3D printing

My new article about 3D printing is now up on RegHardware. For more background, I’ve republished the piece I did for Personal Computer World in 2008 here.

3D printing – the future

This item was originally published in Personal Computer World in 2008. I’ve reproduced it here to link to a new RegHardware piece on the topic, together with my original article on RepRap.

Current projects like RepRap tend to concentrate on reproducing simple plastic parts, but there are interesting developments not that far away. For example, some experiments have already been done on ways that the technology can make circuit boards – or at least lay down the tracks. By using an etch-resistant material instead of the usual plastic polymer, all the user needs to do is take the board and place it in an acid bath, in the usual way – a process that can itself be automated. Longer term, it may be possible to use different heads to lay down layers of circuitry.

Another useful avenue of exploration is enabling RepRap type machines to recycle, so that instead of relying on a fresh supply of polymer to create things, existing objects can be shredded to a granules, and then re-extruded as something else. So, the example children’s shoe that the machine can make could be recycled, with a little extra material, and turned into a larger size.

It’s even possible to grow crops that will provide the raw material to feed a machine. Starchy crops, such as maize, can be fermented to produce the polymer polylactic acid (PLA), which can then be used to create products like cups and containers in a RepRap, which can be recycled or simply composted when they’re no longer needed. On a much larger scale, similar techniques to prototyping are being explored in the construction industry, where machines may be able to create buildings by laying down thin layers of concrete, potentially enabling fast automated construction of any self-supporting structure. Ideal, perhaps, for building on other planets.

3D printing with RepRap

This article was originally printed in Personal Computer World in 2008. I have reproduced it here as an adjunct to a new piece on the topic for RegHardware.

A staple of much science fiction is the universal machine – whether called a cornucopia, replicator, foundry or something else, it’s a familiar concept: a machine that can make anything, and at its extreme, more copies of itself. While a truly universal machine, able to create complex circuitry as easily as a cup of tea or a new swimsuit is still a dream, the idea of 3D printing is far from fiction.

In fact, ‘rapid prototyping’ machines are nothing new – they’ve been used in industry for some years now, enabling design drawings to be turned into objects by laying down successive very thin layers of plastic in a process a little like inkjet printing. But these machines cost thousands of pounds, and are very complex – certainly well out of the reach of the home or even small business user.

Now, though a lot of that could be set to change, thanks in large part to work that’s been carried out over the last few years at the University of Bath, creating a rapid prototyping device that?s not only a fraction of the cost of existing system, but can make many of its own parts and can be built and used without detailed specialist knowledge.

Rapid prototyping

The brainchild of Adrian Bowyer, RepRap stands for Replicating Rapid prototyper. The idea was first outlined in 2005, and the first RepRap built the following year. May of 2008 saw the ‘birth’ of the first child RepRap – one for which as many of the parts as possible were made on the original machine. But before we look at RepRap in more detail, exactly what is rapid prototyping, and how does it work?

Although we mentioned inkjet printing in the introduction, it’s perhaps more accurate to think about plotters; a rapid prototyper is based around an extruder – think of something like a glue gun, controlled by a computer – which is fed from a reel of plastic polymer. The polymer is heated and deposited on the surface below it. By moving the extruder head in the X and Y axes, it can create an outline, just like a plotter.

What makes things 3D is that the surface onto which the plastic is deposited can move up and down – the Z axis – and so by lowering it slightly after the first layer, another can be laid down, and so on, building up a three dimensional object.

The end result is a plastic version of the object, made using a polymer like PLA or ABS. But that’s not necessarily the end of things – not everything, for example, can really be made of plastic – especially those exposed to high temperatures. So in industrial or mechanical applications, the prototype that’s printed can then be used to create a mould, and an object then cast in metal or some or material. Since everything’s driven by computer, directly from drawings created in a CAD package, it can be far more accurate than making models and casting from them.

Introducing RepRap

Commercial rapid prototyping systems cost tens of thousands of pounds; RepRap’s different. It’s designed to be cheap to make, and the whole design – including the software – is open source. The intention is that anyone can put one together, and start making things.

The design is fairly simple, and even the extruder can be made without having to buy too many specialist parts. Items such as the stepper motors that are used to position the head are cheap, standard pieces, and everything’s controlled using standard microcontroller circuits, driven by open source software based on a system called Arduino, a cross platform toolset for linking your computer to the outside world.

The software on the PC drives RepRap, based on the design of an object which can be created in standard 3D design tools, and saved in a format called STL, which is used by many prototyping systems. There’s already a small and growing library of objects that you can download from the website; the project team hopes that in time more will be added, allowing people to create whatever they want, simply by downloading the object description and feeding the appropriate polymer into a RepRap machine.

If the idea is appealing, but the thought of building a machine from scratch is daunting, don’t worry – there are companies such as Bits From Bytes that can supply kits of the mechanical and electronic parts, so all you have to do is put them together.

All the software for RepRap is downloadable from the site, as well as comprehensive instructions. According to RepRap’s Adrian Bowyer, over 1,100 electronics kits have been sold, so there’s already a fairly large community of users.

Building a printer

Building your own 3D printer might seem odd, in these days where we buy new equipment rather than having old stuff repairs. But it’s probably not beyond the skills of most people who have built their own PC, and certainly not those who built their own computer, in the days before the IBM PC.

While RepRap is probably the most well known of the small prototyping systems, it’s not the only one. The Tommelise project at 3dreplicators.com is based on RepRap, but designed to be simpler and made out of wood, with just a few hand tools necessary to build it; the creators reckon it can be made for about $150 (£80).

Another project, Fab@Home is a little more expensive, with complete machines costing around $3,600 (£1,940). It uses a disposable syringe to deposit material, and can be modified to handle two at a time, enabling you to create more complex designs.

And, lest you think that you’ll only be able to build simple things like plastic cups, there are galleries on both the RepRap and the Fab@Home sites, showing some of the things that have been made by both machines – including screw top containers and even iPod covers.

If you simply want to experiment with what can be built by squeezing things out of a syringe, then Fab@Home is probably a little easier to get to grips, especially since you can buy a ready made machine while RepRap, with its polymer extruder, is arguably closer to a traditional rapid prototyping system and will require rather more effort to set up.

While much of what’s being made with RepRap at the moment is just polymer based, it won’t remain that way. Circuits can be included in designs by building small channels into them; RepRap’s Bowyer has created a second head for the system while is designed to lay down track of Field’s Metal – an alloy with a very low melting point – in those channels. That can provide the internal wiring to link components. By including appropriately shaped compartments for things like integrated circuits, they can be dropped in place, making contact with the tracks laid down by the printer.

Similar techniques have been used with Fab@Home too, with conductive pastes laid on the other materials, and it’s this that has enabled users to come up with basic electric designs, such as simple miniature torches, linking batteries to LEDs. Although it’s still relatively early days – the ‘print head’ for Field’s Metal hasn’t yet been integrated into the RepRap, techniques like these will make such 3D printers even more versatile.

One we made earlier

One of the novel things about RepRap – and the reason for the ‘Rep’ part of the name – is that a functioning machine can make all the plastic parts for another one; in fact, the ability to create a copy of itself was one of the main design objective. So whether you’ve bought a kit or made one from scratch, you can set it making spares as soon as you’ve finished it. Or create another to give to someone else.

Of course, with the need to use some parts that can’t be easily replicated, such as the metal connecting roads, wires and the programmable microcontrollers, RepRap and similar systems are in reality a long way from the cornucopia machines of science fiction. But that’s not to say that they can’t have a dramatic effect – especially if, as we mention in the box out, they use as their source material things that can be grown and turned into a polymer without industrial scale processes.

Look around and see how many simple plastic objects – or objects that can be made of plastic – we use every day; combs, cutlery, crockery, containers, valves and so on. Being able to make those cheaply and easily with a computer, a 3D printer and locally grown supplies really could make a difference to a lot of people. RepRap’s slogan is ‘Wealth without money.’ And why bother printing money, when you can print the things you’d use it to buy?

On location

One of the web sites that I run is, shall we say, aimed at gentlemen of a certain disposition, with particular shared interests, and a fondness for the well honed physique.

I’ve been working on a mobile version of the site, and one of the things that the desktop version has is a list of venues – bars where the members of the site will feel welcome and want to spend some time. This is a natural for adding to the mobile version of the site, and for enhancing with code to support geo-location.

That’s the means by which the web browser can determine its physical location – it works on desktop PCs too, but is obviously even more useful on a mobile device. A couple of JavaScript function calls are all that you need to do this, and then you can let the web server know where the user is, and customise the information.

So, on my site the obvious thing to do is to use this to filter the list of venues, showing those closest to the user, first, and omitting all those over a certain distance away. I’ll do a separate web page explaining how that works from a programming point of view.

There are two interesting things from a privacy and openness viewpoint that struck me, though.

Can you trust location services?

To take the latter first, I was joking to the owner of one of the venues that, for a small consideration, I could make sure that his rival always appeared to be 50km further away than it really is. Although it wasn’t a serious suggestion, it did make me wonder – how can you be sure that such information is trustworthy (beyond the actual error in determining location).

If you’re using a big open site like Google maps, then it’s probably quite reasonable to assume that if they were deliberately omitting, say, branches of Burger King that were close to branches of McDonalds, people would notice. But in a closed application, or a site dedicated to a specific topic, it’s much harder to do. And I wonder just what lengths some people will go to, in order to skew the results (and of course, it’s not just geographically, but given the task in hand, that was what sprung to mind).

Privacy issues

The other topic that I’m wrestling with is privacy. The site has a privacy policy that I think is quite clear, and we explain what persistent cookies are used, and what they’re used for.

