Slashdot Mirror
5-Axis Robot Carves Metal Like Butter
kkleiner sends along an amazing video of what robot-controlled machining is coming to. "Industrial robots are getting precise enough that they're less like dumb machines and more like automated sculptors producing artwork. Case in point: Daishin's Seki 5-axis mill. The Japanese company celebrated its 50th anniversary last year by using this machine to carve ... a full-scale motorcycle helmet out of one piece of aluminum. No breaks, no joints, the 5-Axis mill simply pivots and rotates to carve metal at some absurd angles. Every cut is guided by sophisticated 3-D design software (Openmind’s HyperMill)."
I read that as "Craves Metal" and was scared as hell.
I will only be really impressed when it can smoke cigars, swear and run on booze.
...there is no spoon
I don't mean to take anything away from the Japanese who are clearly leading in the robotics industry. Especially with technologies like this, humanoid robots like Asimo, and even those creepy robots that have the bad latex skin, these are all really impressive displays of Japan's prowess in this field. More importantly, the control mechanisms are being refined at both the software and hardware interconnects, so this isn't just "robotics", but rather the whole field covers a much broader scope than merely software or just hardware.
Why isn't the U.S. leading in this area? Why have we decided that we're happy enough building Facebook applications? It's sad to see that we aren't as focused on building real systems that will have an actual physical impact on our surroundings. We took Laertes' ridiculous admonition "to thine own self be true" and turned ourselves and our energies into the very worst of what we are as a nation. We have become exactly what the Japanese saw 20 years ago: a nation of lazy, overpaid workers. And, I hate to say it, we are paying the price for that with our jobs.
It's a nice enough demo for a five-axis mill, but these are hardly new nor revolutionary in any way. These have been around for at least a decade, probably much longer.
You could create one awesome looking suit of armor with that.
Troll is not a replacement for I disagree.
The Japanese company celebrated its 50th anniversary last year by using this machine to carve ... a full-scale motorcycle helmet out of one piece of aluminum. No breaks, no joints, the 5-Axis mill simply pivots and rotates to carve metal at some absurd angles.
This has been possible for a very long time. I've seen 5, 6 and even virtual axis mills decades ago that could do this. The software is easier now and the machines have improved tolerances and speeds but the basic technology has been widely used for ages. Multi-axis CNC mills are absurdly useful but not even remotely new.
In other words, nothing to see here. Moving on...
I would be more impressed if it couls take a chunk of Aluminium and carve out a tinfoil hat with the same thickness as one made out of Reynoulds Aluminium foil. Then I would be impressed.
Tsukasa: All I really want, is to be left alone...
Industrial robots are getting precise enough that they're less like dumb machines and more like automated sculptors producing artwork
No, the engineers who built them and the programmers who programmed them are the sculptors, the robots are simply sophisticated knives. They're tools that humans use to create the sculpture.
It isn't artificial intelligence, it's real. It's the programmer's intelligence.
Free Martian Whores!
I'm starting to get involved in CNC machining (hobbyist level). One of the things that is quite clear is that there are really no good open source CAM packages. For that matter, open source 3D CAD has a long way to go, although I have great hopes for FreeCAD (not ready yet, but huge progress in the past year). If someone out there is looking for a challenge, take a look at 3D CAM, starting with 3-axis milling. Toolpath planning is *hard*. Your problem: Here is an arbitrary chunk of arbitrary metal. Here is a list of arbitrarily shaped tools. Here is the work envelop of your machine. Here is a table of chiploads that won't break the tools. Here is a 3D CAD file. Produce gcode. gcode that will not break the tools, not crash into fixtures, not crash the machine, and can start with roughing cuts to carve the initial block to something close, and plan finishing cuts that give you the desired surface finish at the end. A do your debugging where a "crash" can cost hundreds or thousands of dollars in broken tools and machinery.
Well I, for one, welcome our robot metal-sculpting overlords.
If it weren't for deadlines, nothing would be late.
