Ask Hackaday: How Do You Keep The 3D Printer From Becoming EWaste

One thing we sometimes forget in our community is that many of the tecniques and machines that we take for granted are still something close to black magic for many outsiders. Here’s a tip: leave a 3D printer running next time you take a group of visitors round a hackerspace, and watch their reaction as a Benchy slowly emerges from the moving extruder. To us it’s part of the scenery, but to them it’s impossibly futuristic and their minds are blown.

Just because something says it's a Prusa i3, doesn't mean it is a Prusa i3.
Just because something says it’s a Prusa i3, doesn’t mean it is a Prusa i3.

Nearly 15 years after the dawn of the RepRap project we have seen a huge advancement in the capabilities of affordable 3D printers, and now a relatively low three-figure sum will secure a machine from China that will churn out prints whose quality would amaze those early builders. We’ve reached the point in our community at which many people are on their third or fourth printer, and this has brought with it an unexpected side-effect. Where once a hackerspace might have had a single highly prized 3D printer, now it’s not unusual to find a pile of surplus older printers on a shelf. My hackerspaces both have several, and it’s a sight I’ve frequently seen on my travels around others. Perhaps it’s a sign of a technology maturing when it becomes ewaste, and thus it seems affordable 3D printing has matured.

Printer History Told Through The Medium Of Junk Piles

My hackerspace isn't the only one with a few spare 3D printers! Photo: Mendel Mobach
My hackerspace isn’t the only one with a few spare 3D printers! Ultimaker stash photo: Mendel Mobach

You can tell a lot about a product by how long it remains in use, for example a much higher percentage of Volkswagen Golfs have survived for three decades than have Yugo 55s. Thus a look at these piles of old printers is to look at which models turned out to be the stars and which the lemons. There are several main strands among them, each one representing a different set of failed dreams. First among them are the pioneers, a Makerbot Cupcake or perhaps a RepRap Darwin or Mendel. We even know of one space with a working Cupcake, but these aren’t machines you would use in 2019. Instead they represent the days when 3D printing was still a novelty even to us. If you think of the first machine you saw, perhaps it will be here.

Then there are the surprises, decent workhorses in their day that could still deliver a creditable result but which have simply been bypassed. It was a shock earlier in the year to see a pile of Ultimakers lying idle for instance, but perhaps I needed reminding that it was no longer 2015.

By far the most numerous though are the next group, knock-offs, proving that merely selling something as a Prusa clone does not make it a Prusa. It’s interesting though that not all clones were of low quality, I can think of more than one space still happily using years-old WanHao Duplicators, clones of the Makerbot Replicator.

Finally it’s a constant among hackerspaces that someone will have bought a delta printer and inevitably found it to be problematic. Is it a desire to be different or merely a fondness for printing vases that drives it, but if I had to guess which printer would be gathering dust it would be someone’s delta project.

What Can You Do with all This Junk?

So we’ve exposed a typical hackerspace junk pile as a potted history of failed desktop 3D printing endeavour, but as well as that it’s both a significant problem and an opportunity. A problem because while it has become a matter of ewaste the 3D printers themselves are still perceived as having value — but value that can only be realized if they do something other than sit on a junk pile. An opportunity because they still contain the basis of a CNC machine that is sitting idle.

If we can rid ourselves of the view that an old 3D printer is sacrosanct then we can make something of them, turn them into the basis of something else. We’ve all at times amassed piles of junk before realising that we’ve made a mistake, so there’s no shame in taking a fresh look at that printer pile.

The WanHao Duplicator that's still one of the workhorse printers at MK Makerspace.
The WanHao Duplicator that’s still one of the workhorse printers at MK Makerspace.

At its most basic, a redundant 3D printer is a pile of parts just waiting to be dismantled. Slides, stepper motors, extrusions, and the associated electronics. It’s questionable whether whatever was the cutting edge in extruders and control boards circa 2012 could find a use in 2019, but  the rest should find ready homes in other projects. To dismantle a printer though is perhaps to lose sight of its main event. A ready-made CNC mechanism with generous X, Y, and Z axes and a ready-to-go controller should be something that presents endless possibilities.

Attaching a mini router or an engraving laser to it is something of a done deal, but what else could you do? Make chocolate and fondant extruders with syringes and servos, and print confectionery? Attach an LED for some 3D-light painting? One-off art project or wacky tool, how have you repurposed your disused printer to keep it from gathering dust.

Whatever you think of for your surplus 3D printers, perhaps in all this there’s another story. Our community didn’t invent the 3D printer, but we’ve certainly taken it as our own and run with it. If we’ve reached the point at which they have finally matured, then in part that success belongs to us, to those hackers who have pushed forward their design. We can feel proud of that pile of surplus older printers, because they and the $200 Chinese machine that replaced them represent a 15-year success story from our community. The question is, where will they go in the next 15 years?

91 thoughts on “Ask Hackaday: How Do You Keep The 3D Printer From Becoming EWaste

  1. There’s plenty of people around who would be happy to buy a used one for ~$100… but the few sellers around seem to think that however outdated it is, it’s worth $300-ish … despite as mentioned, something now twice as capable is selling new for $200.

    1. Yea, there needs to be something special about a second-hand printer to demand more than a bit above parts value. Especially with new printers bumping along between 89$ and 139$ (on sale) for something functional. Which, lets face it, is probably the same amount of potential work to build out as something second hand that was someone else’s problem.

        1. I think that may be referring to the lowest possible prices but the analogy still stands firm. Many of these printers are easily surpassed by newer tech and yet people seem to think that they hold their value. Not true in the least.

          The average printer ages about twice as fast as an average computer in many ways mostly due to their single purpose nature. What was purchased at a “premium” price when brand new is now only worth 1/4 of the original price at very best… especially when compared to a comparatively low cost brand new system that has the latest tech and a full warranty.

