Designing a 3D METAL printer - Page 3
Close
Login to Your Account
Page 3 of 4 FirstFirst 1234 LastLast
Results 41 to 60 of 74
  1. #41
    Join Date
    May 2009
    Country
    UNITED STATES
    State/Province
    Wyoming
    Posts
    2,789
    Post Thanks / Like
    Likes (Given)
    5979
    Likes (Received)
    4168

    Default

    By now, on a thread full of metalheads somebody should've mentioned the obvious solution to the curling you get from applying successive hot layers: alternate the deposition passes, or every few deposition passes, with a peening pass. Just like welding cast iron with an air chisel over your shoulder, only the air chisel is driven across the layer by a ball screw...

    I don't need a piece of anybody's Nobel prize. If I'm not home just shove your money under the door. Thank you.

  2. #42
    Join Date
    Jun 2002
    Country
    CANADA
    State/Province
    British Columbia
    Posts
    2,065
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    1342

    Default

    Hi metalbot:
    OK, you've asked some very specific questions about technical issues that I think we can put our heads together to try to answer.
    With regard to a build platform that you intend to move up and down with exquisite precision.
    A 150mm x 150mm platform is not very big and can be moved in several ways.
    You can provide a vertical leg on the platform, making it like an upside-down "L", and guide the platform by bolting precision sliding elements onto the vertical leg of the L, then providing a column on which the L can ride up and down, all controlled by a single elevating screw.
    In this design which is used commonly on machine tools like turret mills, you need control only a single screw, and a stepper motor will do that for you at low cost.
    Alternatively, you can make a 4 poster bed design, where you electronically or mechanically slave 4 precision screws together, one at each corner, and use those to elevate and lower a platform.
    Some woodworking thickness planers use this sort of design.
    You could build a precision scissor jack, move the platform up and down with a single cross screw and map the non-linear motion to control the build increment.
    Other designs will suggest themselves, but these are common ones.

    Your requirement for precise consistent build layers will mean the mechanical components will not be cheap, and you cannot build as rude and crude as many of the homebrew CNC hobby toys out there, so expect to pay substantially for that precision motion capability no matter what design you choose.
    If you don't have access to your own machine tools, budget a couple of thousand dollars to build this part of the system.
    If you can live with looser tolerances you can probably save a bit, but not a lot, assuming you have to farm this part of your project out to a machine shop.

    Now, moving the build platform down is one way, but alternatively, you could move a couple of rails at the edges of the platform up instead, and rest the scraping squeegee on those rails.
    This way the build platform doesn't have to move at all, and when the entire system is packed with stray powder, that may well be the better way forward.
    The main problem with this approach, of course is that to maintain the focal plane of the laser, it has to move in tandem with the rails.

    As far as the powder distribution and leveling; I think you have two choices:
    1) Shamelessly copy those who have gone before (this means you have to get access to one so you can see its guts)
    2) Try the different options and see what happens.
    Rigging a manually powered scraping blade between two rails is simple and cheap to try; ditto for swiping the wife's rolling pin for an afternoon.
    My gut tells me the scraper blade will do better than the roller, but try them both.

    These problems are the straightforward ones to solve; the lasers will be a more significant challenge.
    EOS uses 400W lasers or thereabouts with pretty much continuous output, and you have to steer them either with mirrors and galvanometers or with a cartesian gantry.
    Moving them precisely on a gantry at high speeds without jiggle requires a really robust platform, again not cheap to build and is slow to boot, so it's probably galvanometers and mirrors.
    Now you have the incident angle of the laser pulse and all the complexity that decision produces to deal with, but all the problems are solvable with a liberal application of money or gritty bloody mindedness and perserverence.
    To prevent the optics from fouling, you need to be able to either shield them or suck them clean without disturbing the powder bed; yet another subsystem to design and build.
    You also need to cool the laser system.

    Atmosphere control, as you've alluded to, is yet another challenge; flooding a big chamber with nitrogen or argon is not cheap and requires flow control to keep from blowing stray powder all over the place as well as a mechanism to avoid an overpressure event and subsequent blowout.

    Shielding the mechanical systems from powder ingress is another challenge you can't overlook; if you neglect this your mechanism won't survive very long, or may not move at all.

    So if you intend to move forward with this project, there's much to design and much to build.
    You won't get there without partners, so you need to start looking for buddies who have the skills and equipment to tackle these problems.
    The programming and motion control is going the be the least of what you have to solve, unless you intend to make only a very crude system.