"Leadership" might not translate fast enough to cash in the US to look as though it's worth having. The US metric up to the last year or so, which I hope is beginning to fade, is "can we make our money back on this in a short time?" and the closure of labs like Bell and Xerox PARC reflect this bottom-line thinking. Germans and Japanese alike see nothing "better" in the challenges of design than in those of manufacturing so they have good engineers doing both, and they think longer-term. It's less difficult to sell the leadership argument to their management. The French don't even appear in the contest and that's because all their bright people - who are legion - are theoreticians, they see something not quite nice, or grubby, or something in manufacturing and manufacturing engineers are seen as lower life-forms. If the French could get over that they might place.
The Chinese won't lead, ever, with stolen IP and that's how they do business. They have advanced recipes but when they break, there is no theoretical backing for it. They'll manufacture things a couple lamellae behind the cutting edge until they get over that. Once the ROW catch on you will see the Chinese doing truly wacky things because they will be stealing poisoned IP.
...how to make 10kg product with only 190kg of waste metal !
Can it create clock spare-parts ?
There are scads of youtube videos of multi-axis machining, from impellers to V8 engine blocks, that are several years old. But, way before youtube, in the 1970's, Japanese nine-axis milling machines helped Soviet designers make submarine propellers vastly quieter, meaning subs like the Soviet Typhoon-class were roughly as quiet as American subs had been for a while. The military and export implications of multi-axis milling machine technology was mentioned in US Congress debates at the time: In 1983-1984 the Japanese firm Toshiba sold sophisticated, nine axis milling equipment to the Soviets along with the computer control systems, which were developed by Norwegian firm Kongsberg Vaapenfabrik. U.S Navy officials and Congressmen announced that this technology enabled the Soviet submarine builders to produce more accurate and quieter propellers. So this is by no means new, but it sure is pretty.
Nostalgia's not what it used to be.
Did they glue the transparent aluminum to the regular stuff before carving, or can they treat it to make it transparent later?
but will it blend?
Very nice. But not that unusual for a modern machine tool. Here's a Matsura mill doing much the same thing. It's the software that's interesting.
The current generation of machining software finally has constructive solid geometry that really works. The software can predict where the surface of the work is, as material is removed from it, and can reliably calculate clearances to the tools. I'm very impressed. This really works for arbitrary convex objects now. I've worked on collision detection enough to understand how hard that is.
Coordinating the multiple axes isn't the hard part. That's just relative transformation matrices, which has been done in computer graphics for many years. (Although the newer robot and machining systems understand some of the machine dynamics, and consider inertia. That's new.) It's the modeling of the surface as it changes that's hard.
This is very expensive software, but it's worth it. You need both HyperMill and either SolidWorks or Inventor. You design the part in SolidWorks or Inventor, then use HyperMill to generate the commands for the CNC machine. Total cost is upwards of $10,000. The CNC machine tool itself is relatively dumb; it's just running previously computed moves. The newer machine tools have software to display the 3D model and the tool, so you can check the planned moves against the actual ones when setting up.
Nobody machines consumer products out of solid blocks of metal except as a demo, of course. It takes hours to machine something that can be made in seconds by stamping or molding. Machine tools are used mostly to make stamping and molding dies, and one-off parts. Also, even in modest volumes, you don't start with plain blocks of metal. You cast or forge a blank and machine off the excess.
I thought it was making a T-800 skull.
And I almost bought a one-way ticket to Japan to save man kind.
Steel on toast for breakfast, yum!
Its would've been a much cooler (and scarier) demo if they'd carved out a Cylon head instead of a motorcycle helmet.
This has never happened to me before (for a machine ore piece of hardware).
That helmet is beautiful, but the machine that made it is sexy. I want two. NOW! one on each side of me.
How much is your data worth? Back it up now.
Bite my shiny metal ass!
..., how accurate is it? I've got a project on my 'to do' list which involves cutting a worm and hobbing its mating gear so that it will have an error of better than 5 arc seconds.
Have gnu, will travel.
Ever see gas turbines machined? Goes for cars too. A lot of very cool and useful things wouldn't be possible without awesome machine tools. The physics and math behind these things is pretty amazing too.
If you want to start a revolution, get one of these to sew a shirt.
The USA had these sophisticated machines for a long time (decades) though probably without the current level of software.
They were considered critical national security technology because their primary use was milling components for advanced nuclear weaponry. The specific shapes and ability to form these unusual shapes required for the physics unhindered by manufacturing constraints was considered an essential capability.