          Then again, some of these people are the same ones that expect something like a Star Trek Replicator level of perfection when it is actually basically a machine tool subject to the same adjustments, maintenance, and learning curve of any other complex tool.

          It would be great if more people realized this as the components are still mostly useful and there are still a lot of things they can be used for.

  2. I limit number of ewasted 3D printers by not having one. And for that I have a few reasons:

    1. I think that the whole 3D printing thing is overhyped. I don’t like stuff that’s overhyped.
    2. Anything a 3D printer can do, can be done with CNC mill. The difference is that with mill one must plan ahead, how parts would be made and assembled to form the final product. There is also some work with choosing milling bits, drills, engravers, etc. It’s a bit more involved.
    3. 3D printers are limited in the type of materials they can use. CNC mills can do all sorts of materials, including wood (which I prefer), various plastics and metals.
    4. 3D printers are SLOW. And require constant observation in case something goes wrong.
    5. Most people use them for the same things they could make using hand tools or CNC mills.

    I’d wait maybe another 15 years for technology to really mature…

    1. I think part of the point is that CNC mills tend to be both more expensive and more involved. I’ve never seen a hackerspace with a row of CNC mills, but plenty with numerous printers. A CNC mill is an excellent tool, but only if you have access to one and can use it.

      1. Actually in my opinion and from my experience CNC mills are not that hard to use. It is a bit more involved than 3D printing, but on the other hand you can use a mill for stuff that is impossible with 3D printing. As you can use 3D printing for stuff that would be very hard and complex on mill (and would probably involve multiple different tools, molds, casts and other, fun stuff). You can get an entry-level CNC for about 500USD. Some versions even have true USB controller, which is both a good thing and disadvantage (it works only with Mach3 under Windows, so open source crowd won’t like it at all). There are cheaper machines that use aluminum extrusions, 3D printed Z-axis and spindle assembly and cheap 775 DC motor as spindle, but they are almost useless – no rigidity, especially in the assembly, and by default steppers are underpowered and overmicrostepped. Just search for CNC 3020, which has more rigid construction. I’d add to that another 100-150USD as a future cost for beefier spindle once the 300W one dies. The operating costs are just some lubrication, milling bits and stock.

        An open source alternative are MPCNC and GRBL controller. I did use GRBL and it worked fine. The machine it controller OTOH was almost useless – it required plenty of hacking…

        1. I agree that 3D printing is over hyped and CNC mills are more capable, but for the average person, I disagree that they aren’t that hard to use. Work holding alone is a huge issue, often requiring designing and milling parts just to hold the part you ultimately want to make. Planning/generating tool paths takes a lot of time and consideration and can get extremely complicated if your piece isn’t simple x/y profile shapes. Then there’s the mess of chips, the cost of the tooling – I easily have more invested in end mills than I do in both my 3D printers. There’s the constant changing of end mills and having to indicate new tools between changes to continue subsequent ops. Now if you really want to cut metal, even some plastics, you’ll want flood coolant, which requires a significantly complex enclosure, the shielding of the electronics/motors, etc.

          I have both a mill and two 3D printers (one FDM, one resin). I love the mill and use it more than the printers, but zero people I know would invest the time and resources it takes to learn/use a mill over a 3D printer. They’re just totally different worlds of commitment.

          Lastly, I think you’re off on a bit of a tangent. The article is about what to do with obsolete 3d printers, and I agree with Colin – I’LL TAKE THE STEPPERS!!! :)

          1. – agreed – I wouldn’t put CNC in a similar degree of difficulty (having done both). So many more complications. Work holding, tool sizes to account for (both cutting radius, and clearances before you’ll smash your collet or something into the work), feed rates (sure, 3d printers have them, but generally you can get away with defaults, and only tune if you want). Much less forgiving – don’t put enough thought into a feed rate, depth of cut, or work holding and you’re likely to break (not cheap) things. FDM, you can pretty much model it with a bit of thought into support and how it will print, and let it rip. I don’t generally closely monitor prints, but even if you want to, a RPi on a budget printer gives you wireless printing, and monitoring prints from your mobile device if wanted (given usb or pi cam).

          2. To be honest I cheat a little”
            1. I screw the stock outside the work area to the table. Usually that\s enough. If I need to machine part on different side, I cut a hole that’s negative of the new bottom and pressure-fit the part. Then I just zero the Z axis on the stock top.
            2. I have limited number of tools so I plan the job accordingly to make the least amount of tool changes.
            3. Every new tool in use is separate job, so I don’t bother with tool offsets and such. Just wait for job to finish, replace the tool, zero the Z axis and load a new job.
            4. For both CAD and CAM I use Fusion 360. IMHO the best “free” software for that kinds of work. It has some quirks and issues, like the whole “cloud” thing, tool paths requiring regeneration whenever I open a file I worked on before, etc. Open Source CAD software usually has terrible UI, bad UX and sometimes stupid basic idea making the software even harder to use, arcane and evil.

            As for feedrates that part is annoying, true. The data available is either for the most expensive tools (which I can’t even find here, much less afford them), or unreliable and imprecise. And everyone in USA still uses IPS instead of mm/min, which adds extra conversion for me. I hate Imperial system and it must die. Anyway I figured out some standard default values for stock that I use and it works so far.

            As for crashes, problem usually exists between computer and chair. With my first machine, which was weak, not stiff enough and generally a piece of junk, tools always broke on the material. With the one I have now tool usually is fine, the workpiece or stock or wooden block that holds it gets mangled.

            I need more end mills. Yesterday one job took over 1.5 hours, because I have one end mill that is too big and one that is too small. Fortunately the cheap ones I can get are good enough for materials I’m machining.

        2. Entry level CNC for 500$? In which world?! PocketNC is considered entry level and it is about 6000$ or more like 6000€ after VAT and shipping. For 500US you won’t even get one working axis DIY.