    I'd personally make a scissor lift platform, run the lasers with a hybrid cartesian gantry for gross positioning and galvanometers/mirrors for fine positioning, and start with a plastic squeegee, operated by hand.
    I'd see if CO2 lasers will provide the wavelength and continuous power output you require, and try to source cheap ones from China, accepting the poor reliability of Chinese components for these early experiments, or alternatively try to scare up a junked one and rebuild it.
    I doubt a flashlamp or diode pumped YAG laser will do the job for you; their maximum pulse frequency is awfully slow for this application.
    Good luck with your project; it's going to be a big one, but if you pull it off, it'll be extremely cool!!
    Cheers

    Marcus
    Implant Mechanix – Design & Innovation - home
    Vancouver Wire EDM -- Wire EDM Machining

  3. Likes omegaclass liked this post
  4. #43
    Join Date
    Apr 2013
    Location
    Canet, France
    Posts
    6
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    0

    Default

    Hi lazlo!

    I don't want to disagree but... why would a company which earned $147.359 billion in revenue last year - and is the 4th largest company ever according to Forbes - buy up one of the main sintering companies (morris technologies)? Just to sit on it?

    For run of the mill parts like the vast majority of mechanisms require however, it has zero hope of competing now and forever:
    That is an accurate and obvious statement, I am not sure who said 3D Printing will replace everything?? I'd like to find the bloke... it is just one of those misnomers that inevitable appear around technology like this. The benefits of the technology show up when complexity is needed, it's a simple graph.

    Morris Tech apparently - according to an interview with Morris of Morris Tech - make modifications to the Sintering Machines that they buy in from EOS, I thought that was worth noting.

    That was indeed a very interesting post implex - there is no doubt, these are key challenges.

    (1) It's a process that needs lots of time and a lavish energy budget to make even a single small part, and both go up as the cube of the build size.

    The EOSint M 280 uses an average power of 3200W all the way through to 8500W. That, I assume is largely to do with their 200W Lasers and in built nitrogen generators. The lasers we are looking to use will consume a total peak power of 500-1kW (see how below). That is comparable to having a heater running all day.

    As was said, it is harder to speed up the build time (without using bigger lasers and thicker layers = lower resolution) than to run parallel printing.

    2) The fact that it's a thermal process means distortion and thermal stress in the part as the part is successively built, with a new hot layer being deposited onto a cold layer (which is why my C shaped part curled so badly).

    This is one of our fundamental problems with the process and there is no clear cut way to solve it - save for heating the entire build platform up to 1500C and stuffing up point 1.

    Which brings us to Oldwrenched top notch Ideas.

    (4) There is a basic lack of material versatility in two profound ways; first the difficulty of avoiding cross contamination as a particular powder is changed out to a new, different powder; second, the limited powders available for processing.

    For me... personally... I will be happy to just print in titanium over the next few years, perhaps this is a problem for commercial systems however this just is not enough of a concern yet. The next step up will be multi material printing, combining metals and plastics, an there is a member of our forum who is making a machine to do just that using hypersonic deposition, but that is a completely different system...

    (4) It's severely anisotropic; in X and Y the material properties are quite good; in Z they're terrible, just as is true for every layered build process in any material.

    Quote from the EOS white paper...

    When building parts up layer-by-layer by sequentially
    melting and re-solidifying individual lines of material, one
    intuitively expects to see a resulting structure such as that of
    cobalt-chrome in Figure 6 with „weld lines“ and fine grain
    structure. However lasersintered parts in Ti6Al4V show no such
    layered structure. Instead a dendritic structure is observed with
    crystals oriented perpendicular to the applied layer, and having
    a height much greater than the layer thickness (Figure 6.3).
    This phenomenon can be explained by the laser energy of each
    instead of increasing it. Subsequent investigations indicated
    that this unexpected behaviour is caused by the formation of a
    duplex-type steel structure with interlocked phases of austenite
    and martensite/deltaferrite during the extremely rapid melting
    and re-solidification. Figures 6.1 and 6.2 illustrate this effect.
    During conventional heat treatment, the duplex-type structure
    remains and subsequent martensite formation is hindered, which
    explains the different results.
    vector remelting part of the previously solidified layer below and
    thereby removing the existing boundary, and the subsequent
    recrystallization causing the crystal to grow through the layers.
    It should be noted that the material structure and properties
    depend on the build strategies (e.g. laser exposure patterns) and
    parameters used. The examples quoted above all refer to specific
    build situations and the results may vary when different
    parameters are used.
    I would like to challenge EOS on that claim by having them print two small bars in their machines. One horizontally printed the other vertically printed and doing a stress test along with forged equivalent (control) bars, see what breaks first.

