If you want to make your own gun or harboured a desire to make a boat perhaps the device you need is a 3D Printer. In principle it sounds easy: just download a 3D model from the net, throw it at the printer, and whatever you desire comes out the other end. MakerBot The Replicator MakerBot's The Replicator The truth is, it’s …
Christmas crackers and cornflake packets
It's a great technology for making small plastic gewgaws very slowly and very cheaply. I await with interest the developments that will enable it to work with steel, light alloy, titanium, tungsten carbide, and do it faster than casting or forging can do. (Wow, I've just printed my own oil tanker!)
Re: Christmas crackers and cornflake packets
Metal support is already here.
I was at one of the trade shows a while ago, there is at least one company doing additive manufacturing with metals. The machine uses a different process, fusing a bed of powder with a high-power laser (no sign of any sharks in the model I saw, sadly). This allows for some fascinating light-weight metal structures to be made.
I think the company I spoke to are only leasing machines currently, and they are at the eye-watering, if-Sir-needs-to-ask end of the cost range, but they do exist.
I'm not a shipbuilder but I'd imagine big ships would be made mostly out of rolled steel. Not much to be gained there from this approach. AFAIK you can already do this "additive manufacturing" shtick with metals, if in a machine about the size and probably not quite the price this was ten years ago, but whether the results would be strong enough for, say, a small motor (or model jet engine) I don't know. Anyone? El reg?
Metal printing is also known as Laser Sintering it is pretty quick for one off and low volume parts, especially since it can make as many parts that will fit in the cube at once. I have seen Aluminium and Titanium in use but the process works for many types of metal (generally the lower melting point types!).
These Titanium parts for Bloodhound SSC have been on show in various places, they are a fairly rough build similar to a good sand cast, but the process can produce higher tolerance 'smooth' parts also if required.
Good enough for NASA
Well NASA evaluated the process years ago and said it was good enough for production parts.
Not only that, but SLS (selective laser sintering) can produce parts with superior properties to those made by traditional processes. You can even achieve 'property gradients' if you require them- for example, a beam that varies in stiffness along its length.
. It is used, as you would expect, in the fields of aerospace and motorsport.
Re: Oil tanker?
@ 08:08 GMT Anonymous Coward
Gives details of materials (including titanium, aluminium and steel alloys), specifications, and also applications as end-use parts in aerospace.
"...for a new engine coming out next year, GE is planning to print them [jet engine parts] in cobalt-chrome."
" including 32 different [printed] components for its [Boeing's] 787 Dreamliner planes."
As for jet turbine blades... Rolls Royce blades are a single crystal of titanium (temperature controlled lost wax casting), so I wouldn't expect 3D printed blades to be as good. For a model aircraft engine though, you wouldn't be as concerned about strength/weight/safety.
There's also jet cutting, which can produce complex shapes,
but needs a clear path through, so can't build tubs, etc.
Home printing destroying the Christmas Cracker industry!
Don't worry, they haven't written software to write shit jokes yet!
Who needs a computer program to write shit jokes?
Re: Christmas crackers and cornflake packets
"It's a great technology for making small plastic gewgaws very slowly and very cheaply."
Granted the current set of vulture heads cost 12 pence in plastic which would qualify as very cheaply but there is the matter of the capital cost of the machine which adds about $500 to the price of each of the four heads. Then there is the other expense that everyone seems to neglect, power. Perhaps the good folks at El Reg could cook up another vulture head with a watt meter attached and record how many watt hours it takes. Oh don't forget to measure the preheat time separately since you could do several in one go if they are small enough.
"Looking at the online libraries of available things there is very little, gun parts aside, that is actually useful and most of those are probably best bought in a shop."
There are loads of useful things on thingiverse, you just have to get through all the cruft and have an application for the useful things (they are usually fairly specific replacement parts for cars, appliances and furniture).
I'm guessing you are printing ABS at those temperatures. Try dissolving some ABS in acetone and using it as a glue to prime the bed with... otherwise try printing in PLA at lower temperatures. I have only had a couple of parts come of the bed with PLA and never need to print rafts.