          1. The machine next to my computer begs to differ. I’ve got if for free but owner who gave it to me paid for it less than 700USD 4 years ago. I spent another 130USD to get a bigger spindle once I burned the original one.
            Just search on Ebay or Aliexpress for CNC 3020 and see for yourself.

            As for Pocket NC, it’s not an entry level machine. It’s more like semi-pro machine, being 5-asis type that can machine aluminum at decent speeds. It’s entry level when compared to machining centers that cost 10-50 times more than Pocket NC…

        3. I’ve got a CNC3018. It’s not underpowered or lacking in rigidity. As I discovered when I mid-calculated and accidentally cut though my metal hold-downs.
          Hold-downs are a big issue, as is computing cutting paths for anything but the simplest shapes.
          3D printers can now handle some amount of overhang I believe, allowing shapes which are very difficult without a 5-axis CNC.

    2. To be fair you don’t seem like you are the target market for 3d printers so it makes sense that you wouldn’t find them as useful as say using a cnc like you mention. Every tool has its pros and cons, there are plenty of jobs where an fdm makes a ton more sense than trying to bodge a job on a cnc.

      I do a lot of rapid prototyping and with my workflow dialed in after using a 3d printer for a good 4 years I can turn a design from concept to having a working prototype in my hand in as little as an hour or two (obviously for relatively small mechanical assemblies dependent on complexity that don’t require the lowest layer height).

      Just about all the cons you’ve mentioned for 3d printing can be attributed to user error/misuse and can just as easily affect a newbie using a cnc machine. Of course if you try and use a hammer as a screw driver it’s gonna suck, but likewise trying to use a screw driver as a hammer isn’t optimal either. In the end use whatever tools you have that are best for the job.

    3. 3D printing parts is a lot faster and easier and requires a lot less knowledge and skill than operating a mill, either manual or CNC.

      Slow is relative. Milling a complex part is also a slow process and even a CNC mill requires a lot of user intervention. A 3D printer can be started up and left alone for hours or days (if it’s a well made printer, like the ones I build) while you go do something else. You don’t have to watch a well made printer.

      I’ve built and used 3D printers, and used manual and CNC mills, and can tell you that a 3D printer is definitely faster and easier to use for most jobs. I use whichever tool is best suited for the job, but 80% of the time, the 3D printer can produce an adequate part for whatever I’m doing.

      I think that if you tried one out you’d find that it’s surprisingly easy to use and quite practical. Nope, you aren’t going to make high precision stuff or engine parts on it, but there are a lot of jobs that can be done and done well.

      1. Agree entirely on picking the best tool for each job, printing often being adequate and slow being relative – Just because most cuts on a mill take a handful of mins, for every complex part you can spend hours getting it re-positioned for the next cuts. And many extra hours in part design figuring out how to make this complex geometry in multiple parts for later assembly. With a good 3d printer do the part design (often without needed to care about its geometry at all) and then just tell the machine to print.. Come back tomorrow and its done. With CNC you still have to do the part design but because of the massive limitations a cnc mill imposes you end up having to really spend time thinking about the fabrication process so that step takes much longer.. And then you have to do all the manual setup for each cnc step.. Man hours CNC is hugely worse always (though frequently worth the time as the parts can be made in the best materials)!

        As for precision define ‘high precision’ FDM machines can match or even exceed milling in precision depending on a great many factors like depth and widths of features and value of the machine (a stupidly expensive really good mill will be stiff enough to take the cutting force without deflection but one more comparable to 3d printer pricing won’t be able to easily).. And the uv cure resins can offer stupid precision levels..

        The biggest gain to 3d printing is the crazy geometries that a cnc mill could never do in one part.
        Biggest gain to CNC is nothing to do forming the shape, but the much wider variety of materials that can be worked.

    4. That were my thoughts as well (at least a couple of them), until I received my first filament printer. I now see them as the better machines than CNC mills:

      1. Less waste material – subtractive manufacturing leaves lots of waste.
      2. Not so loud – you can operate a 3D printer without disgruntling your neighbourhood.
      3. Less dangerous to handle – a printhead might burn your fingers, a spindle can hurt it even more.
      4. Routing something in 2.5D with a CNC mitt takes time as well and you can’t let it run without constant observation.

      I should have started 3D printing even sooner than I actually did.

    5. 1. Come back in from the lawn gramps, it’s winter and cold outside.
      2. Well that’s simply not true. Most things, sure, but not when you start taking advantage of the opportunities the tech exposes. Big companies don’t have machine parks full of printers worth millions just for giggles (well, I’ve worked with companies that do, but that’s beside the point :D ).
      3. I’ll give you wood, that’s a bit tricky. But plastics, composites, “rubbers” (I bet machining a block of Shore 50A plastic is pretty tricky, especially when you want details on the sub-millimeter scale) and metallics are all available.
      4. Slow is a relative term. While you are planning, setting up, and afterwards assembling your parts, the printer will have been printing for a few days already. Sure, if we’re talking about a very simple part with little to no planning, a CNC will likely win, but once things start getting complicated that’s a different matter. If we’re talking about a run of many of the same part where the process is dialed in, the CNC will have an advantage, but when we start talking custom parts things change a bit.
      And no, a quality printer doesn’t need to be monitored any more than a CNC mill would need to be.
      5. Hobbyists, sure.

      In the end though, it’s all just tools. You use whatever tool is the best for the job, in some cases that’s a hammer, in other cases it’s a scalpel. There’s no wrong or right, just a sliding scale of applicability and strengths. Many people suffer from the “everything looks like a nail”-syndrome once they get a printer.