    @ Marcus - wow! That is a lot to digest . Thank you for taking the time to write this information!
    You can provide a vertical leg on the platform, making it like an upside-down "L", and guide the platform by bolting precision sliding elements onto the vertical leg of the L, then providing a column on which the L can ride up and down, all controlled by a single elevating screw.
    In this design which is used commonly on machine tools like turret mills, you need control only a single screw, and a stepper motor will do that for you at low cost.
    This jumped out as the most logical way to go. So one would need a ball screw, a stepper motor and 'precision sliding elements'. A couple of thousand dollars (more actually), while not trivial, is expected given the nature of the work. What is the general consensus on chinese ball screws, of which there are a myriad on Alibaba : Ball Screw-Ball Screw Manufacturers, Suppliers and Exporters on Alibaba.comBall Screws

    I would want one with the the finest thread possible. It would not need to be too long either (150mm). These for example: Linear Motion Ball Screw - Buy Ball Screw,Ball Lead Screw,Linear Moiton Ballscrew Product on Alibaba.com .
    Now, moving the build platform down is one way, but alternatively, you could move a couple of rails at the edges of the platform up instead, and rest the scraping squeegee on those rails.
    This way the build platform doesn't have to move at all, and when the entire system is packed with stray powder, that may well be the better way forward.
    The main problem with this approach, of course is that to maintain the focal plane of the laser, it has to move in tandem with the rails.
    I have trouble visualizing this system... maintaining the focal point the full length of the Z-Axis (150mm for example) will complicate things.
    As far as the powder distribution and leveling; I think you have two choices:
    1) Shamelessly copy those who have gone before (this means you have to get access to one so you can see its guts)
    2) Try the different options and see what happens.
    Rigging a manually powered scraping blade between two rails is simple and cheap to try; ditto for swiping the wife's rolling pin for an afternoon.
    My gut tells me the scraper blade will do better than the roller, but try them both.
    Good one! I do live relatively close to Germany, perhaps I'll pay a visit to EOS and ask them if I can pretty please take a peek under the hood (of course be up front and tell them I want to open source what I find).

    Option no.2 is underway. I agree, the biggest problem when dealing with powders at this scale is electrostatic adhesion, so we want a sharp edge and minimal contact with the powder while spreading. We would also want to keep moisture out of any powder we put in the system.

    With regard to the lasers... this is our single greatest challenge and there are two ways we are going to solve it.

    The two main laser types we are looking at are 1 - Diode Pumped Solid State (DPSS) Lasers. This would be a YAG Crystal rod being pumped with a high wattage 808nm laser. 2 -Just using the laser diode on its own (main drawback being a lower beam quality as compared to a DPSS laser).

    We could just as well use a Flash Pumped YAG laser, but flash lamps have horrendous lifetimes and are expensive to replace, which will bite is in the end ( in the form of maintenance costs).

    There are some pretty good threads no this subject if you are interested over on the Metalbot Forums: metalbot.org • View forum - Lasers

    The effective power of the laser is directly related to it's focus. So EOS systems use a 200W laser that has a focus of 100-500um which is really bad for parts resolution, but can melt large surface areas quickly. Just as an example, if we were to use a 20W with a focus of 10um we will have the same sintering ability (all be it slower) as the EOS laser.

    That combined with much smaller powders as compared to those used by EOS should make it possible to use smaller and much cheaper lasers in a 3D Metal Printer.


    60W laser diodes are readily available for ~$350, the trouble is that their beam quality is poor, but tests have show they have the ability to melt metal powders, so this is very promising. To improve the beam quality as has been mentioned we need to put the laser diode beam into a Nd:Yag crystal and turn it into a DPSS laser.

    Here is a 35W un focused laser, you can really see the power: https://www.youtube.com/watch?v=3OZD3UVhEZk

    DPSS Laser cost more than diode lasers simply because they use more exotic materials and are more complex, but it is possible to make one in true DIY fashion for a relatively low cost.
    Moving them precisely on a gantry at high speeds without jiggle requires a really robust platform, again not cheap to build and is slow to boot, so it's probably galvanometers and mirrors.
    The gantry has been discounted. We have a couple of experts in laser scanners (galvos, mirrors, drivers, etc...) working on laser scanners of all sorts .

