Assembling Your Own 3D Printer 129
Posted
by
Soulskill
from the build-to-build dept.
from the build-to-build dept.
adeelarshad82 writes "Following a tour of a 3D printer factory, analysts at PCMag wanted to explore the option of building a 3D printer themselves. With the help of a 3D printer manufacturer, Buildatron, they were able to compile a step-by-step guide on how to build a 3D printer."
!EarlyAdopter (Score:3, Insightful)
I think I'll wait until 3D printers can 3D print other 3D printers.
Re: (Score:1)
Re: (Score:3)
Not me. If 2D toner is more expensive by the ounce than imported Russian caviar, 3D toner will be more expensive than highly enriched uranium.
Re: (Score:2)
That's why you wait until 3rd party vendors swoop in to sell their own replacement toner for a less insane price...
Until they get hit with DMCA lawsuits and other legal bullshit, that is... [wikipedia.org]
Either way, stay far the fuck away from any 3D printer made by HP, Epson, or any other major manufacturer...
Re: (Score:1)
Either way, stay far the fuck away from any 3D printer made by HP, Epson, or any other major manufacturer...
The main problem with the HP 3-d printer driver, is its 12315 gigabytes, and is bundled with the yahoo toolbar.
(I'm 99% sure its HP that has 350 meg printer drivers, but aren't they also the yahoo toolbar people?)
Re: (Score:3)
I'm 99% sure its HP that has 350 meg printer drivers...
They're one of them. Sometimes you can avoid that by skipping the crap that comes with the machine and downloading the "basic" version of the driver. The bundled one is 349.5 MB of UI designed to sell you HP's ink, derived from genuine unicorn tears.
Re: (Score:1)
What will they call it?
The HP Protosmart 3810 e-All-in-One. It will Print, scan and Copy!
Re: (Score:2)
With the current system it's just large rolls of plastic filament in either 1.75mm or 3mm diameter. And 1kg of ABS filament is around 45$
Re: (Score:1)
Yep. Makerbot also sell their Replicator for around $200. I think this is a 2 colour version (dual print heads) and $48 for the spools.
http://www.makerbot.com/ [makerbot.com]
Interesting stuff to make too!
http://www.thingiverse.com/thing:13101 [thingiverse.com]
More pics:
http://www.pcauthority.com.au/Gallery/288483,in-pictures-why-the-makerbot-replicator-is-one-of-the-best-toys-ever.aspx/1 [pcauthority.com.au]
from CES
Re: (Score:1)
:( Ignore the price.... It's more like $1,800 and not $200.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
This is where I exist on my current desire for one of these. Having just spent a few hours hacksawing a plastic gizmo apart then trying to figure out how to mount some more securely fitted batteries to it, I'm struck by the notion that it would've been really nice to just 3D print a slightly more snug battery compartment.
Re:!EarlyAdopter (Score:4, Informative)
I think I'll wait until 3D printers can 3D print other 3D printers.
Actually a RepRap can print 50% of a new RepRap. You just need the metal bits and circuit boards.
Re: (Score:2)
I think I'll wait until 3D printers can 3D print other 3D printers.
Actually a RepRap can print 50% of a new RepRap. You just need the metal bits and circuit boards.
There is also a hack for the original makerbot involving many little printed bolt together pieces. So makerbots can more or less replicate themselves too.
If you allow "subtractive machining" as a 3d printer in addition to your "additive machining" then metal lathes and milling machines have been replicating themselves for well over a century...
Re: (Score:1)
Yes, metal lathes and milling machines can *only* reproduce with their own help. This always made me curious what the process is for creating a lathe / milling machine with finer tolerances than the present state of the art.
Beam epitaxy? (Score:3)
You can pretty much get any tolerance you want.
-- Terry
Re: (Score:2)
Re: (Score:2)
I've always wondered this too - and I suspect your conclusion is in fact the reality, judging from the number of scientific processes which depend on exactly this idea (or are absurdly reproducible by it - such as how STM tips for atomic resolution can be made with wire cutters and platinum-iridium wire - because one atom will always be slightly closer then the others).