      1. Ad. 1. I’m 34, and what winter?! It’s 8 degrees C here! And 3D printing is overhyped. It used to be used only by big companies for prototyping products that would end up being machined or molded (depending on type of product and material). Now every company, every hobbyist and every start-up tries their best to use them as main production tools, even if it doesn’t make any sense whatsoever.
        Ad. 2. Big companies can afford big, powerful 3D printers. They can also have machining centers and other tools that are beyond hobbyists or even some semi-professionals.
        Ad. 3. Show me an entry level 3D printer that could do metals. And I wouldn’t machine semi-rigid materials – I would mill a mold for them.
        Ad. 4. From my experience the slowest operations are always the most complex ones, like 3D carving with ball end mill. Still it might be comparable or faster than 3D printing. The only big waste of time on hobby mill is the setup, like reversing the stock to mill the other side. But that could also be solved with some clever thinking. I’m working on a custom keyboard from wood – to mill the bottom side of keys I’ll mill a hole for them in piece of stock and then put them one at a time and run the finishing job.
        Ad. 5. Most of HaD readers are hobbyists.

        On that last point I agree with you. 3D printer is just a tool, useful in some contexts, but it’s not a replacement for every other tool, especially when we are talking about entry-level, hobbyist machines. Yet people try to 3D print just about anything, even though it makes no sense from the engineering (or economic) point of view.

        1. > *Now every company, every hobbyist and every start-up tries their best to use them as main production tools*

          That’s wild hyperbole. Yes, some people overdo it, especially people who’ve never actually used 3D printing and just have heard of it, but the limitations are well-known.

        2. #3 – Show me a metal-capable mill, that doesn’t require all kinds of upgrades in motors, ways, stiffeners, etc., with tooling, that costs less than 10x what an entry level 3D printer does. You can’t, just like no one can show you an entry level 3D printer that can do metals. The question is absurd.

          I have a Taig CNC mill, which is about the stiffest complete CNC mill you can get before the price increases dramatically. It cost $2600 between the mill and controllers. I spent another $1.2 to 1.5k building an enclosure with flood coolant to cut metals reliably well. And I have at least $1k in end mills and probably another $1k in fixture plates, clamps and accessories. That’s upwards of $6000+. Of course it’s more capable than a $300 3D printer. To suggests that reflects poorly on the 3D printer is ridiculous.

          1. I didn’t try aluminum yet, but someone else did. His machine is almost identical to mine, but he upgraded the spindle to 1kW water-cooled model. He can do aluminum and other soft metals at low feed rates (IIRC 120-240mm/min), 20k RPM and depth of cut no higher than 0,3mm. He uses vacuum to remove the chips instead of flood coolant. I think I might be able to do the same at even lower feed rate and DoC. So I’d rather make molds to cast the part with wax and then use lost wax method. It’s done that way in industry…

            Come to think I actually machined aluminum once. I accidentally crashed into the bed, but tool survived without damage… ;)

        3. Well, I gotta say that you have a good way to go growing up. I’ve been in machine work, hobby and professional, longer than you have been alive, and my introduction to 3d printing (FDM) was in the mid-1990’s.

          You keep dodging and weaving here. Your original claims were that 3d printing is overhyped (opinion, but I tend to agree, especially at the consumer/hobby level, and irrelevant to the article), that 3d printing can nothing that milling machine can (demonstrably false, and irrelevant), 3d printers are limited in the materials they can use (debatable on the whole, but true at the hobby price point, and irrelevant to the original article), 3d printers are slow (debatable– it depends on the material, design, and intent, and, yet again not relevant), and that most people use them for things that hand or CNC tools could be used for misdirection, not relevant, and just plain dumb, as a 3d printer IS a CNC tool.)

          Please, give it a rest. We all know that consumer grade FDM 3D printers are not in the same league as a $US100K Haas. But they can still do things the Haas can’t, and are not comparable, as the Haas is $US100K, and the consumer grade machine is $US300.

          Yes, these machines are used to make a lot of trinkets. So? You need to start somewhere. Ten years ago, the consumer grade machines were really not up to the task for parts with good, predictable mechanical properties. Now they are, within the material characteristics and skill of the user. Even the $US300 units (my low end import has made parts that are in some pretty intense manufacturing environments, for example)

          Really. Let it go. You aren’t impressing anyone.

    6. As several others have said (more politely): Ok, whatever

      To actually address your statements:
      1) Yes, overhyped, but that is not a reason to accept or reject the tech. Happy logical fallacy.
      2) No. Absolutely not the case. This is why, despite having a full machine shop, I also have and use an FDM printer.
      3) Not really. It depends on how much you are willing to pay, just like with a CNC mill. I can pay $US2K and get a desktop, small envelope 3 axis mill, or $US150K on up for a Haas machining center.
      4) Again, depends on the situation and the machines being compared. Apples to abstract algebra comparison, here.
      5) Happy logical fallacy number two. Most people use their car for the same things they could do with a bicycle, or by walking.

    7. I have a CNC machine and 3D printers. The CNC is the thing that requires constant observation. And you left out fixturing from your list, which is by far the biggest pain. I can just do a calculation to get my milling speed, doesn’t help me try to get stuff fixed and flat. Also, there are lots of practically useful shapes that my 3 axis CNC can’t do, and no good open source software for > 3 axis CAM. I’ll continue happily using my 3D printers whenever I reasonably can, thanks.

      1. I use my 3d printer for making fixtures for my cnc mill, as it happens. If you’re not hogging out a lot of stuff but just need to take a thin cut off something, like, say, a lost-PLA aluminum casting that needs a face flycut, a big 3d printed fixture with the inverse shape to the workpiece sure saves a ton of time. I did a custom thermostat housing for an antique car recently, and fixturing something with a hemispherical bottom with a hose pipe attachment sticking out at a weird angle would have been an awful task with standard clamp downs or a vise. Printing two 3d shells that clamped on either side and then fit in the vise was easy and plenty stiff enough to remove 0.1mm off the bottom to get a great surface for the water gasket.

    8. I guess, but I cast an intake for my Austin(lost pla) in one piece that no cnc would have been able to cut for about $500 printer included. So I don’t see the need for a cnc mill. Not to mention the cost in time multiple processes would take for more complex shapes.

