3d printers (Quigley Post)
Some 3D printer options for the designer.
Kevin Quigley runs his own successful product design agency in Shropshire and has been a first hand witness to the evolution of digital tools for product development — which includes but is in no way limited to 3D printing — over the last couple of decades. Never one to shy away from controversy or debate, in this guest blog post for PersonaliZe, Kevin’s views provide honest and realistic insight into how 3D printing, as a service, has contributed to his business and why, until now, he has resisted buying into the tech directly.......
The other day I was clearing out some debris from my office and I came across some prototype parts I had "3d printed" back in 1996 by a third party. I recall the project very well. It was a typical "mission impossible" type project for a customer that was really up against it. The company had a small in house design team who were working flat out on a major project when the proverbial brown stuff hit the fan on another project. Without naming the company or the product, what happened was that a system they had been developing over a two-year period had developed a critical fault during the final testing stages in the field. The problem for my customer was that the product was part of a larger system and their customer had threatened to pull out of a £0.5m contract if they could not resolve the issues with this one product.
I happened to be doing some other work for them at the time, and was on the premises when this came through so I was hauled into their design manager’s office and presented with a seemingly impossible task. Redesign the product, to make it work with the existing ancillary systems and correct the new found faults …. and it had to look like the existing one …. and it needed to be in a product trial in two weeks. As I said, mission impossible, but without the gadgets!
Bear in mind this was 1996 — before widespread use of 3D CAD, rapid prototyping or even CNC. This was a time when most prototypes were hand made.
The hard part for me was not the design side, or even getting the parts made, but rather that there was no margin for error — this had to work first time. So we set to work. We took the 2D Autocad files from the customer, imported them into our 3D CAD system, modelled up the existing system then used this to redesign the product. After a few days we thought we had a solution that would work, but the only way to know was to build it. Now this was a fairly complex system so the only option we had was to get the parts 3D printed.
Back then I used FDM almost exclusively for rapid prototyping — mainly because I knew a company that had a Stratasys FDM machine. So we saved out the STL file, couriered the data disk (yes, a disk; and yes, couriered) to Coventry and waited. A couple of days later I drove down and picked the parts up. Most designers can relate to this, but when you see your design, physically in 3D, and it all goes together perfectly, that definitely classifies as a good moment in your life! I was so pleased I called up the customer, said I was holding a prototype and it seemed to work. He asked me to come straight down so we could try it on the their factory test rig.
A few hours later I arrived and was led to the boardroom, where I was asked to demonstrate the product to the company MD and the problematic customer! Their customer picked it up, fiddled around with it, threw it down on the table, and a small exposed piece of the prototype snapped clean off. "What good is this he said — it breaks too easily!"
That, in a round about way, is the whole issue with 3D Printing. As soon as you go away from the comparative protection of the 2D screen or paper print to the 3D object, there is an expectation of functionality. It should work. Everyone can relate to a 3D object. There is no interpretation needed. There is nowhere to hide!
So moving forward 16 years, I look at that project, and wonder how differently I would have done it today? How easy would it have been to undertake? An interesting exercise (I thought).
In terms of design, whilst the process is the same, the reality is that it is far easier today, due mainly to the enhanced capability modern 3D CAD offers. We use SolidWorks now. Back then we used Form Z. Another factor is the internet, because rather than driving for three hours we can now hold web conferences to iron out details and share the screen — all of which increases speed and reduces cost.
Now for the prototyping. This is where I run into issues. Back then, for five FDM parts, the turnaround was two days. The parts were hand finished and perfectly smooth, well made and solid. These days I would probably use SLS for the fit and function phase, and I might get them back in two days, but probably not. I probably would not use FDM as a bought in service anymore for this type of project as SLS is stronger (in my experience). FDM as a service tends to be costly, and most off-the-machine parts have a distinct layering in them. I could get Objet parts back next day from a local supplier, and there is always the excellent Materialise Next Day service — but these options are not usually suitable for functional testing.
So where does that leave us?
On the prototyping side, I have more options now in terms of where I can get parts made, and what processes to use, but when you are right up against a deadline you always have to go with who has capacity and who has the process you need. For functional test parts that means SLS or FDM. In my experience SLS has superseded FDM as the process of choice for functional commercial bureau parts. Possibly because SLS is a more bureau friendly process in that you can nest components easily, or that there are no supports to dissolve. SLS is simply faster and more productive than FDM for bureau work.
I looked through my old invoices and found the bill from the RP company I used in 1996, then I costed the same parts with an online service today. Back in 1996 we had five parts made and finished in two days for £1350. Today the same parts as SLS would be £540 with a four-day delivery as a standard off machine finish. FDM (via an online supplier) offered £1100 with a five day delivery. Now I know I could have reduced the costs and delivery by shopping around, and, in truth, depending on who has jobs running I could have probably reduced cost by 50% and perhaps a 2–3 day turnaround. So, today’s costs are probably 25% of what they were in 1996, but delivery is much the same or worse.
So just buy a 3D printer?
Yes, having in house 3D printing is a solution to both cost and delivery but let's look at the options available right now and the environment the machine operates in. We work in a small open plan serviced office with a small kitchen/WC attached. This is the ideal environment for an "office friendly" 3D printer — or is it? If we look at what we want in a 3D printer — accuracy, speed, good surface finish, ease of use — and we ignore all the low cost consumer machines (more later), the starting price for "professional" machines is about £6600 for the new Stratsys Mojo. Next up after that you are very much into the £12–16k range, then up near or over the £20k mark. But the issue is that all those machines (other than the Mojo) require post build finishing, curing and/or cleaning facilities to finish the parts and remove support materials. All of which are unsuitable for a small office with no ability to access a plumbed-in solution, and, to be frank, I have ingested enough toxins in my life so far for the sake of design (Magic Markers rather than mushrooms I hasten to add!) that the thought of installing a cyanoacrylate infiltration unit for a ZCorp printer does not fill me with joy. Stratasys seems to have recognised this on the Mojo by making the support removal system a standalone kettle-like device. Simple, effective and exactly what this market needs.
But the bigger barrier to implementation is cost and build size. Looking back over our prototype builds in the last few years most could fit into a 250 mm cube and all would fit into a 500 x 500 x 300 mm platform — but that is a BIG machine. Yet another issue is not size but materials. Over 30% of our needs are in clear parts. This restricts the options yet again and drives up the price. I'm not doing well here….
So just buy a cheap 3D printer and compromise….
As SLS has risen to be the bureau system of choice, FDM has become the dominant process for the new low cost 3D printer market — most, if not all, low cost consumer machines use the process. So the obvious solution to speed is to buy a low cost 3D FDM printer? Right? No. I will explain why.
Getting back to my earlier point about 3D parts having nowhere to hide, this is the problem I have with low cost machines and the media hype around the whole explosion in 3D printing. Let's be brutally honest here. Most sub £2000 3D printers have small build volumes, course layer size (0.3mm min typically) and finished parts have poor surface finish. Compared with an off-the-machine SLS or SLA part, most cheaper machines are dire. I could not show these parts to a paying customer because they have become used to receiving a higher quality of component.
Even as an in house checking solution I think we would soon get fed up configuring a low cost machine and putting up with the compromises these bring and getting hold of print samples from low cost machines is difficult, and to be honest the ones I've seen don't impress me much. And yet, I do see some light at the end of the tunnel.
In the last 12 months there have been some very exciting developments in the 3D printer market. For me, the one with the most potential is the Blue Printer from Denmark, which promises an SHS process (like SLS but with a different heat source) in a reasonable build volume for under €10,000. Snapping on their heels is the Xeed from Leapfrog offering a wireless printing FDM machine with a very decent build size for under €6000. At this time I've not seen either machine in action, but I am assured they exist! Even further down the price range are the machines by Delta Micro Factory, which offer breakable support FDM for just over the £1000 mark.
One of the issues, as a buyer of this technology, is simply getting access to the devices to see them in action. Whilst I can call Objet, HP or Stratasys tomorrow and go to their offices to see a printer in action, this is not generally the case with the low cost options. This makes the buying process tricky. I am not buying this device for fun — it has a job to do, and I really don't have any desire to spend my evenings fiddling with extruders, platforms and stepper motors trying to get a low cost system tweaked to the stage it produces semi reasonable prints. I want the device to "just work", and just as important, to just keep on working.
So, despite my issues above, in the next few months I will likely be buying a 3D printer, and that choice will be dictated by the things I decide all purchases on — will it make my life easier? Not ROI, not quality, not cost. Will it make my working day simpler and more enjoyable. I'm not keen on being a guinea pig for new technology so the Blue Printer and Xeed may be termed as potentially risky options (but do me a deal guys — I'm prepared to beta test if the price is right :-) ). The Up! 3D printer offers fantastic value and seems robust, and to be frank, at that price I can take a punt. The Mojo would be, I suppose, the sensible choice, given the pedigree of the company that makes it and the availability of local support should things go wrong.
I will be looking at options in the next few months so I'll keep you posted!
Looking at the bigger picture….
So my rambling thoughts have made me consider the whole 3D printing media buzz that seems to be around at the moment. I "get" 3D printing. I understand the back-story to the hype. But that is all it is. Hype. Better people than me have said the same kind of thing but I honestly do not believe we will ever reach the Utopian state of 3D nirvana touted in the media when Joe Public fires up their 3D printer, selects a spare part for their washing machine and hits print. That is, to coin a phrase, total bollocks.
Of course there will be the techno geeks who will do this, in the same way that there are PC geeks now who enjoy delving into computer code, but that is not the mainstream market. There was a feature recently by some poor misinformed individual stating IKEA would be quaking in their corporate boots by the threat of 3D printers churning out wardrobes and other products. Well, no. If anything, IKEA will see this as an opportunity. Only organisations with the financial muscle IKEA has can really exploit this ‘print-your-own’ market. Rather than having individuals printing off bits you will have the IKEA website with configurable 3D printed products that you order online and collect in store. The parts may or may not be manufactured in store, but they would be collected in store — if only because IKEA know that when you are there you will partake in a few meatballs, wander around the warehouse and probably spend something to justify making the journey in the first place. You see, the whole 3D printed consumer thing amounts to this — a means to get you to spend more on "properly manufactured" parts.
The very harsh reality of 3D printed consumer products is that the part quality is rubbish. Take iPhone cases. I can buy 1001 different cases on Amazon for under £5, or I can spend £30+ on an SLS printed bespoke version that has all the quality of a 35 year old Triumph TR7. At the TCT event last year I was given a 3D printed pen, which was a spiral design that would have been hard to manufacture any other way. As a designer I recognised this and understood the reason for using the SLS process to make the part. The quality and finish of the pen was still poor compared to a moulded plastic version, but it was a compromise I was prepared to make given the complexity of the design. Yet if you look on Shapeways or any other online 3D printing "shop" most of the parts could be made using standard production methods. To me, that is what is wrong with consumer 3D printing. Tat is still tat, no matter how it is made.
I actually wonder if consumer personal 3D printing will bottom out in the next few years as more and more early adopters realise that part quality will not change much and running costs do not justify the experience. As the media continues to make bold claims I think you will start to see the IKEAs of this world begin to bring 3D printing into their offer, but only on their terms. Sure, your Makerbot-owning consumer can model up an IKEA vase and print it, but IKEA will start offering a bespoke 3D printed vase you configure on their site, pay for on their site and collect in store — and that vase will be printed in full colour incorporating your nearest and dearest's face on a state-of-the-art 3D printer. Given that scenario, 99% of consumers will take the IKEA route rather than the DIY version.
I am sure others will disagree and tell me I am talking nonsense. But ask yourself this. Do you really want or need a 3D printer in your house? I don't. I have enough junk without my kids printing out 101 bits of tacky plastic rubbish. At least with the IKEA type approach I suggest, you, the consumer has control over what you do best — how to spend the money. Not convinced? How many cheap inkjet printers or laser printers sit unused because their owners either cannot justify the cost of consumables, or they realise they can print their own photo books at Boots or online for a fraction of the cost ….. and that is a product where the quality was good!
So, 3D printing is a fantastic technology. For the designer it is an essential technology. For the consumer it offers things that cannot be done any other way. But I honestly do not see it ever becoming like the PC and being installed in most homes. Instead I see it as a bigger industry altogether based around businesses like IKEA, supermarkets etc. So despite all the media buzz around low cost printers ask yourself where is all the money? Where are the £10m investments in 3D printing companies? The money is not following the low-end machine market, it is following the high end production machines and the sales channels. This is not like the early days of the PC where companies like Apple grew from zero to billions in a matter of years. There is no IP (most of it is open source), there are many players in the market and most critical of all (I think) there is just not the consumer demand to spend £1000+ on a cheap printer, then another £50 a pop on consumables.
For business use, 3D printing makes perfect sense at the low end or the high end (depending on your needs, as I explain above). But for mass-market consumers? No.
NB: This blog post was originally submitted as a comment on Rachel's invitation post last week, but we thought it was too good (and too long) to go there.