An obvious example of something useful would be the spare parts that white goods manufacturers charge for... A door latch component on our tumble-drier failed, it has warped in the heat (poor material choice). The Whirlpool website was very slick, and smoothly offered to take £18 off us a replacement latch, a part that really could have cost no more than 20p to make.
Were 3D printers more common place, this business model (be competitively priced at retail, charge plenty for spares) could be dented.
I gave a shout to Ben Heck the other day, (in relation to an article about Valve exploring new games controllers) and I will do so again: he routinely uses 3D printers and CNC machine for a variety of projects that I am sure Reg readers will approve of (such as a C64 style case for a Raspberry Pi), and produces videos about his exploits. You can do worse than Google his name and check out his website.
An obvious example of something useful would be using a laser scanner on certain miniatures, and then applying that information to the 3D printing process. It could just about print money with that concept. So El Reg, HOW ACCURATE AND DIFFICULT WOULD THAT BE?
I know a bloke who needs a rocket-plane...
Thank you .....
... for an interesting and informative article. With an initial purchase cost of about $2000, it's not a bad price for serious hobbyist equipment and a very low price for any small company that could use such a capability.
Christmas is coming....... :)
Re: Thank you .....
And here is one with a bill of materials totalling $500, a 'Delta' robot, - imagine an upside down camera tripod, and the XYZ position on the head is controlled by the lengths of the legs. This is another open-source (make it, improve it, let us know) 3D printer, and the link will give you 3D models, schematics, BOM, and software:
Re: Thank you .....
reprap.org is the place to go if you think you have the nouse to build a 3d printer yourself... expect to stump up about £500.
Print a case?
Yes, yes you should. For your raspberry pi.
Now for a similar set-up for multi-layer PCBs, and printing phones would suddenly be much closer. Not sure if this'll end up in every home, but the corner shop would be good enough. In fact, document replication shops will be with us for a while, but not quite as much as once, perhaps. This something for them to foray into?
Re: Print a case?
Printing multi-layer PCBs. Brilliant. Perfect killer app.
Re: Printing multi-layer PCBs
Better still, print the components too. Who needs a silicon foundry? ;-)
LPKF can do it but you need a lot of equipment.
As alternatives to Sketch-up, you can also consider both Wings3D and Blender.
Both are free (open source) and quite easy to use. I use Wings3D for some game modding, and I know others who use Blender for the same. Either would do the job nicely for conjuring up models for printing.
Autodesk Inventor Fusion is available as a free (as in beer) download, in the form of a technology preview:
This is representative of the new movement in CAD, the combination of Parametric (think drawing board, or Mechano) and freeform modelling (think clay). Previously, people would shunt models back and forth between packages such as Solidworks and Rhino.
It expires for PC users on 1st April 2013, no date specified for OSX. It is the first of the big CAD players to produce a Mac version, an interesting development.
>It is the first of the big CAD players to produce a Mac version, an interesting development.
Bentley MicroStation used to run on the Mac, but was withdrawn around '97 for some reason.
There are other uses
It is getting popular in the model railway hobby.
Short runs of unusual wagons.
Base coach bodies for sticking etched sides on, interiors done, underframe not bad, etched brass sides look good, in smaller scales.
Parts for models.
Not just prototyping
The economics for use in production would seem to be dependent on the cost and ability to recycle the raw material, the volume of production runs and complexity.
Direct deposition/3D printing has the advantage of very little wastage. So if you are using high grade titanium and know that your wastage can be recycled but only as a downgraded material saving about 90% of the amount needed to mill an aerospace turbine housing from a solid block of aerospace grade titanium is very interesting.
As long as you are within material and printer parameters some forms of complexity become almost a non issue - hence the use by jewelers and so on.
The volume of the run is also important - shouldn't be too long before dental work is a large user of 3d printing - quick bespoke crowns and so on . No need for one - off casting.
(anyone know if anyone can print bronze? should bring back that craft at a much lower cost)
Given the range of material one might wish for I suspect bureau printing (eg shapeways has a material list http://www.shapeways.com/materials ) has an interesting future.