Re: (Score:2)
Re: (Score:2)
Which is the silly part, as the metal bits are the most expensive part of the machine (happy Ultimaker 3D printer owner here)
Re: (Score:2)
50%? I think you might be seriously underestimating the importance of "the metal bits and circuit boards". ;)
Re: (Score:2)
I don't know about that. You couldn't go to 45 nm, but 1 um is a pretty accessible feature size by UV light based processes, and more then enough to produce basic microcontrollers. The real problem is finding sufficiently pure semiconductor material - but on the other hand, at 1um you're far less susceptible to having your chip wiped out by an imperfection.
Re: (Score:3)
They took an opensource project, added some shoe polish to some parts, then put a huge metal box around it and cut their name into the box.(also, price markup).
Re: (Score:2)
Like RedHat? Or Apple?
Re: (Score:2)
I wish someone would come up with a nice easy and clean way of making PCBs at home. Etching with chemicals and/or UV light is not much fun, and it is tricky to get good tolerances and consistent quality from cheap equipment. Then you have to drill it, deal with vias somehow and assemble without a silkscreen.
Re: (Score:2)
CNC is the answer to this really (though don't take my word for it, since I have a small CNC currently non-functional while I try to rig up a better spindle for exactly this type of task).
Re: (Score:2)
Can't they self replicate? (Score:1)
Can't I just have someone with a 3D printer print me out a 3D printer of my own? If it takes 24 hours to print one out, it would only take a few weeks to print out a million of them.
Re: (Score:3)
Well they can print parts... you still need to assemble them. Good luck doing that with millions of them in only a few weeks.
So it's an assembler bot you need? They're a bit harder to assemble and it normally takes a woman 9 months to produce one and then lots of further work until it can actually assemble things itself, but with enough women you should be able to produce your assembler bot army in a few weeks.
Complicated (Score:1)
Looks very complicated, but I suppose that once you've built your first you can simply use it to build all future printers for nothing.
Suckers :)
Re: (Score:1)
How original!!!
Yeah, your comment is very original. It's only the third time you make it in the discussion of this story. :-)
Re: (Score:3)
Sorry guys my 3d printer made three copies of bennomatic, now look what happens.
Re: (Score:2)
Re: (Score:3)
Re: (Score:2)
Hmm (Score:4, Insightful)
Looks a lot like someone put a RepRap Prusa Mendel in a box, and pretended it was a new product.
Re: (Score:1)
Funny how a Kit ends up looking like the product isn't it.
Re: (Score:1)
FEATURES:
*Based off of the Prusa Mendel architecture
So yes, they put a Prusa in a box.
Re: (Score:2)
Looks a lot like someone put a RepRap Prusa Mendel in a box, and pretended it was a new product.
Hilariously, that's how they self describe themselves, more or less
http://buildatron.com/about-buildatron [buildatron.com]
If you know the relationship of Ubuntu to Debian, then buildatron to reprap is a pretty close analogy. Not exact, but close enough. I like them. I'm still building my own Mendel by myself instead of getting an assembled model from buildatron, but not being one of their customers doesn't mean I can't think they're good people. (Am I too negative?)
Re: (Score:2)
Making cash from someone elses product,call it your own by slapping a catchy logo onto it.
Good going America, you're doing it the chinese way, only a shitton more expensive.
Re:Hmm (Score:5, Informative)
Indeed. And the worst part is, it's silly expensive!
If you want a 3D printer DO NOT GET THIS. Get something from Makerbot or Ultimaker, they sell easier to build kits that give higher quality prints for less money. RepRap is a fun project, but it takes quite a while to get usable results (lots and lots of tweaking). I have an Ultimaker myself, and took me 8 hours to build and get my first print working.
As for people wondering about the quality of these kinds of machines: http://daid.mine.nu/~daid/IMG_20120125_211716.small.jpeg [daid.mine.nu] this is printed on mine.
Re: (Score:2)
Re: (Score:2)
The utilimaker kit is €1194, $1500 in today's trade. If the quality is that much higher, seems like a wise choice, but it's not cheaper than the other kits out there.
Re: (Score:2)
3D printers suck (Score:5, Insightful)
Re:3D printers suck (Score:5, Informative)
To some extent you can bolt an extruder onto your existing CNC mill. That is my "rep-strapping" plan. Lots of details at this link.
