude design and shipping and compare injection molding (red line) and 3D printing (blue line). The major cost with injection molding is the mold, itself, which in this case, costs about $8,500. In a convenient statistical breakoff, manufacturing runs of less than 50 units yield a lower total cost than for 3D printing.  As can be seen, there are few economies of scale in printed parts, aside from perhaps amortizing design and reducing inventory costs. Unlike molding, the 75th part costs as much as to produce as the first unit. Having said this, manufacturing runs greater than 50 may also be cost effective for 3D printing when factoring in inventory/stocking costs.

What does this analysis suggest? Extremely limited volume and custom, one-off, high value parts and products will be increasingly produced with 3D printing. We can also expect that as 3D printing costs decline – as they surely will – more and more spare parts and limited production parts will be fabricated by printing. Another important cost factor not included here is that transportation costs – and carbon footprints – should be reduced. Many injection molded parts are made in China, requiring extended supply chains and logistics which are costly in both financial terms and environmental impact. Printed parts are also easier on cash flows. With greatly reduced upfront manufacturing costs, entrepreneurs can invest in products with less trepidation.

A Longer and More Winding Road to Personal 3D Printing?

A Mixed Bag of Making

I have touted the prospects for digital fabrication and more specifically the economic benefits of the additive manufacturing revolution enabled by 3D printing. Great strides have been made to bring 3D printing out of computer lab and into the R&D shop. Thousands of firms now use printers costing from $15,000 to the millions to create rapid prototypes and finished parts. But beyond the computer lab, hacker space and R&D department how does printing fare when released in the do it yourselfer’s shop?

Options for 3D Printing

While it’s quite easy to print readymade files and use fabbing service bureaus to outsource prints, my preliminary verdict from personal experience and some frustration – is that there is a ways to go yet before personal 3D modeling and printing is simple, effective and ubiquitous. I suggest the situation with 3D printing is a bit like personal computing about 30 years ago and I’ll explain why this is and what’s needed to transform the industry.

First let’s look at the Maker’s options for 3D printing assuming you don’t work in the rapid prototyping industry or a computer or fab lab type of environment that has access to experts, high end software and hardware. Let me also note that you don’t have to be creating your own models and printing them on your own printer to be involved with making. However, many, if not all creative types will at some point want to create and print their own 3D designs, if not in their own office or shop, then somewhere local that may not possess a high level of expertise.

Fabbing Options Now (also see diagram above):

  1. Existing Model Printed at Service Bureau: Find an existing free model on the Internet and send it to a fabrication service like Shapeways or Ponoko (or local fab lab) who will print and ship the finished product back to you in a couple weeks. There are thousands of models available. Presumably as the inventory of models grows there will be ones that meet many aesthetic and functional needs.
  2. Existing Model Printed Locally: As with method above download a freely available model and print it on your own printer or at a community fab lab, where you might get some help.
  3. Create Model and Print with Service Bureau: Create your own design with 3D software and send it off to a local or cloud based service bureau. This assumes your model is valid for printing. Some services like Ponoko and Shapeways are willing to help evaluate and repair a printable file for a modest upcharge or subsciption.
  4. Create Model and Print Locally: Design it and make it on your own studio/shop’s printer. I assume many makers want to do this; personal expression drives making. And this is where things get tricky.

Not Quite a Personal Factory – Yet

Wishing to investigate the personal end of 3D printing I  purchased a fully assembled Makerbot Replicator www.makerbot.com and it was delivered about a month ago. Makerbot has been the poster child for 3D printing and has been featured in dozens of publications and TV shows during the past year. The new Replicator model is a vastly improved over the version I purchased and built from a kit about two and a half years ago. I took the Replicator out of the box and printed a preloaded 3d model file from a supplied SD card in less than an hour. Alternatively, I could have downloaded a free prepared model file from the Internet (www.thingiverse.com) and printed it from the SD card that slides into the Replicator’s SD slot. Once you have a model file you don’t even need a computer to run the Replicator – an on board touch pad and LCD display the necessary menus for printing. This is all pretty easy.

But what if you want to make something of your own design? Curiosity, self expression and problem solving drives creative types and inventors They want to solve a unique need they’ve identified or express their own creativity; not just copy someone else’ model. This is where 3D printing becomes considerably more complicated. Printing a unique, freshly minted 3D design still takes considerable effort and most likely a lot of trial and error.

Is It Like 1982?

I’ve previously suggested that the current status of DIY 3D printing may be like that of the personal computer 30 years ago. In 1982 you could buy an IBM PC or an Apple but it lacked a graphic user interface and a mouse. Word processing software existed in a couple of text only applications such as Wordstar or Wordperfect and Visicalc was the only spreadsheet. Printing was a rudimentary, dot matrix affair. The introduction of the Mac in 1984 began to revolutionize the situation, with graphical user interface, mouse and well integrated word processing and other applications. We know how quickly things changed after that.

The fundamental problem with 3D printing now is that it’s not a simple process to get from the model to a printable file. The model file needs to be translated into a fully closed triangular mesh. Imagine you must represent a solid volume, your model, with many small triangular shaped pieces of paper that are cut and glued together, creating an exterior shell of your solid object or assembly. The resulting construct, called a stereolithography (stl) file, should be a “watertight” model or it can’t be parsed in many thin layers and printed.

The opportunities for holes and other errors not readily apparent to the eye become rapidly clear in such a construct when considering any shape more than a simple box or sphere. And fixing the holes can be a confounding job. I, still, have not been able to create printable files from two models of my existing portfolio, after many iterations and the deployment of a program designed explicitly to find and fix such problems. I expect more research and effort on my part will resolve the problems, but if the personal printing industry is to explode then these problems need to be addressed with innovative products.

What’s Needed

While a number of good 3D modeling applications are available, in both the paid and free categories, none I’ve come across offers a completely smooth and foolproof workflow from model creation to print file generation. Some helper applications offer model file fixes but from my experience, they still overlook issues that turn up when compiling the code (G code) instructions for the 3D printer. There’s no doubt I need to become a better modeler for 3D printing. However, the Maker world of the near future should give me the option of printing my objects as simply as sending this page to the printer.

Factory in a Cloud; Part 2

Fablamp combines 3D printing with 2D laser cut acrylicIn the previous post I outlined the prospect for cloud based do it yourself (DIY) digital manufacturing. I described a test of this capability using the fabrication (fabbing) service, Ponoko, to make the parts for a lamp of my design, using laser cut acrylic sheet and 3D printed polymer. My goal for the project was not to create a production ready consumer product. Instead it was to test a process of prototyping and potential manufacturing, using remote resources. This post picks up on the fabbing process, once my design files had been submitted and accepted by Ponoko.

Delivered Goods:

I received the flat laser cut pieces (purple plastic in the photos) just a week after my order. They met my expectations, with accurate shapes and nearly smooth, polished edges. However, it took almost 3 weeks to receive the 3D printed polymer housing for the LED lamp (the white finned piece in photos). When this part arrived I was reminded that my knowledge of file preparation for fabbing leaves something to be desired. In translating the computer model’s smooth twisted surfaces of the housing “blades” for printing I had over – simplified the file, resulting in faceted rather than smooth curves. This was not my original design intent but on second consideration the faceting creates an interesting texture. The durable polymer material I had chosen has a slightly rough, but not unpleasant, texture. Since I wasn’t certain of the exact attachment point of the flat lamp leaves onto the lamp housing housing I had left the attachment holes off the laser cut pieces; opting instead to measure and drill these holes in my shop. Once drilled, the 1/4 inch holes in the leaves press – fit perfectly over the “buttons” I had cast into the lamp housing. From there, the lamp was easy to assemble and wire.

Results:

I now have a lamp prototype that cost about $250 and took 3 weeks to complete, not counting computer modeling. I could not have easily created this object using conventional methods or materials. In theory I could have hand cut and finished the leaves of the lamp on a jigsaw and drill press but this would have taken hours and the accuracy would have been nowhere near perfect. Perhaps I could have carved or molded the central lamp housing out of plaster or polymer clay but this would also have been time consuming and again, the accuracy would not have been good. So, $250 for an accurate, attractive, working prototype is pretty cheap.

Had I been more knowledgeable about computer modeling and file preparation for printing I could have achieved a more finished piece for the lamp housing. Moreover, there is a range of materials available; Ponoko offers smooth, shiny finishes in a variety of colors and materials including ceramic. So, in theory, achieving a consumer ready product is not beyond reach. I should also mention that Ponoko is not the only option for outsourced fabbing; Shapeways offers similar service.

Lessons:

What’s to be learned from this experience, over and above surmounting the technical requirements for making finished parts that exactly match your expectations.

Not Exactly Rapid Prototyping; A three week wait for a prototype is too long. Prototyping usually depends on fast iteration. I may have been able to shop around and find a quicker service but I doubt I could have found one that could have delivered a part in less than a week under $250. By comparison, Makerbot Industry’s new Replicator promises an out of the box 3D Printer for about $2000. If the print quality is near the quality of the part I ordered then about 8 more prints from a fabbing service would be the equivalent of a purchase.

Manufacturing Is a Possibility; Even with a long delivery time certain types of custom goods could be outsourced to a manufacturing platform like Ponoko. Nike’s custom shoe program promises delivery in 3 – 4weeks. A quick scan of Ponoko’s inventory of design offerings by various makers suggests that jewelry, small housewares and home furnishings are popular areas. Can one compete with IKEA or Target on housewares and furnishings; no. However, perhaps more fair comparisons are Design Within Reach and other high end purveyors of artisanal modern home furnishings sold in fairly small quantities.

Additive Manufacturing; Unlike laser cutting or CNC routing which is basically just a faster and more accurate means of cutting or carving away something that can already be done by hand or less automated machines- and these are no small feats – 3D printing allows the making of shapes and assemblies that might be otherwise difficult if not impossible to create. Complex nested geometries that mimic biological structures are possible. Now, people offer elaborate, biomorphic jewelry pieces, printed in materials including precious metals. However, before long printing of human body parts and organs is likely. For now, beyond prototyping, jewelry and luxury furnishings, additive manufacturing favors small quantity, high value parts and assemblies in the medical and aerospace industries. What other niches cry to be filled?

mission

We develop and market energy efficiency strategies and technologies. We focus on the building and transportation sectors, which account for more than two thirds of the energy budget.

contents