But still - me want. fab@home material syringe seems like fun but the reprap seems more sensible ....
This might help, a discussion on the open source 3D printing site RepRap about the very question you ask. They are talking about 'Bronze Clay', particles in bronze suspended in a binder. After modelling (usually by hand) the 'clay is placed in a kiln to give you fairly dense parts. A silver version is also available.
Even these guys haven't found a way to print the material directly, you could print your self a mould. Your shapes would be limited by the usual mould design considerations, except you could consider your mould to be expendable.
Re: Not just prototyping
"The volume of the run is also important - shouldn't be too long before dental work is a large user of 3d printing - quick bespoke crowns and so on . "
My dentist will probably have one as soon as they are commercially viable. He likes new toys that improve his performance/efficiency.
Several years ago he surprised me when preparing a tooth for a crown. Instead of taking a mould he took a 3D picture - and fed it into a 3D milling machine. We had a chat while the crown was being milled - then the finished piece clicked neatly into place. There were a selection of blanks in different enamel shades so that the crown matched the teeth.
Dental is only one field Renishaw work in, having grown out of metrology in the aerospace industry. They are probably most famous for their ruby-tipped contact probes, for measuring turbine blades and the like, but are moving into surgical robots.
Re: Re Bronze
Was only a side thought - its a complex process http://www.rosalindcook.com/clay-bronze-sculptures.htm at the moment. Sculpture -> mould -> wax -> mould -> bronze. Looks like it could be shortened somewhere but given the engineering could be a challenge - maybe printing the wax ...
Re: Not just prototyping
On prototyping (or er playing more honestly in my case) ...
Should also give a hat tip to openscad - free interpreted simple programming tool for 2d-3d extrusion and constructive solid geometry (think basic set operations with geometric shapes). Easy and fun with only basic programming and cartesian math skills - most reg readers would surpass that I'd guess!
Can get it to export STL files for printers and the like (although some post processing also by free tools).
Re: Not just prototyping
Dental work using ceramics and cutters to create crowns on the fly is well established. My dentist has been using this stuff for about five years now (from Siemens, can't remember the trademark). I have about 3 or 4 in my head. The last time from start of treatment (removing the old tooth) to leaving with a perfectly fitting crown was about 40 minutes... Also cheaper.
Re: Re Bronze
"Sculpture -> mould -> wax -> mould -> bronze. Looks like it could be shortened somewhere but given the engineering could be a challenge - maybe printing the wax "
Have been waiting for the technology to mature for the last decade. Although there are haptic tools to create the sculpture model in a PC - that doesn't help those of us who use hands-on clay. The real missing link is a 3D model capture using ordinary photographs.
That still leaves a question mark over how to smooth the stepping ridges of a printed/sintered/hardened process without softening the fine detail.
Re: Not just prototyping
You beat me to it.
In addition to having the manufacturing fully automated my dentist has a screen in front of the chair so you can see the 3D model and then watch the progress of the cutter as it creates the crown. Not only much faster and accurate than the old way of doing it but fascinating to watch.
Read Rule 34 by Charlie Stross to see where this might lead
Charlie Stross uses 3D fabbers as a major plot element in his novel "Rule 34". It is disturbingly plausable.
Things to come
if you consider the Makerbot to be the equivalent of an early 8 pin dot matrix printer think what the the equivalent of a current inkjet printer will do. The current range from Objet are interesting but the price is way beyond my budget.
Re: Things to come
And in other news, Stratasys (3D printers) and HP are parting company, without the collaboration bringing any dramatic reduction on the price of the machines. My old school's first laser printer, an Apple, cost them £5000... I was hoping HP might get the unit cost down to that sort of price.
What's the latest progress on producing 3D computer models from a series of digital pictures of an object? There was something from Cambridge - that was supposed to work with a set of photographs which did not have to be tightly controlled perspectives.