      Also, wood? what’s that good for? You mean cnc router?

      1. Couldn’t you obtain the spare part?
        And I never wrote that 3D printers are useless or bad. They just ain’t for me, at least not for now…

        Yes, wood. And yes, a 3 axis CNC milling machine. Today I did a soap mold from piece of alder fence. Most of things I do are 2D and 2.5D cuts and carvings, some engravings too. But from time to time I do a 3D carvings…

        1. “Couldn’t you obtain the spare part?”

          Maybe [Red-Fathom] was making a stream lined induction, or adding a blower/compressor, or turbo, or fuel injectors to an engine originally carbureted…

        2. It’s for a 67 Austin fx4 they don’t make upgrades and the intake bolts to the exhaust with a “hot spot” to heat the single tinny zenith carburetor. It floods out alot.

          See I have no use for 3d carvings, but law of the hammer.

    9. I have a CNC mill. I’d like to see you mill an engine head with internal water passages. I 3d printed it in multiple pieces, placed in baked sand cores, closed it up, coated it in Plasticast, burnt out the 3d printed form, and cast the head. Then I put it on my CNC mill and cleaned up the sealing faces. (It is possible to mill some internal water passages if you’re willing to accept welding on side panels afterwards, and then dealing with welding-induced geometry changes, but it is not easy and you can’t get good coolant flow.)
      I don’t have the money for a Haas. As a result, my 3d printers are _much_ faster in making a complex shape than my CNC mill is, because I have to make multiple fixtures to hold the workpiece securely. And the total price of both 3d printers was less than the price of a replacement servo motor and amp for the mill.

      1. I was actually considering making my own model internal combustion engine, or at least a few steam engines of different constructions. But I can’t cast metals in the apartment building. Something about fire safety…

    10. 3D printers and CNC machines complement each other very nicely. A CNC is more involved and time consuming but you get a part with better finish and strength. The noise and dust also shuts down that section of the house when it use. 3D printers are easier to use and require far less operator time. Almost set and forget.

      For a more professional looking product I’ll use the CNC for the exterior and the 3D printer for internal parts. 3D printing is also great for testing the fit and look of CNC parts before you spend the time and effort to machine it. Both good machines but I don’t feel that either of them is a replacement for the other.

    11. I think it is not over hyped. Not anymore. It certainly was but the market has settled down into well defined feature sets, use cases and reliable manufacturers.

      At the beginning it was very hard to be sure the manufacturer you bought your printer from would still be in business the following year. This was important with various old-school thought trying to grab parts of the market and bringing the same s**t as the inkjet market – closed source drivers, locked in filament, etc. Those companies are mostly dead and forgotten, and I don’t think anyone will miss them. Yes I made the mistake of buying an XYZprinting printer as my entry into the hobby. It was initially great but the filament lock-in became unacceptable to me because after the hype they started to fade so I couldn’t even readily buy their stupid RFID tagged spools. Yes there were some “hacks” to change that but really they weren’t feasible, it is extremely difficult. So I tore it apart, chucked the injection moulded casing into the recycling bin and had a bunch of nice motion control bits that could be made into anything as long as I had a… uh… 3d printer.

      So I got an Wanhao i3+ and it was my workhorse for 3 years. I upgraded and upgraded it until there was barely any of the original left. And that is the ultimate thing with 3d printers – with the huge amount of open work and sharing in the community they are almost never obsolete being able to literally print their own upgrade parts.

      I started out printing stupid dogs and things of Thingiverse, eventually I was designing my own things and having a device that turned my CAD creations into a physical thing is a feeling nothing short of magic. I started designing things for friends, everything from new parts, cookie cutters, replacements for silly things. You have no idea how amazed someone is when they have broken a stupid clip on a TV remote or clothes dryer and you say “Oh that’s no problem, give it to me and I’ll just print you a new one over the weekend”. I have saved many times more than my printers value in being able to replace small plastic parts of more expensive items.

      When you have a 3D printer and are happy to help people you are never short of friends. Especially those in other hobbies like drone racing where they break small plastic parts frequently and the manufacturers charge unreasonable amounts for replacements because they would rather you buy a whole new one.

      Where it does not make sense for someone to setup a whole supply and logistics chain for a small item that is where a 3D printer has been a game changer.

      I don’t buy a case for a raspberry pi, I can have one in ~2 hours off my printer bed. I don’t have to worry about trying to find small volumes of metal shim stock (which is very difficult in many places!) or cutting my thumb off in a CNC or having a big bending jig and whatever just to make a better stand for my iPhone.

      It is absolutely untrue that a CNC can do all the same things as a 3D printer. It can not produce hollow voids. It can not resolve complex overhangs or internal details. Unless you have a $6,000+ 5 axis CNC you can not even deal with some simple features that a $200 3D printer can do in stride with far less power usage, noise, mess and space.

      1. You almost convinced me to buy a 3D printer. Unfortunately they are still a bit expensive here and expensive to operate too. Actually I think they are great if you need a single, custom part or prototype. I did that twice by ordering them from makers who run businesses off their 3D printers…

        As for things you can do with 3D printing, well it can be done with substractive machining as well as with additive. The way you do it is a bit complicated: you split your design into pieces hat can actually be machined on the machine you have, and then you assemble them into one part. And you can’t actually make overhands and internal empty spaces with FDM printer without adding plenty of supports or lattice fillings. You can them only the way I can. With FDM. that is…

        1. With the right filament, cooling, a darn good extruder (I was using a Flexion which is about the best out there) you can achieve impressive bridges and internal features. It is true you can’t do a 90 degree overhang on an edge which you could do with a CNC and the right bits.

          But I’ve also never seen a CNC make a benchy boat in one pass without tool or fixture changes ;)

          It’s all just the right tools for the job, but loads of us weirdos out here enjoy 3D printing just for the sake of 3D printing.

    12. heh I wanted an adapter bracket to shove a AM4 cooler on a AM3 board, not having the board or heat sink on hand (lunch break at work) I looked it up, first one that popped up was a 3d printed model

      when I got home I flattened the mesh printed it out to scale and made it out of 3mm acrylic as it was just 2 almost L brackets with 4 holes in each, took a whole 20 min and I even sanded the edges nice n smmof. That would have taken at least 2 hours to print according to cura

      1. None near me, AFAIK. Nearest hackerspace or similar thing is about 50km from my town…

        I ordered a 3D printed parts thrice. First part was flawed design and didn’t last very long. Before I redesigned it, I’ve got the item I tried to build from scratch. The second part was actually a new X-axis assembly for my first, terrible CNC. Original was also made of 3D print, so I thought “why not”? First print was beautiful, very smooth, I paid 25 bucks and got smoothing for free. Turned out the person who made it messed up scale in one direction so all round holes were oval. I discovered this after hammering in a linear bearing. Second attempt with different maker worked much better…

    13. Irrelevant. Just because you don’t use them doesn’t mean they aren’t going to be millions of peoples ewaste. Furthermore your comments are inaccurate and misleading. I have fadal 3016 and a $350 fdm printer cr-10s. They both have their place.

    14. I politely disagree…

      1. I will agree that the 3D printing trend was far overhyped for the average person to use. These are not “Star Trek Replicators”, they are machine tools

      2. I can point out a plethora of shapes that are not possible in any way via a CNC mill and can show supporting documentation from the industrial sources. The simplest being complex geometries within an internal cavity within a single piece part.

      3. Most *Hobbyist* 3D printers are limited in materials. Industrial ones are not, many are already printing many metals including stainless steel as well as ceramics and others NOT possible through CNC mills.

      4. CNC mills (and lathes) are SLOW too, requiring vast attention to detail in planning out the sequence of steps required for a given part as well as tooling changes. there is also the issues of pre-setup and post production operations as well. Most of them also require attention in case “something goes wrong” because some failures can be far more catastrophic and destructive.

      5. Interesting that you cite “most people” when engineers have been using 3D printing for rapid prototyping for decades… mainly due to the aforementioned time involved between iterations for standard machining practices and difficulties thereof.

      Considering that the tech has actually been around for several decades, much of the tech is already fully mature… the difference involved is bringing it down in price from the $10,000 to $50,000 industrial grade machines into the hobbyist market… THAT is where it is still fairly new… THAT is where the new markets are opening up… and the very capabilities it offers are the reason why it is growing. It is still a pricey endeavor to get a decent quality manual mill or lathe, not even talking about full CNC variants, plus the time and learning required to operate the equipment.

      Finally, this is yet another machine tool to add to the arsenal of existing equipment… it has advantages and disadvantages just like a mill or lathe (CNC or otherwise) Each one has a niche that it was created to fill.

    15. Amen. I have often wondered why I might need a 3D printer. Either I’m not that creative or I can make a part some other way. I’m more “electrical” than “mechanical” anyway. A mini-mill, a mini-lathe, and a few other tools do the trick if I need something substantial.

  3. We have a few old, dead printers at the makerspace- they are available to members to take home and fix or to take the parts to build a new printer, but no one is taking them. They still prefer to just buy a cheap Chinese printer. The printers at the makerspace are seeing less use as more members buy their own machines to use at home.

    If you have an old controller, a couple steppers, and a few pulleys you can always make a sand table. Only two motors and two endstops, so wiring and configuration are no-brainers, and the simplest of 3D printer controllers are easily up to the task of drawing the patterns in sand. You don’t have to write software to generate patterns- it’s already done and ready to use. Sandify generates gorgeous geometric patterns and if you’re a RPi fan, don’t miss Mike Dubno’s advanced sand table software.

    https://github.com/dubnom/sandtable
    https://sandify.org/

    1. We’re still seeing pretty heavy use on the 3D printers at the hackerspace. I believe it’s because, in this case, the maintenance on the printers can get handled by people (a few of us, anyway) who know what each of them need and everyone else can “just use” them.

      Most of the folks have absolutely no interest in maintaining machinery. Can’t say as how I blame them; it can be a right pain.

  4. We have a few dead printers at the makerspace and I can’t give them away. No one wants to fix an old printer and very few are interested in designing or building their own using the parts from the old ones. The working printers at the makerspace are seeing less use as more people are buying their own machines.

    If you have a couple steppers, a few pulleys, a controller board, and a power supply you can easily build a sand table. Software for generating patterns is readily available (Sandify and Mike Dubno’s excellent RPi software). Since there are only two motors and two endstops (no fan, thermistors, or heaters), wiring and configuration is super easy, and even a basic controller can handle drawing two dimension patterns from gcode.

    1. You have some pretty lame assed hackers if no one gets excited over the steppers and drives etc. I can off the top think of a few things that would be fun to do with an old 3D printer, and that is assuming you don’t want to use it as is.

  5. Isn’t this balanced against all the dot matrix printers that were converted, or unsuccessfully converted, to 3D printers (and xy plotters)?

    I can remember when Don Lancaster went on about “Santa Claus machines”, mind of vague at the beginning but clearly now the idea of 3D printing. I think there wereeven some construction articles in Radio Electronics or the like with rudimentary printers made from dot-matrix printers.

    Everything starts out expensive, complicated, and heavy. Demand causes streamlined construction, and better specs, which brings in more buyers which in turn streamlines further. Few buy in at the beginning, they come in when prices have dropped a lot, and things have become simpler, andmaybe different.

    A friend bought a VCR in Nov 1980, still expensive and big. But it never changed much except for price and size.

    Few bought the Altair 8800,tnough demand was enough for the rest to follow. Each iteration brought in more buyers, but the biggest wave came way later. Lots of people use cellphones as their computers, and their “needs” are quite difference. They in effect waited till what they wanted finally arrived

    1. Yah, if I had them sitting around, they’d get all kinds of things bolted, ziptied, or even *gasp* duct taped on. Like pens, lasers, dremels, “earwax” scopes, spotwelders, Heath Robinson/Rube Goldberg arc deposition rigs…

      1. Also while they still shoved plastic I’d probably have a few days of “massively parallel” production of some generally useful hardware parts, brackets, knobs, standoffs, clips, fill up a few small bins.

  6. I’d just like to say I’m relatively new to using a 3d printer I have and can run a mill and a lathe. But that being said those machines are awesome and imagination is your only limitation. As is with a 3d printer but printing your own 55inch wingspan rc plane on a 3d printer is so bad ass it’s not even comparable in my thoughts. My take on it is just keep pushing the limits at one time the only machine that could recreate itself was a combination drill mill lathe with products like peel I think the printer could pretty much do the same thing and that’s something to say in my mind.

  7. I own a commercial pre built 3-D printer that I am soon going to submit as a Hackaday project as it has become too expensive to maintain with the manufacturers parts. I originally payed AU$1500 2016, with the aim of it being an educational experience. It has done that extremely well, and worth my original investment to me.

    However I had to choose between continuing to buy expensive parts and servicing from overseas, or buy a cheap Chinese knockoff for home. I had already recently built an Anet-A8 for our Men’s Shed (a Makers Space for old buffers mainly) and was very pleased with what I had learnt from building and running it. It is now an AM8!

    So in the end, to tackle e-waste and extract maximum learning from my early expensive machine I decided to strip the electronics out, with its proprietary software, associated harware etc and I have been rebuilding it with eBay bits, a RAMPS 1.4, Marlin 2, OctoPrint and Cura, a classic combination to break free from the original company (that deserted us in Australia). It is now 100% maintainable with very affordable eBay parts. I have learnt an amazing amount from this ground up effort.

    I am hoping it will soon make the grade as a Hacker day article to inspire others to up cycle financially unviable proprietary 3D printers with new electronics and open source software where the underlying mechanical technology is really quite good. It has been a great hacking experience!!

    See you soon. RobW

  8. Why dis the deltas? I bought (kickstarted) a DeltaMaker and at a time when everyone else was complaining about all the ways their 3D printers didn’t work, it just fired up and produced usable prints right out of the box. I bought a delta on purpose because it didn’t suffer from the one problem every other design at the time did which was over-constrained axes causing binding or slop. That and it was a very stiff machine. While they took a long time to come through, when they did they sent a very polished product.

    1. That was my experience too. The makerspace had one donated by a member and after many years of use it only just had the PSU fail (after four years of up-time). I think the only maintenance on it was the fan catching a screwdriver at one point.

    2. Same here. The only printers at our space that WORK are the deltas!

      We have, like everyone else, a shelf littered with cartesian carcasses. And there simply aren’t enough other motion projects to use all the guts we could harvest from them.

  9. Haven’t been impressed with the stuff done by hobby 3D printers which is usually some cheap plastic box,panel or some flimsy widget shown here on Hackaday. Nothing that couldn’t be done by hand or a lathe/milling machine.

    It’s not to say 3D printing is hype, I’ve seen the stuff done by corporations in the medical field that is very impressive. But they don’t use a dodgy $200.00 hobby printer or junk found in their parents basement to achieve their results.

    In regards to those cast off 3D printers, who in gods name wants one of those horrific time sinks? There is a reason they were dumped by their fanbois. They didn’t work, tended to catch fire, etc. I’ve read enough articles here on Hackaday to keep my distance from this fad.

    Maybe in another decade as the printers get better they’ll be useful. Even then I’d stay away from the cheap ones. There’s a reason why they are cheap and that means cut corners. It boils down to the same reason there are more Volkswagen Golfs on the road than Hugo’s – quality and design.

    1. Yesterday I printed a heddle hook for my wife’s new loom, and a knob for my neighbor’s motorized wheelchair joystick. I could do both of those in my metal shop. However, my wife tried out the heddle hook and said “oh wait I need the hook deeper so the yarn doesn’t come out” and the neighbor said “that’s the size of the old knob but my hands don’t grip as well as when I was younger so could you make it larger?” and half an hour later, in each case, I had a revised one that they liked. So then I milled a hook out of stainless steel and cut a knob out of aluminum on the lathe and they were both the right size. PLA is waaaaay cheaper than tooling for cutting stainless steel, and my cnc 3d printer takes waaaaaay less time than my manual lathe to produce prototypes. And the girl we were babysitting over the weekend was sure thrilled to have a 300mm high Yoda 3d printed to take home, and clearly thought it was the coolest thing she’d seen. I don’t know how long it’d take to cut a 300 mm cube of aluminum on my cnc mill because there’s no way I would ever do that for a toy for a kid. Utility, amusement, and satisfaction.
      As the owner of a legitimately nice machine shop, I seriously don’t get the widespread criticism of 3d printers. I average 2 hours of print time a day, because there are so many fixtures, enclosures, prototypes, and jigs I need in a hurry, and 90% of them don’t need to be metal.

      1. I also have a manual benchtop milling machine I converted to CNC, and a delta 3d printer that I built. In general they’re pretty complementary tools, and suggesting that one is better or worse is kind of missing the point, as they’re different tools.

        Like, which is a better cutting tool, a chisel or a saw?

        I think one valid criticism is not the machine itself, but of a lot of the users. It’s pretty common for over-enthusiastic new users to try printing everything, including parts that would be better cut out of available materials. Look at how many “drones” are on thingiverse that are clones of fibreglass/carbon designs, using similar dimensions, etc.

        In general if you’re printing sheets of material, or uniform tubing, you’re doing it wrong. Cut your tubes with a hacksaw, and then print brackets to attach it all together and print the odd shaped pieces that aren’t otherwise available.

        Also, trying to print big things in many pieces and gluing them together has always seemed kinda silly to me.

  10. I still have my 2014 Mixshop Kossel(ish) delta kit printer. It was good enough to print its own upgrades. I’ve only gone through about 5kg of filament, but that’s 83 different objects. Mostly utility stuff…a bunch of project enclosures, some soap dishes (does any commercial soap dish actually fit with any sink anyone onws??), just a bunch gadgets and replacement parts. About half of them turned out alright and are still in use. I’d absolutely say it’s been a great trip…PLA is really a crap material and FDM is a giant hack, but these things I make are *useful*. And mostly they’re things I couldn’t possibly get as a regular consumer…a lot of them could have been milled out of wood or steel but my collection of woodworking tools and skills is lacking.

    Anyways, I can’t imagine throwing that printer away. Not to say it hasn’t produced some waste, though mostly it’s plastic parts I’ve thrown out so far. But I’ve already got a replacement hotend awaiting installation, and I kind of hate RAMPS so I’ve been lusting after a smoothieboard. But the stepper motors, the little bits of hardware, the long extruded aluminum pieces, I wouldn’t be surprised if I’m still using them in a decade.

    On the other hand, I love a commodity that just works, so I could see myself seduced by a $300 new printer someday. It’s just that my printer is a known quantity so far, and that’s really where the value is. I’d have to learn anything new!

    1. “On the other hand, I love a commodity that just works, so I could see myself seduced by a $300 new printer someday. It’s just that my printer is a known quantity so far, and that’s really where the value is. I’d have to learn anything new!”

      Better the devel* [sic] you know, than the devel you don’t!

      *development

      B^)

    2. Yea, there really isn’t any new hotness. Commodity printers just spoot hot plastic through a tip.. so unless it’s something that moves that plastic in a fashion that’ll make *your* work better you’ve probably got the best setup possible!

      And in my experience maintaining the things is the biggest issue with the printers. So a printer you’re comfortable working on is far better than one you might not be.

    3. Pretty sure a soap dish isn’t supposed to be “part of the sink”. There intended to sit on the counter for those without sinks with the depression to hold a bar of soap or don’t want to clean the mess from a bar of soap from that depression.

  11. “Old” 3D printers? What? I’ve still got my original 3D printer. It’s been “Ship of Theseus”ed to the point where there’s very little of the original printer left, but its parts are kept around as spares or reused for other projects. Why would you get rid of a 3D printer (or even deem it obsolete to the point where throwing it out is even considered as an option)? You have a 3D printer! Upgrade the rigidity. Upgrade the control system. Change to zero-backlash drive screws. Add another head. Experiment! You’ve got this amazing tool. For crying out Pete’s sakes, USE IT!

  12. Oh no. Author should definitely put them old cranky plywood case printers out on the porch and put an advert in craigs or upcycle to have them hauled away and ‘properly’ recycled. Clearly unuseable trash. Dont stress the ol back too much and save a few steps to garbage can. That and save myself a few minutes washing garbage out of it.

  13. Easy: Buy a printer that’s so expensive you don’t dare replacing it. (Yeah, this is going to be a rant)

    We have an Ultimaker 3 that used to be a wonderful machine. Low maintenance, always worked. Then again, for close to €6000, you’d expect that, wouldn’t you?

    Sadly, that amount of money buys you little over a year of fun. Then, even after oiling/greasing everything, replacing the build plate, replacing the sensor board and calibrating it twice, it’ll still fail auto-leveling about 1/3 of the time. And Ultimaker tells you to just try it again. It’ll work the 2nd or 3rd time, yes. But troubleshooting is not what we bought this machine for. Annoying, but no real issue yet.

    Then, after a few months, it’ll stop updating. It will simply refuse to update its firmware from USB anymore. So you contact them again and get a whole procedure book for exhuming the main board and reflashing it. Fun times, now it doesn’t update AND the screen decided to use hashtags instead of some random characters.

    Whoops, the non-updating thing was a bug from Ultimaker. It won’t update over USB, but Internet updates work fine. The whole flashing procedure? Nah, never would have helped anyway. The screen? Yeah, you did that yourself by flashing the machine. So now we’re using a laptop to share our company wifi over ethernet to get that darn thing updated until they finally fix their bugs introduced with the “cloud” features we never wanted nor would ever use. (Fun fact: they actually removed our ability to monitor the temperatures in Cura to introduce it. Now we’re stuck with some slow website served from the printer, that doesn’t tell you anything. )

    And wouldn’t you know it. A month later it starts to randomly stop mid-print because it’s unable to communicate with the hot-ends. I2C connection seems messed up somewhere, but even after two new print cores and a new hot-end assembly (only way to buy the replacement cable), so €500 ex VAT worth of “consumables” not covered in warranty later, it still craps out about once a day. And since, by now, we’ve had it for two years, it’s our problem.

    The 3d printer we bought for less than the replacement “consumables” for the Ultimaker cost us, is doing the bigger prints now, since it can manage to keep its act together for more than 4 hours. For the small prints, we still use the Ultimaker, since it’s too expensive to throw away.

    But trust me, the day that thing finally gives up the ghost, I’ll buy it a nice new axe or bat to get intimate with, until there’s no e-waste left.

  14. Most of the comments seem distracted by the idea that a printer will always be a printer – you have a pile of stepper motors, controllers, power supplies, drive gears/screws. Take it all apart, put it in a pile and challenge yourself to build something that *isn’t* a printer with it (or give it to the local schools’ robotics team).

  15. Certainly in the UK, but I expect it is the same the world over, youth groups are taking massive steps in terms of teaching STEM within their organisations. In my experience (as in instructor for the Sea Cadet Corps) I would welcome any donation of a 3d printer (although my unit is lucky) to show our engineering classes how the technology works. Just to clarify, I’m not asking for anything myself, however there are people out there who would still be very thankful for it.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.