    To prevent the optics from fouling, you need to be able to either shield them or suck them clean without disturbing the powder bed
    A very important point to note.

    flooding a big chamber with nitrogen or argon is not cheap
    Thankfully it is not a massive volume to fill, you can buy over 1000L of argon for under $130. Perhaps sealing off the build chamber will reduce argon consumption further.

    Powder ingress is a problem. But there are only three parts that will potentially suffer from this, the powder spreader, valves that drop the powder from the powder holder and the build platform itself.
    You won't get there without partners, so you need to start looking for buddies who have the skills and equipment to tackle these problems.
    That is why this project is open source! There is no way it would have any chance being closed. Thankfully I have been pleasantly surprised by people who are willing to help and have the type of equipment we need! But this project covers so many fields and disciplines and there is such a long way to go...

    I'd see if CO2 lasers will provide the wavelength and continuous power output you require
    Unfortunately CO2 lasers have a long wavelength of 10.64um which is I'll suited to metals, high powered CO2 lasers will not even mark metal, which is why you don't see them used in metal engraving machines. YAG lasers or laser diodes, as has been mentioned, are our best option (1064nm of 808nm).

    Thank you Marcus, that was a truly helpful, well informed post and has gone a long way to sorting things out in my mind.

    Have a great day!

    Jethro.

  5. #44
    Join Date
    Apr 2013
    Location
    Canet, France
    Posts
    6
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    0

    Default

    Just a quick update for the interested... there are some top notch ideas going around!

    We have decided, for now at least, to put a "laser scanner" on the back burner and go with a Gantry. So some form of CNC software in combination with a slicer will be used.

  6. #45
    Join Date
    Feb 2013
    Location
    Houston TX
    Posts
    3,371
    Post Thanks / Like
    Likes (Given)
    15
    Likes (Received)
    1502

    Default

    I want to see a finished part before I sell the machine shop to buy one.
    Have you solved any of the technical problems like weak parts, horrible finish, and taking forever?

  7. #46
    Join Date
    Jan 2003
    Location
    Canada
    Posts
    10,603
    Post Thanks / Like
    Likes (Given)
    4341
    Likes (Received)
    2793

    Default

    Matsuura lumex avance-25 is indeed " where it's at " was sent the brochures a while back, pretty neat system with integrated finish milling to size.
    Didn't ask the price....

  8. #47
    Join Date
    Dec 2013
    Location
    indiana
    Posts
    3
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    0

    Default

    I have been reading the comments here and it seems very few folks here have anything relevant to the title of this post except for a few. i believe it is possible to build a unit the size of a large tool box and produce the quality seen in the 600K units. first we must identify the kind of laser used and if it is cost prohibitive can we use other laser types like green lasers, or what type of conditions must exist inside the work envelope to maximize the bonding/melting of the power? how about instead of powdered metal a fine metal wire be fed into the laser heat zone to do the same thing?


    I for one do not want to see the big boys owning this technology for themselves such that they can charge whatever price they want for the service they can only provide. when the people have the power of production governments have less power when the government owns the power of production we have devastating depressions.

  9. #48
    Join Date
    Jan 2011
    Location
    Estonia
    Posts
    676
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    174

    Default

    Someone decides "I will build a 3d metal printer"
    No prior knowledge about mechanics.
    Main concern is "ball screw"
    All the problems which are currently unsolvable "will be solved in future" because "anything I dont understand should be simple".
    And of course "the will is most important" and "somebody has to do it" and "fuck the big companies".

    Lets say, if Woody Allen decides that he wants to be a pro-boxer and beat up Valujev, then of course all he needs is "strong will" ???

    PLEASE DO YOUR HOMEWORK AND DONT PUT UP DRAWINGS MADE IN PAINT OR CORELDRAW !!!

    "Believing" that you can do something gives no bonus points, DO something.

    Ok, now to details:

    How do you prevent powder "dusting" when blade goes over it ? Increasing pressure will result in jumping particles and uneven surface. Low pressure results in grooves. This is probably the main issue of your machine, ballscrews etc is piece of cake.