Most of the time and money is in the 3-d robot part that does all the positioning. An extruder is actually pretty cheap.
http://reprap.org/wiki/EMCRepStrap [reprap.org]
The other issue, as a machinist, I can verify that accessories that a mill requires are about twice the cost of the mill, and the accessories a lathe requires are about the cost of a lathe. So, I bolt my $350 rotary table to my $500 mill, stick the $50 7/8 inch gearcutting arbor in the spindle, clamp a $50 expanding arbor into my $125 chuck attached to the rotab, supported on the other side by my semi-homemade tailstock that cost about $100, then stick a $25 involute gear cutter in the arbor, I'm not gonna add it up, but just to cut a simple gear out of a blank machined on my lathe, is gonna cost almost as much in accessories as the mill itself. Then add a bunch of clamps, a decent vise, a much of endmill holders unless you're one of those collet people, it adds up, man.
At least theoretically a "utilimaker" thing only requires the additional purchase of a power cord.
Re:3D printers suck (Score:5, Informative)
A friend who has done this tells me it's somewhat of a waste of time because the 3D printer wants to move an order of magnitude faster than the max speeds most mills are capable of driving. (Since it's adding toothpaste, rather than cutting away metal, maybe that's not so surprising.) He ended up buying a Thing-o-Matic (and doing a *lot* of re-engineering to get it working reliably) but now he's thrilled with it. It doesn't replace his mills and lathes, but it sure is a convenient addition. He's all oh the thumbwheel broke off my micrometer: I'll print a new thumbwheel bracket. The windshield mount on my recumbent broke, so I'll print a new one. He's printed plumbing parts, cookie cutters [softsolder.com], centering adapters for optics, replacement bar handle clamps [softsolder.com], you name it, and there's no setup or clamping or accessories or anything like that -- not even alignment. He just emails the completed gcode to the machine and goes in twenty minutes later and takes his new item off the stage. I'm dead envious.
That might be different if you're using a servo-based mill with fast ballscrews, but for steppers with fine-pitch threads, well, my CNC is pretty rattly and jiggly when it's driving around at 10ipm and his Thing-o-Matic can run at 500ipm.
Re: (Score:3)
It doesn't replace his mills and lathes, but it sure is a convenient addition.
Ah, the sign of someone who knows their trade, rather than a boy playing with toys: knowing the right tool for the job.
Re: (Score:2)
He's all oh the thumbwheel broke off my micrometer: I'll print a new thumbwheel bracket. The windshield mount on my recumbent broke, so I'll print a new one. He's printed plumbing parts, cookie cutters, centering adapters for optics, replacement bar handle clamps, you name it
Has he printed any foundry patterns yet? That's the idea that I'm hot for, but no one else seems to like or try that idea, as far as I know.
I want to print a pattern, cast it in Aluminum or maybe advanced zinc alloy (or maybe someday I'll afford the fancier safety gear to cast brass) and then machine my new "whatever" casting on the mill to get a finished part.
I can already do that with lost foam, by making a foam pattern on the cnc mill, but my god the pillar of smoke from burning styrofoam is an instant
Re: (Score:2)
I think that'd work great except that the surface texture of his patterns is rough and that's going to be hell for drawing the pattern from the sand. I've done a *lot* of green sand pattern casting. You need A: draft and B: good surface finish. (How good? The castings I do, if I put a piece of scotch tape over a lathe center hole in a turned form, the finished casting has the zigzag end where the tape dispenser cut off the tape clearly visible in the metal. Any surface imperfections will be finishing i
Re: (Score:2)
Oh hey someone's doing it: http://hackaday.com/2012/01/31/3d-print-in-wax-cast-in-metal/ [hackaday.com]
Printing powdered wax into a wax form, and doing lost wax casting.
(All the advantages of lost foam, with much less smoke. This does limit you to castable refractory, which gets expensive, but they have castable refractory that can handle platinum, so you could do steel with this.)
Re: (Score:2)
Do you have some recommendations for a combo late/mill in the $1.5k category? I was unaware they had gotten so cheap.
Re: (Score:3)
Re: (Score:2)
Re:3D printers suck (Score:4, Interesting)
If you are REALLY hardcore, you can 'almost' build a mill entirely from scratch. First you need to build a smelter, which is not very difficult, then make some parts out of wood... cast them, pour them, sand... sand... sand... then there are tricks for making parts that are totally smooth that again involve a lot of sanding. In the end you have to buy bearings, the chuck, some other odds and ends. You can make the motor, but you're really trying to do it the hard way if you do. Once you have all of that done, you have a mostly aluminum mill. Which you quickly use to make steel parts for your next mill, because it sucks. But it can be done. The point is, once you have a mill, you can make nearly anything given enough time.
Re: (Score:2)
Why not just buy a superior mill in the first place? The cost will work out to be about the same and you save a lot of time.
My guess would be - for the experience.
Re: (Score:1)
Re: (Score:3)
*I* certainly have never seen a mill/drill under $1500 that's useful for anything other than a boat anchor. A coworker got a Smithy Granite 1224 that he thought was fairly useful, but it sure wasn't $1500. (And of course that's a pure manual machine, so you still have to add at *least* $500 of electronics to get CNC -- and for that same $500 you can buy a complete 3D printer kit [hackaday.com] and be printing stuff in two hours, they claim, whereas it took me about 20 hours to convert my manual mill to CNC.)
Re: (Score:3)
All the info you need to get you started. All those 7x10, 7x12, 7x16, etc. lathes come out of the Sieg factory in China then resold as Harbor Freight, Grizzle, Jet, and many other brand names. Slightly different motors & trim levels, but the same basic machine. Great starter machines that are pretty capable. Not as robust as industrial machines, but very capable none the less.
Sieg also makes some mills and lathe mill combo units. Good little starter equipment with a robust comm
Re: (Score:3)
I've used 7x10's. I've used my Atlas -- same model as yours. I'm sticking with the Atlas, and my coworkers who have 7x10's come over and look at the finishes I can get and sigh with envy. I consider most of the 7xwhatevers to be parts kits that could be turned into a reasonable lathe if you replace all the plastic changegears with steel ones (at $40 each) and likewise the plastic gibs and don't mind that you still can't get the compound to feel solid, and spend a bunch of time scraping away at little bit
Re: (Score:2)
Re: (Score:1)
Re: (Score:3)
I presume you mean manual mill, not CNC mill. Even so, I'm not sure it exists at $1500. Smithy makes some combo machines. I'm not sure I'd want a combo machine. I'd rather have a separate mill and lathe. For mills, IF (big if) you can find a well taken care of 50 year old Bridgeport or clone thereof, that is the way to go if you have room for it. But you need to know what you are buying, because if the spindle bearings are shot you are looking at an expensive repair. That largely depends on how well
Re: (Score:2)
Well, if every feature fell on a 1/16 inch location, sure. But if features tend to fall on a 0.010 or 0.100 grid it gets annoying.
Re: (Score:2)
The problem with a mill is that you are starting out with a large block and taking off pieces until you have your shape. So you end up with a lot of wasted material (of course you could probably re-cast the scrap shavings into a new block but...). Whereas the 3d printer uses only enough material to make your shape, and it can include hollow parts inside. Some of them can even make fully assembled movable parts (like say a crescent wrench).
As for the material, it would seem that eventually one would come
3D printers suck FOR THE MOMENT... (Score:3)
I see your point, but I think your argument is a little bit like saying (circa 1990) "why would anybody shell out hundreds of dollars for an ink-jet printer, when for the same price you could get a really nice set of drafting tools? And you could choose whatever paper and ink you like, instead of producing a fuzzy mess that runs when you get a drop of water on it."
Re: (Score:2)
except when inexpensive (for the time) inkjet printers did hit the market full force in the early 90's they where a godsend, cheap color, near laser quality, fast, and most importantly didnt sound like an industrial machine raping the transmission in your car
Re:3D printers suck (NOT) (Score:2)
two comments....
I have built both a CNC router and a 3D printer. It is *much* harder to build a milling machine because of the mechanical stiffness that the machine requires.
Take a look at the printed parts used to build a reprap - the parts have very sophisticated shapes that are *impossible* to make on the typical sub-$1000 2.5D CNC machine.
Re: (Score:1)
1) You can't get something CNC for that.
2) Harbor freight (1.5K mill linked to below) doesn't sell tools, they sell tool _kits_. If you're not prepared to disassemble, align, and otherwise fix all the stuff they busted, you're screwed. I'm a half-owner of one, I know.
3) Tool pathing is still expensive / highly skilled. At your price point, you can't just turn a 3D model into a path that a mill can make.
4) Design constraints are different for the two. You can't mill internal voids.
None of which is to sa
Re: (Score:3)
Okay.. so how do you create a single-piece hollow sphere on your combo lathe/mill? It seems to me that there are absolutely things that you can do with a 3D printer that you couldn't do with a lathe/mill..
It doesn't necessarily mean that it's "better", but it is different. And as one of those "non-mechanically minded" folks you mention, it does seem simpler to me from both a hardware and software perspective.
Re: (Score:2)
I don't have a 3D printer myself, but I did get some stuff printed last week. I just sent them the model, and 2 days later I had my first ever 3D prints. Awesome.
When our CEO saw one of my prints, he was really interested. He asked if we could perhaps use the process to do faster and cheaper creation of future prototype models, and the answer is yes. I guess his little engineering hobby is just for kids though, it's not like it's made him a millionaire or anything.. oh, wait..
Re: (Score:2)
WTF are you talking about? The context of this thread is 3D printing vs CNC/lathing, and pointing out that both have their pros and cons, and can in fact complement each other quite well. Who cares if it's home or "professional"? We're talking about fundamentally different methods of creating parts.
The home kits will get better over time, just as 2D printers have. At this point in time I don't think the cost of home printers are worth it personally, hence I sent my designs elsewhere to be printed.
I don't se
Re: (Score:2)
Oh, and the hollow sphere thing was what we call an "example". Ping pong balls are made in two halves, and have a significant weakness across the seam. If you can't think of any uses for being able to print objects within other objects with no assembly required (fully assembled ball bearings being a cool example), or being able to print a single piece item with a single piece external skin and an internal lattice structure, etc, that is entirely your own lack of imagination at fault. Feel free to never util
Re: (Score:2)
Re: (Score:2)
Re:3D printers suck (Score:5, Informative)
I have a Makerbot Thing-o-Matic, and have also used smaller CNC milling machines before. There really are some crucial differences:
1) There's an actual fundamental difference in the types of objects that can be created with an additive process like the ToM uses, and a reductive process like a CNC milling machine uses. Creating lightweight, hollow objects is basically impossible with a mill but trivial with my ToM (I frequently print out multi-layer gear objects for mechanical widgets I'm building).
2) I can operate a small 3D printer in my apartment. It's roughly as loud as my laser printer, and doesn't produce any mess/dust-clouds/bad smells. I would never use a CNC mill in my living room. At best it is way louder, produces way more mess, and basically requires a shop-vac to be running the whole time.
3) Finally - having used a range of both cheap 3D printers and cheap (and not-so-cheap) CNC milling machines - I can get a low-end 3D printer like a makerbot that actually produces useful parts for under $2000. Ever CNC mill I've seen for under $3000 has been crap. If you can spend $5000 (and have at least some background in machining) you can get a pretty good small CNC milling machine and maintain it.
Re:3D printers suck (Score:5, Informative)
Well, I see you have been modded +5 insightful. Yet, you don't seem to have any experience wither either mills or 3D printers, at least you don't claim to have any.
I own one of each, and have access to other CNC mills and 3D printers.
Guess what? They both have their place. They both have their limits. Use a machine tool within it's limits, and you will be happy. Try to push beyond its maximum work envelope, and you will be frustrated. I have done a lot of good stuff with both. My clunky Makerbot Cupcake has printed a lot of robot parts and other stuff. It is quick to draw something up and bang it out. And it lives in my living room. Guess what, my mill doesn't live in my living room. So I'm very happy with the clunky state of at-home 3D printing. Do I want more resolution and a bigger work envelope? Yup. But I've still done a lot of good stuff with it.
CNC mills are great, but it also is a whole heck of a lot more work to go from a drawing to a part. And more expensive, too, by a lot. I could buy several Cupcakes for what I have invested in cutters, collets, measuring tools, vises, clamps, etc, etc., not counting the mill itself. And there is no comparison between the learning curve. You will be a 3D printing expert long before you have mastered creating G-code for CNC milling.
As to your cost comparison, there is no $1500 CNC mill worth having. I've seen the output. I've talked with owners. I've done the math and understand the work envelope. $1500 spent on a RepRap style machine can do a lot of good stuff. $1500 spent on a CNC mill.... is a sloppy, weak columned, backlash-plagued wimp with a work envelope so small you can't produce parts as big as you can on a cupcake, and you *still* haven't bought any tools. The $1500 CNC mill can work in aluminum and free-machining brass. The RepRap can work in ABS and PLA. Well designed ABS or PLA parts can be pretty strong, and can be glued up into strong large parts.
Face it, all you have said is: "Grapes are awful, they don't taste like chicken."
PS. In case anyone is wondering "Well, what *is* the cheapest CNC mill worth owning?" I would say choose between a Tormach PCNC 1100 or a Mikini 1610L. This is what Sherline owners move up to after they have figured out that the Sherline can't do what they wanted to do. This is not because Sherline is bad, or that Sherline lies in their spec sheet. It is because beginners don't understand what the Sherline spec sheet is trying to tell them, and so they don't realize it can't come close to doing what they think it will. The other thing to remember is that when you go to buy a dial indicator or a carbide cutter or some other widget, it costs you exactly the same amount of money whether you are going to use it on a Sherline or an Akira-Sieki. CNC machining is a spendy hobby, that's just life.
Re: (Score:1)
Re: (Score:2)
Acetone makes a reasonable cement for joining pieces. I suspect melting the whole works would degrade the material properties, and besides you need a mold so you might as well start with a molding process. 3D printing works by heating the feedstock until it is in its "plastic" phase and will stick to itself, allowing you to build up structures. They end up reasonably strong, but not as strong as an injection molded part where all of the melt is packed into the mold under pressure, and freezes at roughly t
3D printers Aren't (Yet) For Noobs (Score:2)
You make good points. However, neither a 3D printer nor a lathe/mill is really geared for someone who isn't prepped to put in a lot of time learning the tool. Then, there's the issue of right-tool-for-the-job. I think 3D printers are still best suited as a prototyping tool, although once tuned up they can also work well for creating plastic parts that would be a wee bit of a challenge on a mill (eg. gears). I've seen print tests with porcelain mix that put out complex shapes ready to fire... a niche I doubt
Why are these so damn expensive? (Score:2)
I know it's a cool idea to 3d print the parts but can someone please finally mass produce fully assembled units so we can have $200 3d printers already?
Assembling the machine is the easy part (Score:3, Informative)
Re: (Score:1)
I'd suggest #reprap on Freenode for being the easiest way to get information like this. In fact, you should try out Printrun and Slic3r.
Re: (Score:1)
Re: (Score:2)
Assembling Your Own 3D Printer (Score:1)
Re: (Score:3)
Here's a good start:
Mindstorms Autofabrik [youtube.com]
Print Escher stairs? (Score:1)
Give that this device basically lets you print 3D objects you can draw on the computer screen, should I expect to be able to finally have a real life Escher stairs and the such?
Re: (Score:2)
Re: (Score:2)
sure can, you just have to model it in 3D first
Re: (Score:2)
Am I missing something (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
No kidding. From the summary, I thought this would be a guide to sourcing parts. Once I started reading, I quickly started losing interest since I didn't already have the kit to assemble. Next week: "Applying toothpaste to your own toothbrush."
Oops, sorry. Car analogy: "Refueling your own hatchback."
Costs still seem to high... (Score:2)
I'll get me one, as soon as it gets cheaper to print out a WH40k army, than to buy it. Bonus points if it comes out sufficiently coloured.
At that point, I'm all set to jump on this new piracy train WOOOOP WOOOOP
Some Built it yourself links (Score:2)
http://www.kickstarter.com/projects/printrbot/printrbot-your-first-3d-printer?ref=live [kickstarter.com]
http://www.kickstarter.com/projects/relativedesign/reprap-self-replicating-3d-printer-fuel-the-moveme?ref=live [kickstarter.com]
I use my 3D printer to print itself -"SkyNet' (Score:2)