No doubt these machines illegal to export and probably restricted from commercial use.
Also they had to collect all of the shavings, because they were milling nice stuff like plutonium, uranium, beryllium and high explosives.
This is so far beyond the simple CNC (read paper tape here) controlled mills we has access to when I started collage. Absolutely amazing. Watching it tilt both the product and the mill in different directions while the mill cutter was inserted into a cavity struck me as a little scary. It looked like it was going to snap the tool off. The compute horse power in this thing must be huge and the designers must have had a blast with developing this product.
does he carve butter like metal as well?
A std 5-axis mill doing what it should do.
Chinese puzzle balls
Donald 'Duck' Dunn: We had a band powerful enough to turn goat piss into gasoline.
Sexy. Yes this is not really impressive unless you find this sort of thing impressive. (I do). Unique? Not really. Difficult to duplicate? Hardly. Fun to watch? Definitely!
This subtractive method of production is pretty old. When we can actually produce in quantity parts with this same precision from a medium in an additive fashion we will have accomplished innovation.
"It's because they're stupid, that's why. That's why everybody does everything." -Homer Simpson
If you look closely in the video, you'll see Deckel-Maho on the machine. It is a German machine. Seki is providing the CAM software. At most, Seki is providing the control, but it looks like Siemens is the default control for Deckel-Maho.
http://www.dmgnippon.com/query/internet/v3/igpdf.nsf/fa8158c4d08f8585c12576a9005695f3/$file/mailing_dmu_hsc_us.pdf
Nobody machines consumer products out of solid blocks of metal except as a demo, of course. It takes hours to machine something that can be made in seconds by stamping or molding. Machine tools are used mostly to make stamping and molding dies, and one-off parts. Also, even in modest volumes, you don't start with plain blocks of metal. You cast or forge a blank and machine off the excess.
That's not entirely correct. Machined parts are actually relatively common in the motorcycling industry for a number of reasons.
- Often the part is a one off, or small batch product as you mentioned.
- Billet parts are in demand for their perceived beauty, or value.
- Often, machined parts are used for their high strength to weight ratio.
Some of the following machined products are widely available in the industry:
Upper and lower tripple clamps
Hand Controls
Foot Controls
Engine Covers
Engine covers are an especially interesting case. Most modern cast aluminum covers are very lightweight and thin. In the event of a crash, they often wear through, leaking oil (on an active race-track,) permitting dirt into the engine, and rendering the vehicle inoperative.
Many race organizations now require aftermarket covers on many bikes.
A lot of silly products, such as reservoir caps are also available as CNC machined parts.
Mis-read the article. I'm not sure what Daishin Seiki actually does, looks like a prototype shop. They use hyperMill (tm) from Open Mind Technologies http://www.openmind-tech.com/.
So it is basically Daishin Seiki's demo of what they are capable of with a Deckel-Maho (German) machine and hyperMill (US)CAM software.
Granted, they appear to be multi-national.
http://www.openmind-tech.com/en/the_cam_company.html
I'm lost as to why Daishin Seki is getting credit here, other than a poor write up on someone's blog and a cool demo.
And on a personal note, I need to get my reading comprehension checked, or check my meds...
A helmet?
About five years ago at Oshkosh, Williams International was showing-off a compressor turbine hub assembly for their EJ-2 engine. It was milled from a single piece of metal; hub, compressor blades, everything. One piece.
A 'helmet.' Pbbbbbttttt. . . .
Regards;
Impressive stuff. Their next project now can be the infinitely detailed surface of the Mandelbulb:
http://www.skytopia.com/project/fractal/mandelbulb.html
Why OpalCalc is the best Windows calc
if it saves 10% of the time of a $100k/yr machinist or tool programmer. And those packages do a LOT better than that. $100k for the software would probably be economically sound, except that the mfr would have to figure out a way to sell it in pieces, because $100k capital investment is pretty steep for something that you can't easily repossess or sell at auction.
Or the mathematician that informed the programmer. Or the teacher that taught either. Or the hack that built upon an idea. Or the maverick who first chipped stone. Or the second maverick that observed and copied. Or the high tech black slab that eventually resulted. Or the promethian lesson the slab eventually imparts to some clever observer.
It's information all the way down, son. Consider it a backup strategy.