  10. Likes N/A, digger doug, Oldwrench liked this post
  11. #49
    Join Date
    Feb 2009
    Location
    Midwestern MN/Wi USA
    Posts
    1,182
    Post Thanks / Like
    Likes (Given)
    68
    Likes (Received)
    315

    Default

    One of the challenges is the material dynamics of being heated and deposited. Expansion, rapid cooling, material flow....yadaaaadadaydyaydyadadaydyadaydaydyaydyayd yadyaydaydyada It really goes on forever and you compromise in the end. But these are the reasons youll never see ABS plastic being put down in .100" layers. The same reason you have ultra expensive raw materials to start with. Did you know a few lbs of metal sinter material can cost thousands? Besides making it cheaper you better count on adding the cost of raw materials because in the end the machine should produce a working model.

  12. #50
    Join Date
    Sep 2013
    Location
    Canada
    Posts
    12
    Post Thanks / Like
    Likes (Given)
    3
    Likes (Received)
    1

    Default

    I don't think that a ballscrew is really necessary, they're typically used as backlash eliminators. since you won't be having rapid up and down movements, backlash shouldn't be an issue. not only that, but it's a loaded system, so if you went with a precision leadscrew it would always be forced down. as long as you pick up (measure) the location of your print table before every print, wear on your leadscrew nut would be negligible.

    I would have a feeder piston and a printing piston of the same size and run them off of the same stepper. as the printing piston drops, it creates a void of a certain volume and the feeder piston rises, creating an excess of material of the same volume. the roller comes across and flattens it.

    The feed and printing pistons could run on nearly identical mechanisms, cutting down on design time, with the exception that the feed plate doesn't need to be anywhere near as parallel. a fixed leadscrew in the centre of each plate would do the trick, with linear track to keep parallel with the printing plane. cut some gear teeth onto the leadscrew nuts and put a thrust bearing under them and you're laughing. if you need greater accuracy than running a stepper on quarterstep will give you, add in a gearbox.

  13. #51
    Join Date
    May 2007
    Location
    KC
    Posts
    1
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    0

    Default

    What ever happened to this?

  14. #52
    Join Date
    Feb 2012
    Location
    southern Or.
    Posts
    588
    Post Thanks / Like
    Likes (Given)
    592
    Likes (Received)
    404

    Default

    All we need is a few breakthroughs and it will take over the world.

  15. Likes digger doug, Oldwrench liked this post
  16. #53
    Join Date
    Aug 2006
    Location
    New Zealand
    Posts
    1,061
    Post Thanks / Like
    Likes (Given)
    58
    Likes (Received)
    208

    Default

    Hi
    I saw a question about the issue of evenly spreading a thin layer of power and problems related to static charge.
    Every laser paper printer uses static charge to evenly distribute toner on a roller. Just wondering if there is the possibility of technology transfer here.

  17. #54
    Join Date
    Dec 2014
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    87
    Post Thanks / Like
    Likes (Given)
    9
    Likes (Received)
    11

    Default

    Quote Originally Posted by SND View Post
    Matsuura lumex avance-25 is indeed " where it's at " was sent the brochures a while back, pretty neat system with integrated finish milling to size.
    Didn't ask the price....
    I did... 850K or so. By the time it is installed and you buy material more it, budget 1M.

    AG

  18. #55
    Join Date
    Sep 2010
    Location
    Vancouver Canada
    Posts
    611
    Post Thanks / Like
    Likes (Given)
    390
    Likes (Received)
    187

    Default

    Interestingly Hermle has developed a machine Hermle MPA - Technology but they seem to be only offering it as a service rather than selling machines. Maybe they're spending a few years working with it to get all the bugs out before selling them. And I'm sure they'll be in the million range as well.

  19. #56
    Join Date
    Mar 2006
    Location
    Yorkshire, England
    Posts
    317
    Post Thanks / Like
    Likes (Given)
    9
    Likes (Received)
    27

    Default

    Quote Originally Posted by ewlsey View Post

    The metal printers I have see use a bronze matrix to sinter the base metal. So if you want to make something from steel, it's only a fraction steel, and only a fraction as strong.
    Really?
    You've not seen the Solid Concepts 1911 then? They tested it successfully with factory rounds without heat treat ;-)

  20. #57
    Join Date
    Oct 2005
    Location
    erie,pa
    Posts
    7,985
    Post Thanks / Like
    Likes (Given)
    14735
    Likes (Received)
    3757

    Default

    I tried to read each post, and I didn't see this mentioned:
    Sciaky | Metal 3D Printing Solutions & Welding Systems

  21. #58
    Join Date
    May 2005
    Location
    CA
    Posts
    846
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    127

    Default

    Quote Originally Posted by rcoope View Post
    Interestingly Hermle has developed a machine Hermle MPA - Technology but they seem to be only offering it as a service rather than selling machines. Maybe they're spending a few years working with it to get all the bugs out before selling them. And I'm sure they'll be in the million range as well.