Re: 3D photographs
I had a play with the demo, but didn't have need enough to get to grips with it. You need to print out some sheets with which to calibrate your camera, since it need to correct for lens distortion. Modelling humans (or any moving subject) works better if you use several calibrated cameras at the same instant. A few low-end Canon compacts would probably do the trick, since there is a low-cost hardware hack that allows shutter release through their USB ports.
An alternative if you want a 3D scan of your head are those people who set up stalls in shopping centres offering a 3D model of your head made by forming bubbles in a cube of glass-like resin. The laser scanner they use outputs *.XYZ files, literally a CSV file defining 3 co-ordinates for each point in 3D space - ask nicely and carry a USB stick. The easiest tool I've found for creating a surface from this 'point cloud' is the 'drape' command in Rhino (trial version allows 20 saves).
Otherwise, full power to the Kinect hackers. The Reg has a real stuffed vulture. The Reg can borrow a Kinect I'm sure. Maybe the Reg can attempt to make a 3D scan of their bird of choice and publish an account of their efforts?
Gun and model of Zuckerberg's head
Obvious combination to start off with.
Re: Gun and model of Zuckerberg's head
Print Zuck's head in wax, and recreate the ending of Raiders of the Lost Ark with a heat gun....
Re: Gun and model of Zuckerberg's head
Wouldn't that be terrorism?
Even home 2d printing isn't all that great
3d printing used to be known mainly as "rapid prototyping", it's now being used ever more for "rapid manufacturing" and this makes a lot of sense for low volume products.
Various parts of the media, from geek websites to the FT, have been putting forward the notion that 3d printing will revolutionise manufacturing by putting a 3d printer in everyone's home and allowing for decentralised manufacturing. I don't see how this is even remotely plausible.
I always think of home inkject/ laser printers when the notion is put forward. Would anyone seriously consider printing and binding a book at home to be preferable to ordering it online and having it delivered the next day. Consider the cost, time, reliability and hassle. Even for a custom book you would still be better off going with something like blurb.co.uk (think about the difference in quality, TCO and cost of you own time). And this is with technology which is fundamentally a lot simpler to 3d printing and has had decades to mature and become dirt cheap.
I can imagine industrial scale 3d printers, which would be the equivalent of the printing machinery used by the likes of blurb, being used more widely for spare parts and such but the notion of it being done at home seems fanciful to me.
Re: Even home 2d printing isn't all that great
" Would anyone seriously consider printing and binding a book at home to be preferable to ordering it online and having it delivered the next day."
Bought a "print on demand" book. The out of copyright material was already online as a very slow scrolling PDF. The content was almost entirely black and white photographs which the "print on demand" rendered as predominantly black splotches lacking any detail.
Re: scale matters
Would anyone seriously consider ordering 1,2 or 3 pages online and having them delivered the next day to be preferable to printing them at home?
Print a 3d printer, half the size?
Come ON! Where are the suggestions of, "If we just get a machine to make itself half size, nanotechnology will be a reality!"
I miss the days of Tomorrow's World crazy technoutopianism.
Re: Print a 3d printer, half the size?
That's before everyone seemed to realize that that nanotechnology doesn't follow standard forces and delves in to the exceedingly odd world of quantum mechanics. The technoutopianists ignorance made them think we were much closer to mass produced technology at the atomic level. That said, we have learned a great deal about making stuff smaller, and due to that we have the dense storage and processing capacities in our current products.
TL:DR, making stuff smaller is hard.
I'm with you, Conel... RP was my dissertation subject more than ten years ago, and I still can't think of a convincing end-use that justifies the hype it has attracted in the last few years (with no real advance in technology).
Objects are usually better if mass produced (the tolerances of injection-moulded parts are incredible) and if you want to customise an object for a person (a hearing-aid, for example) taking a direct cast of the body part is usually the better option.
Obviously the hobbyist / designer part of me is excited by this stuff, and I can identify a few niche applications for the average household (see tumble-drier door latch, above) but nothing that would really justify everybody owning one. I'm hoping to be proved wrong, though!
Something to keep an eye on is the printing of body parts or very small items... but again, these are not consumer applications!