Location, it seems to me, is something that needs to be covered in this too. So, does anyone out there have any best practise guidelines for how location data should be used on a website? Obviously this is going to vary from site to site and app to app – for example, the dating app Grindr is based around the idea of showing you who’s in your area, constantly updated, so users are aware it needs that information.

But a site like mine, where location is used in slightly different ways, and hasn’t been used before, perhaps needs to treat it differently.

This is what I’ve come up with, as the basis of a location privacy policy:

1. We will only attempt to determine your location on specific pages, and only when necessary.

2. We will always explain why we are attempting to determine your location (eg, to show you nearby venues, or to let other users know which city and country you are in).

3. We will never show other users your exact position, nor will it be stored in our database.

4. We will always allow you to view and edit any location information before deciding whether or not you want to share it with other users.

Icecrypt T2200 Freeview HD box only £60

If you’re quick – before midnight today (8th March), you can pick up a FreeviewHD bargain.

The Icecrypt T2200 Freeview HD box is on offer at Play.com for £59.99. It’s a pretty respectable bit of kit, though it lacks Dolby Transcoding. Once you’ve got one, you’ll find some beta firmware here, which allows you to record to a USB hard disk.

You may want to read my review on RegHardware.

Patrick Whitfield, November 1967 – 5th March 1991

Twenty years is a long time; almost half my life now, but some things continue to shape our lives for many years after they happened.

I was going to write something new about this, but decided instead to repost something buried away in an old version of my website. You may also want to read a related post from last summer, with my thoughts regarding the lack of charges in the Ian Tomlinson affair, which contains a bit more about the PCA investigation.

There is one member of the family that merits special mention; Patrick Whitfield, my late twin brother. His death and the events surrounded it changed my outlook on many things. When bad things happen, you may suddenly realise that the things you believe in really aren’t quite what you thought.

Patrick died when his life support system was turned off on the 5th of March 1991. He had sustained fatal head injuries, resulting in brain death, after “leaving a police van through the rear doors, while it was in motion.” He was taken to Addenbrookes hospital in Cambridge, where my mother and I waited from Thursday until Tuesday the 5th, though it was clear from Friday the 1st that he would not survive.

Unanswered questions

As is usual when there is a death in custody, an investigation was conducted. And, as is the case in the UK, that was carried out by another police force and, as family, we’re not entitled to see the results of that investigation. The officers involved in stopping my brother, who had failed a breath test, would not answer questions at the inquest. The investigating officer told us “There’s nothing secret in the report, but it is confidential.” The Crown Prosecution Service wrote suggesting that “although an offence had been committed, it was not of sufficent gravity to merit prosecution,” and Cambridgeshire Constabulary spelt our surname wrong when they wrote to offer condolences.

Social justice

This horrific experience, and the careless attitude of assorted state agencies, fundamentally changed my view of many things. I was unemployed at the time, and since I missed signing on because of being in hospital, and then arranging a funeral, I lost all benefits. I was told by the unemployments office that you are allowed one day not looking for work in the event of a death in the family. If I’d known I was going to be away for longer, I should have filled in a holiday form!

It may be that the officers involved did everything by the book. Perhaps the book is wrong, or perhaps they did make a mistake. But without being allowed to see what the investigation revealed, we will never know.

We need an open and accountable police force – that doesn’t mean a witch hunt, it means answers to simple questions, like “Who said what to whom on the night? Who did what, and when?”

We need a social security system that doesn’t withdraw benefits from people just when they need them. When bad things happen to people, they should be supported – not told to fill out a holiday form. It seems to me that bureaucracy is so intent on keeping figures down that it’s more important to stop a claim than consider whether or not someone needs help.

Inquest is an organisation that campaigns about deaths in custody.

HarperCollins statement on library eBooks

I wrote earlier this week about new proposals from HarperCollins which would mean that library copies of eBooks expire after 26 loans. Today I’ve done a short news piece for RegisterHardware.

Since it was only a brief item, and there wasn’t space to include the HarperCollins UK statement in its entirety, I’m reproducing it below:

HarperCollins has always been a firm supporter of libraries in the UK, working hard to ensure that our books are available in both physical and digital form. We also work closely with The Reading Agency and include libraries as a regular element of all author tours. With a rapidly changing digital market, establishing the right arrangement for lending e-books, which supports the interests of both readers and authors while providing a sustainable future, is not straightforward and many models may need to be tested before we reach the optimum outcome. The deal which HarperCollins US have struck involving a cap of 26 consecutive circulations on a “one copy, one reader” basis, is one potential solution, and preferable, we believe, to simply prohibiting ebook library lending as some publishers in both the UK and the US have done. We continue to work with the PA on this issue, with the aim of reaching an industry-wide solution.

Update: I’ve asked HC to clarify whether or not – as OverDrive told me – the 26 loan limit will apply in the UK, as their statement isn’t 100% clear on that. I’ll update when I hear more from them.