Guestblog by Joris Peels, 3D printing strategy consultant
The Luxexcel team asked me to provide some information for people who are setting up their first 3D printing production lines or are commercializing their first 3D printed products. Many people are told that with 3D printing you can make anything and that it is easy. I’m sorry to have to rain on your parade but this is not the case. There are many design and process constraints with 3D printing and Additive Manufacturing processes. These constraints, problems and issues often differ greatly depending on the 3D printer used or the process used. For individual geometries and applications guidelines and roadblocks may differ significantly as well. Settings, ambient temperature and many other factors also play a part. On the whole however 3D printing is relatively straightforward. If one takes the time, designing and 3D printing a part on a desktop machine can be done by many people. Commercializing a product however and manufacturing it in a repeatable way according to standards is an entirely different matter. People with Engineering and 3D Modelling experience have a lot of advantages when starting out on such a project. Even they however often overlook important things to consider in 3D printing. I have over the years noticed many problems arising from a few key roadblocks that are often underappreciated in starting commercial 3D printing manufacturing projects. These roadblocks are actually fairly obvious for the most part. I’ve selected the ones that I see occurring most often of that do the most harm.
Redesign for 3D printing:
This has been said before many times by many people but there is a clear reason for this, it is probably the most overlooked single issue with 3D printing projects. In order to reap the maximum benefits from your 3D printed product, it must be redesigned for 3D printing. If you simply take a design made for a subtractive or mass manufacturing process and 3D print it then you will often have a comparatively overpriced and underperforming part. If however you make optimal use of the 3D printing process once you understand it, you can optimally design parts for use with 3D printing. This will make them less expensive and give them higher performance. The first GE 3D Printed titanium aircraft bracket below has no business case since it adopts the same design as traditional brackets but would require the introduction of a new manufacturing process. The second bracket however has the same strength but is much lighter in weight. This reduces the weight of the aircraft which means that the manufacturer can make a plane with a lower seat mile cost prompting airlines to wish to buy their plane over others. Both 3D printed parts, one is useless, the other potentially a game changer.
Your 3D prints are a box of chocolates. You never know what you’re gonna get.
3D printers often don’t have closed loop control. Processes are straight out of the box, unable to achieve a high degree of repeatability & reliability and give users sufficient process control to obtain true manufacturing level results. In order to create a process that does this you will have to integrate or build software and processes after you have developed an understanding of your manufacturing process.
QA & QC or FM.
Boring manufacturing things such as QA and QC processes have to be developed well in advance. Ideally being implemented as soon as you have a machine. You will need to 3D scan, surface scan, CT scan, measure or somehow after the fact check the dimensions & surface of your part. During the ramping up to production and the initial test runs these processes provide very valuable information to implementation teams. Too often however they are either seen as an afterthought, final step or implemented relatively late in the process. This lets teams miss out on important learnings concerning the material, machine, settings and process.
Tea, Earl Grey, Not.
Post processing is a required step with nearly all 3D printed parts and all processes. The amount of post processing and the amount of labor involved varies wildly per process and part. There is no completely automated post processing solution available at the moment. Your magical 3D printing journey into the future of manufacturing will involve men with carts walking from station to station. The wondrous adventureland that is 3D printing also includes men with brushes brushing parts clean or water jetting support material away. Typically the costs of a final part are 30% comprised of labor costs.
Do you have CAD people?
This may seem to be an obvious point. Often however organizations don’t have enough 3D modelling or CAD talent internally. CAD people are added late to teams, letting them and the team miss out on valuable learning opportunities. Often initially test parts are used but without the ability to iterate and improve parts quickly teams do not learn to master the process. One of the key advantages of 3D printing is the ability to do many iterations in a relatively short timespan to improve your product. This is a new organizational behavior and the quicker it is appreciated the quicker you will reap 3D printing’s benefits.
3D Printers, made for University Professors, sold to factories.
Traditionally industrial 3D printers or Additive Manufacturing machines have been sold to service bureaus and universities. For service bureaus, 3D printing was their bread and butter. They wanted to take machines apart and understand every element of their inner workings. It was their business and passion. They wanted lots of settings and had broad institutional knowledge that made them want to tweak their machines. University professors often wanted to hack, modify, change the machines they used in order for them to fit their research. This means that, the machine you are buying is very much lab and not fab. They can tell you a 100 times that it is a manufacturing machine that manufactures all your manufactures. This is not the case, yet.
If you want to do true manufacturing using 3D printers you must be able to service your own 3D printers.
David Leigh, a 3D printing pioneer who was one of the first people to industrialize 3D printing processes for Direct Digital Manufacturing has said it more directly and clearly than I could. When I interviewed him for the journal 3D Printing and Additive Manufacturing he stated, that in order to do Direct Digital Manufacturing, what is needed is, “Quality systems, manufacturing and engineering mindset, in house technical capabilities. Most service bureaus rely on the machine experts to maintain their machines – you can’t do that and maintain efficiencies in the production process.”
Nesting, not only for the birds.
One of the key business drivers for 3D printing businesses is machine utilization. The optimal placement of parts in every build you do is a critical consideration for any operator. There is software available to help you with this. But, there will almost always be a human element. There are many different considerations, some parts may fail if they have a different orientation in the printer. One part may look visually much more appealing if printed horizontally whilst the other would be better vertically. Parts may be stronger or have much lower surface quality depending on their orientation in the 3D printer. In some cases a large vase like part may be printed with many smaller parts in it, greatly increasing the productivity of that build. In other cases these parts may fuse or be unable to be extracted requiring a rebuild. Through years of experience many organizations have learned to make these decisions well. Especially if you are going to be making lots of completely different parts then the human element here is very important. If you make many thousands of similar parts, millions of hearing aids for example, then these will tend to be similar enough that software can automate much of this process. Even then however experience by operators and software people is still important.
Effective build volume is not the same as build volume.
Since orientation and nesting are such important facets in a 3D printing business calculating what parts cost and what a theoretical yield would be from a machine is quite complex. It differs greatly depending on the process, parts and the variability of parts. One critical flaw I’ve seen in business cases is that at one point, very early in the process, someone has input build volume numbers from a manufacturer. They look at their website and see that the printer is 200mm by 300mm by 200mm. They then calculate how fast the machine prints and then try to extrapolate a part yield from this. They then use this to underpin their entire business plan or business case. These numbers then carry from slide to slide and Excel to Excel. I call this the tears in the boardroom scenario. For some reason this is often not discovered until it is very late or too late. Once proper calculations are done the business case may completely fall apart. Space is often required between parts to stop them fusing together or to let them cool properly. Often space is required by parts on the edges of builds. Support may take up space. If you want to 3D print 30 Eiffel towers horizontally your business case may look wonderful. But, test late in the process and find out that they do not print properly in this orientation and this can be a shock. Find out that you now have to print them vertically and only four fit in the machine per build and it could be more than a shock.