    Yes... the additive method has been around for a while: Gas dynamic cold spray - Wikipedia, the free encyclopedia

    More specific examples here: Current Trends in Cold Spray Technology: Looking at the Future - Materials Today

    ... and very typical of Hermle to test for quite some time before releasing a product (machine).

    PM

  22. #59
    Join Date
    May 2016
    Country
    UNITED KINGDOM
    Posts
    25
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2

    Default

    Quote Originally Posted by metalbot View Post
    Hi lazlo!

    I don't want to disagree but... why would a company which earned $147.359 billion in revenue last year - and is the 4th largest company ever according to Forbes - buy up one of the main sintering companies (morris technologies)? Just to sit on it?
    The number of times that I've seen a middleware technology company bought out by one of the big monopoly companies just to shelve the technology... it's not remotely uncommon.

    Got to bear in mind that the main cost of designing jet engines is having access to the prototype parts to test them. Being able to slash at design iteration time would facilitate many smaller players to enter the market. GE are smart to bring the company in house not just to use the technology but to lock competitors out.

    Not to say that GE won't be making jet engines using the parts, potentially not just for prototyping but for production. Also bear in mind that GE are a big company, they don't just make jet engines, and they may be planning on refining and marketing the machines themselves. We are talking about Edison's behemoth here, they're a technology company rather than being tied to one particular market.
    Last edited by bishopdante; 06-06-2016 at 12:18 PM.

  23. #60
    Join Date
    May 2016
    Country
    UNITED KINGDOM
    Posts
    25
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2

    Default

    Quote Originally Posted by CarbideBob View Post
    Why does the field of 3-D printing feel so much like robotics and AI in the 80's ?
    Or... the internet in 1997.

    Quote Originally Posted by CarbideBob View Post
    "All we need is few break-throughs and these will certainly happen given the pace of technology".
    This despite the fact that those with unlimited cash have not been able to do it well.
    I do think it will work eventually, but I don't think I will be around to see it.

    Not to dissuade people from going into it.
    I can see it as a great way to suck money from investors in the short term. I know some guys who made a lots of money chasing artificial intelligence and robots in every house.
    If I believed in this I would not be going open source, always good to market "magic" to those with big wallets.
    Bob
    hype-cycle-general.jpg

    ^ this diagram illustrates what's called the "hype cycle".

    For any new technology it's a long, arduous road from the peak of inflated expectations through the trough of dissillusionment before the plateau of productivity can be entered.

    For the internet, we're maybe entering the plateau of productivity today for *really basic webpages* (and shady government surveillance). The 1995 dream has been delivered. Most people forget how terrible 1995 internet was.

    For VR it'll take a good while longer, but anybody who works with a screen and mouse will explain that proper 3D interaction isn't a luxury, it'll be a massive time-saver. Half the reason people want 3D printed prototype parts is because you can't really learn much by looking at a rendering on a 2D screen.

    AI... that's been enormously over-inflated in terms of what people should expect from it, it'll be a long time before AI has a sense of humour or style, but self-programming self-maintaining code is already becoming something of a necessity, codebases are becoming sufficiently unmanageable at scale that a lot of people are working on using AI to do a lot of the heavy lifting of nuts 'n bolts programming. Creating a production-ready technology is about learning the limitations and using each tool in its proper place. We're no more likely to eliminate the mill/lathe than we are to eliminate pencils and pieces of paper in a design office. Of course I have a wacom, but sketching is more often than not done on paper. Technology doesn't get replaced, it gets expanded.

    Additive machining, ie 3D printing will never be a productivity monster, but it'll be able to produce parts of immense complexity, the most important feature being that it'll make metal-matrix composites much more controllable.

    In terms of additive metal, sintering isn't the only viable method:

    Last edited by bishopdante; 06-06-2016 at 12:21 PM.


Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •