3D printing in the home: Reality Check

03 December 2012

While working on his home office this month Al Dean has been pondering the growth in awareness of 3D printing. With the mainstream media talking 3D printers in the home again, is a reality check in order?

There’s nothing like spending evenings and weekend doing carpentry to focus the mind. There’s something about the whir of a chop saw or clack of a torque ratchet to get your mind wandering while you carry out a task that’s more physically than mentally demanding.  I found myself sharpening up a chisel this weekend and realised that I’d broken my jig some time ago. Of course, my first thought is, I’ll wing it.

Then I got to thinking about how nice it would be to have a 3D printer again so I could model a jig up, print it and do the job properly. I also thought about how there’s been so much talk of 3D printing revolutionising the world these days.
Only this month, it has been in the Guardian and the Economist - and something bothers me deeply. So here’s how I think it’s going it might pan out.


Let’s start with the assumption that people have access to a 3D printer. Let’s also assume that the printer in question uses one of the current, commercially available build processes that’s safe for the home environment. That means plastics - no-one wants lasers and near melt temp metals powder in the home.

So. You’re at home. You break something in the home. A knob off your cooker, perhaps a component from your vacuum cleaner. A part that a simple gorilla glue and duct tape job won’t fix. Most people either ignore it, jerry rig a solution or order a new one.

But you’ve got your 3D printer sat in the basement not doing a lot. You fire it up, but what next? If you’re lucky, you have the remnants of the part. Let’s do a roll call. Busted Part: Check. 3D Printer: Check. See what’s missing? Yeah, the data to print out.


This is one of the most commonly used analogies whenever the 3D printing revolution is talked about and for the most part, it’s a nonsense.

Most manufacturers would rather give you their first born child rather than post the 3D geometry of their products online. Spares and servicing are a revenue stream and they’re not going to disrupt that.

So you head to whatever sources are available. Google Warehouse, Thingiverse, GrabCAD, whatever and hope that there’s a match to your part.

The cooker is seven years old. Your vacuum cleaner went out of production a while back. Chances are you don’t get a match. You think “I’ll model it up myself.”


No. It’s not. If you’re modelling a cooker knob, you’re in safe territory. It’s a few simple features. If it’s something more complex, these days the chances are that it’s a single part, but one that serves several purposes - part casing, part aesthetic surface, part structural or functional support. And guess what - it’s broken into five pieces.

Two of which you can’t find.

Good luck with that. But you’re smart, you’ve got a design or engineering background. Most people haven’t.


So, time for another check. 3D printer is warmed up, 3D data is ready. Let’s load it up and get it building. You’re typically building with ABS, right? Let’s deal with this one right here.

3D Printing isn’t injection moulding. ABS, melted, fed through a nozzle and layered up, partial hardened, then more ABS added on top isn’t the same as an injection moulding. It’s nowhere close.

Injection moulded parts are created in tool-steel moulds, the material is almost liquid at the point injection and they’re shot into that mould under high pressure. That gives you a wonderfully happy mix of surface quality and a very strong, homogenous part.

3D printing parts are vague approximations of the real thing. The build process introduces weaknesses between the layers and the layers themselves aren’t typically 100% solid. But the part you have is intended for injection moulding.

Even if it is built using ABS in the real world, just replicating that form in 3D printed ABS isn’t going to give you the same part. It’ll look the same, but that’s about it.

And there are many more exoctic materials in production used, before you even get onto filled polymers and such.


If it’s a non critical part, then you’re likely to be in luck. It’ll function nicely assuming the parts don’t delaminate, that it’s not exposed to too much UV light or temperature variation. If the part is in a more harsh environment, such as a handle, perhaps a casing or a fixture, then it’s not going to withstand the pressures and forces acting on it in day-to-day life in the home. Why? Because you’ve not redesigned it to account for dramatically reduced material strength.


So, what’s the end result. Some parts will be fine, but if you’re doing anything vaguely complex, vaguely interesting, then chances are you’ll end up with either another part that doesn’t fit properly, doesn’t perform it’s function properly or will fail sooner rather than later. What you’ve essentially printed is another broken part - it’s just not broken yet. Designers and engineers will know all of this - it’s our job after all. The average consumer, even those tech-aware enough to have a 3D printer, won’t.


Don’t get me wrong, I love the idea of 3D printing getting into the hands of more people. I love the idea that these machines allow the “so inclined” to be able to build parts and to experiment without the costs of traditional manufacturing processes. And yes, I’m missing a crystal ball to see a time when you can 3d print parts with production intent material or at least a variant that has the same mechanical, thermal and UV properties. But at the moment, you don’t. So, ultimately, I’m a bit down on the whole thing and I’d like to see a little more realism and fewer flights of fancy.

There’s also a question of safety. You build the new handle for your vacuum cleaner. The part builds nicely and it fits to the existing parts. Then you carry it up the stairs. Because of the weaknesses in the material and build process and the fact that these haven’t been accomodated for, it splits, shattered under impact, whatever. I just hope there’s no-one behind you when you do.

The mainstream media outlets talking about 3D printing is a good thing, but I do wish they’d look into what they’re espousing. If the Economist talks about 3D printed hammers, then something is deeply wrong. I’m sure they double check their financials. Why not this?

Comments on this article:


These are good issues to get out in the open.  Some assorted thoughts:

* It seems to me that most folks don’t fix their own appliances.  My father does, typically using a process he refers to as “Kevlar equipping.”  It involves wrapping the outside of the broken part with Kevlar and fiberglass and a nice coat of epoxy or resin.  Works super for things like wet/dry vacuum hoses and Cuisinarts. 3D printing certainly isn’t a better technique, yet.

* To gauge initial markets where 3D printed parts might take off, I’d look at markets where DIYers already buy a lot of parts.  What subsegments of that market are most ready for 3D printing?  Home improvement?  Automotive?  Woodworking?  Ah haha.

* I suspect that 80% of the worlds plastic is used in the production of toys.  As I kid, I greatly preferred kits to prefab toys.  There’s probably something to be done here, and perhaps TinkerCAD is on it.

* I completely disagree with you RE melting metal with lasers in my basement.  I am waiting to buy a home 3D printer until the prices on SLS and SLM become reasonable.


Posted by Blake Courter on Tuesday 05 2012 at 04:29 PM

The problem with this article is that the basic premise and assumptions are wrong. The first sentence is “Let’s start with the assumption that people have access to a 3D printer.” Fair enough. But it then equates that to *owning* a 3D printer. That’s something of an antiquated notion isn’t it? I have access to a 3D printer (in fact, lots of them), in the same way I have access to the world’s largest encyclopedia, the world’s most detailed atlas, virtually every programme Channel 4 have ever made… Doesn’t mean any of them are sitting in my home.

Once you accept that access can mean buying a service, the rest of the points begin to unravel. Even if a manufacturer doesn’t want to give you their 3D data (and that in itself is debatable in an uncertain future), it’s easy to conceive of a system where an approved supplier prints your part, so you never see the data itself. It doesn’t matter that your cooker is seven years old or your hoover is out of production - unlike physical parts a manufacturer can store data for free. No need for the consumer to worry about the material, that’s specified in the data that the consumer never sees. ABS isn’t ABS isn’t ABS it’s true, but if the part’s been designed from the beginning to be replaced (indeed, if it was 3D printed in production), those concerns drop away. I could go on, but I think you get the drift. It all comes down to the notion that “these things haven’t been accomodated for,” which might be the case today, but why is that a valid assumption for the future?

Okay, I know I’m being a bit unfair, the title was 3D printing in the home, and what I’m describing isn’t strictly that (though there’s an argument that says that as far as the consumer’s concerned, if they’re accessing the internet at home, it *is* at home). But you’re shooting at a target that’s far too easy to hit. The Economist wasn’t able to predict a global financial meltdown a month before it happened. I’m not sure why anyone should trust their judgement on where technology and manufacturing might be in two decades time.

Posted by Matt Sinclair on Wednesday 06 2012 at 01:17 PM


I get your point entirely, but I don’t think that owning a physical object in the home is an out dated idea at all and that’s specifically what I wanted to talk about in the article. Yes, there’s a tonne of service providers, but its not the same thing as having the machine, on site, in your house. It gives more freedom to experiment and the cost differential can be pretty dramatic - on the basis of a set of two parts I built on an UP 3d printer recently, I could either order 3 sets from imaterialise or any of the other vendors, or buy my own machine.

Parts, in the main, aren’t designed for 3D printing - that’s the whole point. And the economies of scale for mass produced items mean that it simply isn’t going to change. And what benefit would there be for the vendor to do the work in the first place?  I get the enthusiasm for the idea of being able to produce your own parts and how it “could” change things in the future. But unless there’s a huge change in the economics of 3D printing (spares would be horrendously expensive when 3D printed compared to traditional techniques, again due to economies of scale), then the status quo won’t change. That doesn’t mean that things won’t change, but I just don’t see this being the Industrial revolution that many are talking about.

Cynical views maybe, but they’re based on what’s available now, what’s gone before, and frankly, material performance and manufacturing techniques.


Posted by al dean on Thursday 07 2012 at 11:08 AM


I would bet you have a LOT more room for a laser melting machine than I do. The idea of being able to produce metal part’s is incredible attractive, but the process, from the rather explosive nature of powder based processes, the need for extraction, the mess involved (those powder cakes come out pretty damned hot), mean you’re looking at another entirely different level of user requirements. When you do get one, let know, I want to come play with it mate wink


Posted by Al Dean@ on Thursday 07 2012 at 11:16 AM


I totally agree with the point about the next Industrial Revolution; anyone who was at the Product Design Innovation conference last week will know it was the first question I asked the panel. What I don’t understand is your insistence that things simply aren’t going to change in future. It’s true that most parts aren’t designed for 3D printing, but then it’s only 20-odd years since the first 3D printing patent. How many parts were being designed for injection moulding 20 years after it was patented? In 1892, if you’d judged injection moulding on the technologies available and the material performance of plastics, would you have concluded then that the status quo wasn’t going to change?

Posted by Matt Sinclair on Thursday 07 2012 at 01:00 PM


There’s a lovely example of what you’re talking about at Ironbridge. When they came to build the bridge, they had the iron coke/smelting process nailed by Darby’s grandfather, but you look at the joints on the bridge, they’re from the wood working world, not that of traditional metalwork - why? Because they’d never done it before. Process/Design mismatch. Wonderful thing.

Its not so much that I’m insistent that things won’t change (ok, maybe a wee bit), more that I can’t see why they would want to change in the mainstream. Don’t get me wrong, I know all too well that denying the potential runs the risk of my sounding like a complete arse in roughly 5 years time - but I struggle to see how 3D printing (which scales in a linear manner) is every going to replace the speed and costs of something like injection moulding.

Yes, you could make bigger faster machines, more automated machines (I saw a prototype 3d printer that have a helical build platform to provide constant build volume a while back), but can you honestly say that you can see a 3D printing machine that’s going to be able to manufacture a, for example, cell phone casing, every two seconds in a homogeneous solid and reliable material?

That’s why I’m dubious. It’s not that I can’t see a place and a need for 3D printed parts, but there’s a lot of talk about how its going to revolutionise things. I don’t think it will. It has potential to change a lot, to bring to processes and capabilities in ways we both can and can’t imagine. But I also think there’s a lot of nonsense talked about it too.

Don’t get me started on the “freedom from design for manufacturing” aspects - that’s another one that properly winds me up.


Posted by Al Dean on Thursday 07 2012 at 01:47 PM


Okay, I knew I was going to get drawn into this more wink I think the key is this sentence: “can you honestly say that you can see a 3D printing machine that’s going to be able to manufacture a, for example, cell phone casing, every two seconds in a homogeneous solid and reliable material?” To answer honestly, no I can’t. But there’s a whole load of factors that go into that two second (well, maybe a bit more) cycle time which need challenging.

First of all, what’s coming out of the tool is hundreds of thousands of identical parts. Whether people actually want identical products is probably for another discussion, but if you’re trying to manufacture and sell unique products, even with only slight differences, then injection moulding tooling isn’t a big help. Bespoke Innovations and MakieLab, which you’ve featured recently, wouldn’t have a business model without 3D Printing technologies. My feeling is that they’re just the beginning.

Secondly, that cycle time is a bit misleading, in that it doesn’t take account of the weeks spent producing and setting up the tooling. If you’re in the business of manufacturing hundreds of thousands of cell phone covers, then injection moulding is still faster. But if your business is manufacturing lower volumes of un-identical products… What’s more, getting the cycle time down is important if you’re producing for stock and trying to amortize your tooling costs. But if you have no tooling, and if you don’t manufacture a product until *after* someone has paid for it, cycle times are much less important.

Thirdly, a solid and homogeneous material? Why would I want that if 3D printers could make micro-porous or functionally graded or micro-mechanical materials? I can imagine a cell phone cover which is shock resistant in the places most likely to be hit when it’s dropped, soft in areas that touch the face, acts as an antenna and which can be rolled up when not in use. I can’t 3D print one yet, but I can’t even imagine injection moulding one.

There’s that saying about how people understand the future - we over-estimate the short-term change and under-estimate the long-term. I think that’s the problem with articles from the Economist etc; they’re fixated on how 3D printers will upset copyright or bring jobs back to America or get everyone consuming again, because those are the short-term problems its journalists like to imagine they know how to fix.

Posted by Matt Sinclair on Thursday 07 2012 at 05:41 PM

Al, you don’t mention 3D scanning as an input choice. 

What 3D scanning is not, is a simple “scan it then print it” approach.  What it is, is a way to get complex shapes into a 3D model that can be printed, with a level of accuracy dependent on the device used to scan.  Not necessarily easier than modeling the part in CAD, but definitely faster when you factor in the need for accuracy. 

Most scanners aren’t priced for the consumer or prosumer, but there is the NextEngine and the DAVID scanner.  If your tolerances are loose, there’s even 123D Catch.

Posted by Tom Charron on Tuesday 12 2012 at 10:15 PM

If they ever become reasonable to own, then I think 3D printers will be like counter-top bread makers that were popular last decade. They’ll seem like a wonderful idea to consumers, until consumers come to the realization that it’s just plain easier to buy a loaf than to make it.

Posted by Ralph Grabowski on Tuesday 12 2012 at 11:26 PM

Hey All,

This is a great discussion. I am crtainly a novice, but i just wrote 2 articles for Seeking Alpha about 3D Printing:

The first one is about how it might ruin the toy industry.

The second is about my belief that Apple should buy into the industry.

Let me know what you think.
Matt Cilderman
@mcilderman on twitter

Posted by Matt Cilderman on Friday 22 2012 at 01:54 PM

Hi all

Something that everyone forgets to mention is that the surface finish of any 3d printed part is crap compared with an off tool part. One of the big things with individualisation is always going to be getting a unique & quality object. 3d printing just cannot do this without a pair of hands adding time to it & then the costs go out the window. I think Al is pretty much bang on with his reasoning, but yes things will change & improve in time. How much time is the big question.

Posted by Steve Willmott on Monday 02 2012 at 09:48 AM


I think smoothness depends on the 3D Printing technology being used and how much you want to pay. Objet claims layers of up to 16 microns (0.0006”) which is a pretty smooth surface.

Posted by Mark on Monday 02 2012 at 09:06 PM

Al thinks software is wonderful but is slow to take to hardware - at least as far as 3D printing is concerned. Oops sorry Al. He ignored low cost 3D printing for 3 years until he got his hands on an UP! and now is a partial convert. (Al - you need to try our Maxit as well - the first 3D printer specifically designed for education).

Those of you above who have posted rejections or partially contest his worldview are, I think, closer to the truth. Also, I see 3D printers at home being more a toy for kids than a tool for adults, and that adult requirements are more likely to be handled by the local hardware (ie 3D printing) store.

Finally, we are in discussions with a university to jointly develop the Maxit so that it can also handle syringe / augur delivered materials, including foodstuffs, Playdoh, ceramics etc. So another of Al’s premises – that all 3D printing will be plastics - is incorrect. Anyway, here at A1 (that is A1 Technologies not Al Dean) we tend to print PLA rather than ABS, and is therefore biodegradable while being made (at least our version) from wood fibre, and therefore similar to ABS (albeit printed and not injection moulded) in strength and utility.

Posted by Martin Stevens on Thursday 05 2012 at 11:45 AM


Let’s straighten a few things out.

Ignored low cost 3D printing? Not quite right. I’m happy with the fact that we’ve focussed on the higher-end machines (I won’t point out the article I wrote about RepRap in 06/07). We’ve built parts, we’ve seen what they can do. As a result, we can draw a conclusion on how each performs. As for the entry level machines, I waited till I could spend some time with at least one them.

After all, that’s my job.Try something out, make an informed decision and see how it both fits into the cost model, ease of use and workflow of a design professional (who are our key audience, thought this is expanding outwards).

I could have done it differently. I could have just read a few articles on the internet, picked up a few samples at a trade show and made a comment about how wonderful they are because they’re cheaper. Which is a better approach, is down to the reader to decide. So I’ve not “ignored it”, I’ve done my job properly.

Let’s deal with 3D printing as a toy for kids. Really? Show me one other “toy” that melts plastic, usually in an unguarded state, at 260 degrees C. I’ve got two small kids. They love a bit of 3D printing, just as much as they love Lego. Am I going to let them go at it on their own? No. I like them with all their fingers unburned and no near molten plastic in their hair, thanks.

I’m with you 100% on the schools potential. Already in the US, shop class has gone. It’s not been merged with textile design or food prep. It’s gone. Forever. And I don’t want to see the same done in the UK. If adoption of 3D printing (which, frankly, is still, after 18 years since leaving college, an utterly entrancing thing to behold) gets a few more kids interested and engaged with the process of design and engineering, them it needs to happen. And I appauled what you’re doing with the Maxit. Absolutely. And I’d be more than happy to have a play with one, any time.


Posted by Al Dean on Thursday 05 2012 at 12:16 PM

Humble apologies Al at the implied slight.

You can allow your kids to download a Lego or Duplo file, as we have done, and then use our Chameleon 3D design package to design a little garden on top of the brick, and then print it; it is now a unique part. It is quite difficult to burn yourself on a 3D printer, but we are also investigating PLA which melts at 50 degs C. When we have a printer for food, Playdoh etc, of course, there will be no heat involved at all.


Posted by Martin Stevens on Thursday 05 2012 at 12:36 PM

Better late to the table than never. I have a couple of budding teenage industrial designer daughters at home who’d relish the thought of bringing their manga/anime creations to life. Even a monochrome 3D printer affords the opportunity to paint or coat the finished article to resemble the intended result. Then they’d sell them on DeviantArt.com and make Daddy a fortune!

Posted by Mike Edwards on Monday 17 2012 at 02:10 PM

At Last somebody has finally realised the emperor is naked! As a maker and somebody who understands the appropriateness of materials and process, I have long stated that these are very useful tools to help impatient people visualise 3d objects, that’s all they are. They are brilliant technologies developed by brilliant people but marketed even more brilliantly! They also have fundamental and so far incurable flaws. I think the next step is to ask. ‘Has time compression really made our lives better’? The answer is simple if you have recently purchased a product which either does not work out of the box or has broken with ease or is just not fit for purpose. If it has accessories or power supplies which are unobtainable within 6 months or ink cartridges which are not made any more!
People need to rationalise technologies into their purpose and their benefit and not try to turn them into something they are not! And people need to value their products and feel familiarity with their world.

Posted by Chris Hill on Wednesday 03 2013 at 09:48 PM

Mechanical Engineer with experience on 3d modeling for manufacturing with CAD design especially on Solidworks , 3d models for 3d printing, 3d animations on Cinema 4d and Renders for marketing photo-realistic pictures .

Skype .(JavaScript must be enabled to view this email address)
.(JavaScript must be enabled to view this email address)

Portafolio website link.

Linkedin Profile

Posted by Noelia Walter on Sunday 19 2015 at 03:25 AM

Leave a comment

Enter the word you see below: