3D Printering: The Things Printers Don’t Do

3D printers are amazing things, but if one judges solely by the successes that get showcased online, it can look as through anything at all is possible. Yet in many ways, 3D printers are actually quite limited. Because success looks easy and no one showcases failure, people can end up with lopsided ideas of what is realistic. This isn’t surprising; behind every shining 3D print that pushes the boundaries of the technology, there are misprints and test pieces piled just out of sight.

If you have ever considered getting into 3D printing, or are wondering what kinds of expectations are realistic, read on because I am going to explain where objects come from, and how to recognize whether something is a good (or bad) fit for 3D printing. The important thing to understand is that printers have limitations, and to get a working idea of what those limitations are. The result will be a better understanding of what they can do, and what problems they can reliably solve.

3D Printers Have Limits

I recently had a talk with someone who wanted to know if a 3D printer could help with a problem they had. As I listened to them describe their needs, I realized I had in a way heard it all before many times.

My colleague actually had a fairly good idea of what printers could do, in theory. But they had very little grasp of what printers did not do, and that disconnect left them a bit adrift when it came to practical applications. To help address this gap, here are some tips that can give anyone a working understanding of the things 3D printers do not do well.

They Do Not Create Objects Effortlessly

Repairing household items is a common use case, but 3D printers do not work like photocopiers for objects, nor do they magic up replacements for missing or broken things. There is currently no practical way to take a few pictures of a broken part and have someone print a new one, nor is there a quick and easy way to make copies of existing objects.

Ideally, if someone required a replacement part to repair a household item, the process would begin with looking the part up online by manufacturer and model number. Then a user would download a 3D model of the missing or broken part, and print a replacement at the press of a button. We’re not there yet. There are plenty of 3D models available online, but we are far from having libraries of user-serviceable parts for manufactured products readily available for download and printing.

3D printers can only create objects from 3D models, and 3D models get made by someone using a CAD program. Creating a 3D model needs to happen first, because without a model a 3D printer is useless.

Making a Thing Means CAD Work

If an object doesn’t already exist as a 3D model, one must be created. Fortunately, the internet is already home to a staggering number of models for useful gadgets, tools, and knickknacks, ready to be downloaded from places like Thingiverse, PrusaPrinters, MyMiniFactory, and others. These models already exist and are — for the most part — ready to be 3D printed.

However, if one needs an object to interface with something else (for example, a replacement part for an appliance) then a 3D model for that object most likely does not already exist. It will need to be carefully designed from scratch, and reverse-engineering of a mechanical design is a process that will involve a lot of careful measuring and testing in addition to the CAD work. Access to both the thing being fixed, as well as the broken part being replaced, will probably be needed. The job could be anywhere between an afternoon’s work, to a multi-day effort.

3D scanning techniques, like photogrammetry, can be useful, but scanning is only a tool to aid other design work; it doesn’t yet remove the need to choose a CAD package and start designing. A good example of how 3D scanning can assist the design process is showcased nicely by this project to 3D print a custom control panel to fit onto a complex shape.

Some Things Do Not 3D Print Well

To print reliably, a 3D model must be designed with a 3D printer’s strengths and weaknesses in mind. One must always play to a tool’s strengths, and 3D printers are no different. Just as a table saw is the wrong tool for cutting curves, so too are some shapes and part geometries not easily 3D printed.

The two types of 3D printing that are most accessible to hobbyists are filament-based (FDM) and resin-based (SLA). Both work by building an object layer by layer, starting from a flat build platform, with each new layer being laid upon the foundation of the one before it. Because of this, some things print more easily and reliably than others.

How can one know whether an object will be troublesome to 3D print without having a lot of experience? Below is a simple checklist of potentially troublesome features. The more items in this list an object matches, the more likely the object will have challenges.

  • Does the object lack a flat surface to use as a base, or does it have a very small base relative to the rest of the model? Such models are often more complex to print than those with stable, flat bases.
  • Is the object very large, or very small? Object size can be an issue, and may depend on the printer and material type.
  • Does the model have thin walls or fine details? Thin walls are often weak points.
  • Does the model depend on tight tolerances and exact dimensions? If so, it may require experimentation to get right.
  • Are there protruding features that are not well connected to the rest of the model? The more parts stick out, the more challenging it will be to print.

For people who work with their hands, here is an evaluation method that is as intuitive as it is simple: Would the object be easy to build out of wet sand, as if one were building a sandcastle? If so, then it will probably 3D print just fine.

3D Printers Are Great, So Long As You Play By Their Rules

3D printers do not run flawlessly every time, nor do they operate in a foolproof manner. Operating and maintaining a 3D printer is not difficult, but it is a skill acquired through experience. It is entirely possible to damage a printer by running it carelessly. Ideally, one would simply press a button then sip a margarita until the machine spits out a perfectly-finished part. Unfortunately, this isn’t true of 3D printers any more than it is for any other power tool.

To be clear, 3D printing is one of the best things to have happened to hobbyists over the last decade, and the benefits are not limited to those who design objects from scratch. For example, printing miniatures for tabletop gaming is a niche that has probably single-handedly propelled hobbyist SLA printing to where it is today. As a result, hackers around the world have reaped the benefits, making it easier than ever to add an SLA printer to the workbench.

Successes are great, but knowing what 3D printers are not good at is also important. With a better idea of what printers do poorly, a thoughtful hacker is not only in a much better position to decide whether buying a 3D printer is a good idea, but will also have a better idea of how many beers that friendly printing favor might be worth.

52 thoughts on “3D Printering: The Things Printers Don’t Do

    1. I would say because it’s more of an art than a science, and there’s a lot involved in a ‘simple’ 3D print that goes much more above the contents of a page. If you press print on a laser printer, chances are it will actually print and give you a result. Only in extreme circumstances might it catch fire. You may have to do some zooming, or perhaps select portrait or landscape, but you will get a result. Do the same with a 3D printer and you have several outcomes which include anything from a plate of spaghetti, a successful print, or even fire… similar to how someone might interact with a CNC machine…. in fact since 3D printers run gcode, let’s just go there, 3D printing is not “machining” as it’s an additive process, however 3D printing is not just simply printing either….. Thus… 3D Printering, as you really have to ‘futz’ with things to get the result you would like.

      1. i woke up to a plate of spaghetti this morning. usualy wakeing up to food is great but not so much when the spaghetti is from a printer.

        at least it didnt blob all over the tip so theres that i guess.

  1. As a mechanical engineer, most of the times, there is no parts that require a single manufacturing process. 3D printing is not usually the last stage and require combining parts, bolts, inserts. Similarly, 3D printing can be a previous stage to glass or carbon fibre layout, by printing the moulds. This is complete natural, the same way that one would not expect lathes to manufacture cubes. This does not mean lathes are bad.

    1. A better analogy would be “the same way one would not expect lathes to manufacture Klein bottles”.

      I actually had a hard time coming up with something solid they can’t make, so I picked a continuous surface made by glassblowing or 3d printing.

      Lathes make cubes quite well actually, look up a turner’s cube. They make polyhedra as well with a small jig.

      1. As a mathematician I’ve had to reply here: NO! for a Klein bottle you would actually need a 4D Printer, and for that there are some key mechanical components still missing. How the rails and bearings of that machine would look like?

    2. Mechanical Engineers… if you only had a clue how backwards you do things for Machinists to fix. Your blueprints come in looking like optical illusions because you don’t bother checking your own work.

      Keep your work on the positive plane quadrant and don’t try to be clever with mixing polar and cartesian data. We still know a dremel head attached is 7mm but uses a 40 threads per inch pitch rate.

      Your first project should be to design and install
      the water faucets you use daily in your home.

      If it is easy to fix and doesn’t leak after 15 years of daily usage, and only needs the silicone grommet replaced then you are allowed to call yourself a mechanical engineer.

      But if it does leak? Thanks for making the world a worse place.

  2. Indeed, they are very good in solving Problems, NOT REDUCE them. I’ve introduced them 7 years ago in our research-lab and created several RepRap types. They are a perfect example on how every new technology intended to make things easier are priced in eventually and finally taking for granted. And at that point they start to make you work even more again. What started as watching hours and hours and be amazed by the machine building thing won’t build on the workbench ended with the fear each morning how well that box worked in the night-shift while you asleep.

  3. At the makerspace, a lot of 3D printing noobs are under the impression that they can use a 3D scanner to copy almost anything and then print it. They look very disappointed when I tell them that affordable 3D scanners suck and they’ll have to learn CAD if they really want to put a 3D printer to good use. Some make the effort, others go away, never to be seen again.

    1. I see the same with metalworking training.

      There are a lot of people with deep dreams and shallow intellectual curiosity/intelligence. They always have a plan for a youtube channel.

      I recently met a guy who wanted to learn to read blueprints so he could make some kind of side business making them for someone somehow, he was never very clear on it, and stuck around for all of 20 minutes training, and left, and told me he gave up on the idea later.

      People like this are called time vampires.

      1. Aww, that’s a very negative way to look at people who don’t know who they are yet. I’d say they’re dreamers looking for the right dream. At least he didn’t ask you to teach him for 5 years before giving up. He surely doesn’t thrive on wasting others’ time. If he did, he’d hide it better.

        1. I actually love teaching people technical skills.
          I joined my local hackerspace to volunteer my time like this, because I love helping people make things.

          I just spent 3 hours training 3 total beginners the basics of cnc router work, software too. I deeply love seeing their faces when they finally get to make their dreams.

          I didn’t mind actually teaching him at all. Its when people ask for my time, showing interest, but make no effort to understand the depth of what they’re getting into- yet already have elaborate plans and a business model based around the idea that they’ll be able to master it in 10 minutes. For some reason, they always have a youtube channel prepared to monetize too.

          I run into these types a lot in hackerspaces, and life in general.

          Very few people make any real non-flaky, serious attempt to learn technical skills beyond a 5 minute googling session on the toilet anymore. That is why I love the crowd here- people actually apply their effort here!

      1. Yup. Looking at support groups for 3D printers, I see lots of posts by folks who’ve broken a part of their printer, and think that they are out of commission until they can obtain or otherwise print a replacement part. It’s like they didn’t realize that (the right) glue can solve a host of problems like this. In fact, using a proper repair technique can often given you a better part than just using a replacement (such as when you glue on a reinforcing brace).

      2. Yeah, there’s a tendency among 3D printing noobs to think everything should be 3D printed. I recall seeing some fittings for face masks to make them seal tighter around your mouth. The process of making one involved using photogrammetry to “scan” the face, the generating a model from the scan, then printing a thin plastic frame with a ton of support material. The whole process probably takes hours and the printed frame is lucky to survive breaking off the support material. Someone wanted me to start making them for people at work (I’m a dentist) and I suggested that bending a piece of 12 gauge wire would be faster, easier, and adjustable to fit anyone.

        I use metal parts to build my printers. I have access to a milling machine but instead of designing parts that require a lot of fancy milling, I just cut pieces of rectangular aluminum tubing and use the mill to accurately drill holes and to finish the cuts. It work well and I never have to worry about parts melting or breaking.

        3D printers have their place, but it isn’t every place.

  4. I think the masses want 3d printing to be like regular printing, as example, you could draw out a picture on any particular program, use colors or not, hit something like print screen, or use whatever program, and have the system ‘just print’ whatever it is you drew onscreen. A glossing over for sure, but in order for 3d printing to hit the masses, where the masses don’t necessarily want to learn CAD or even print something not unique to themselves, this is the level of simplicity required to do so, imo. Remove some obstacles in current 3d printing to bring it more in line with the simplicity regular printing offers, then perhaps it will become a fixture in many households, maybe even placed next to the regular printer. It seems much easier for people at large to diagnose regular printer issues, like sizing issues, setting issues and such, to where whatever artwork they drew and the final result doesn’t meet expectations. Make a few small changes until the desired results prints out. 3d printing seems to have many more issues that can arise that require much more depth of problem solving skills that the average person is willing to put into a ‘quick print’. When these issues encompass hours, or even a day or two, the effort might just outweigh the benefits of whatever one wants to print. What I mean by ‘quick print’ is something on a whim, for fun or maybe just for the sake of taking an image from the mind and bringing it into reality, fun stuff, not necessarily a specific part to fix something, or a new invention and the like.

    1. Personally, I think the concept of 3D printers as house hold items as a fairly odd thing.

      I don’t think we will ever see 3D printers be as common as inkjet/laser printers currently are. (And printers in general seem to be fading out of society, at least where I live…)

      The main issues of 3D printing is generally that it will largely always require a fair bit of trouble shooting. Not just for the model, but also the machine itself. I don’t know how many times I have been asked by friends to fix their paper jams, but it happens, so how much more issues can people get into with a more complicated machine.

      3D printers are a specialty tool to say the least.
      It is on par with a lathe, or milling machine, or a laser cutter, or an oscilloscope.

      A 3D printer isn’t a screwdriver, nor a drill.

      Most people don’t have all that many tools, let alone more niche ones.

      Though, in regards to CAD software, there is sure a lot of improvements that can be made in that area.

      1. I like how the discussion about 3D Printing always tend towards the comparison to paper printer. And then I sawan obvious rule to apply when it comes to the amount of issues depend on complexity of the printing object: “square-cubic rule” – if you double the object complexity, on paper it squares, in FDM there are about 8 times more issues ;)

        1. With increased complexity comes more things to take into consideration.

          Together with a lack of patience in most people, it makes for a rather nice recipe for lackluster results.

          And the thing holding 3D printers away from the mass market is currently not price, considering that one can actually get a rather competent printer for under a 1000 dollars these days. Even SLA printers that used to be “expensive” are cheap these days…

          But that lack of a high price doesn’t make them any nicer to use.
          It doesn’t make 3D modeling any better.
          And trouble shooting the model and/or the machine isn’t easier these days then it were 5 years ago.

          Though, for people who has patience and see “valid” use in it. Then it can be a well worth thing fiddling with.

          And yes, with more dimensions comes more potential issues.
          Both in the thing one wants to print, and the machine used to print it.

  5. If you want to create objects go to tinkercad.com, the possibilities are endless but keep in mind the limits listed on this article. If you need a quick print and don’t feel like designing then go to thingiverse.com and search what you need.

    1. I agree. I still use OpenSCAD for more complicated/precise objects, but Tinkercad is perfect for quickly designing something.
      It’s the MS Paint to SCAD’s Photoshop, sometimes you just need a simple tool for a simple job.

  6. The wet sand metaphor is the perfect way to describe why some designs won’t print (without supports). Dual extruders with one dedicated to water soluble support filaments add capability for some complex designs at a cost.

  7. Had somebody ask me if I could FFF-print some suspension bushings a little while ago. I told them that was probably out of scope, perhaps fatally so. Some people have a lot of faith in plastic.

    1. I once saw a Triumph motorcycle (maybe out of the 1960s) that had its transmission case welded back together.
      That case must have shattered into a dozen or more pieces before it was welded back together.
      I don’t know if any of the pieces were larger than 4 inches across!

  8. Quick Tip: 3D Builder built into Windows 10 is shockingly useful for simple designs. I tried learning Fusion 360, but the learning curve was just too steep. OpenSCAD and 3D Builder let me still make, modify, and print most everything I’ve needed.

    1. I agree about Fusion 360. It’s unnecessarily complicated. I’ve managed to trick it into outputting what I wanted a few times, but it takes way more effort than I feel it should. I also prefer OpenSCAD. Fusion 360 could take a page from their book and make the dimensions much easier to manipulate. If I click on an object, there should be a way to change the length of any measurement quite simply. If I make a mistake and the object deforms, that’s on me. I feel like people who build this software don’t focus on usability as much as function bloat. UI will always trump a new trick.

  9. I measured and modeled and printed a plastic screw bushing for the dash pad on a 1982 GMC pickup in less than an hour. A new nylon one was only like 35 cents, plus several dollars and days shipping. I needed the thing NOW. Then I uploaded the STL to Thingiverse.

    Another quickie job. A replacement for the yarn break sensor bracket of Brother electronic knitting machines. Brother never sold that often broken bracket as a separate part, it only came with a metal ‘mast’ with various other parts.

    I received a mast with broken original bracket and yarn sensor around 11 AM. Before 4:30 PM I had the part modeled, printed, and shipped back with the mast and other parts. That included two test prints of the sections that interfaced with the metal bracket on the mast and the yarn sensor’s clip before I printed the complete bracket. No adjustments to the full 3D model were required after the tests.

    Another project was some knobs for a “Full Classic” car that was being restored to debut at the Hersey, PA Concours de’Elegance show. I had to work from deteriorated originals to get close to original shapes. Designed 3D printed shells with internal features to hang onto urethane resin filling. Also did fixtures to hold the shanks of drill bits and threaded end of a bolt in the resin to make mounting holes. To cap off the job I printed formed blocks to hold the knobs in a vise for drilling the holes larger as needed. I included the test prints as extras so the restoration shop could experiment on finishing and drilling and mounting methods before taking on the set meant for the car. About 2.5 days spent on it for $500 = my 3D printer was then free plus paid me a nice profit.

    What do I use? Caligari truSpace 6.6 and a pair of Harbor Freight digital calipers, and experience using trueSpace since it was first ported from Amiga to Windows. Now how to get Microsoft to open source trueSpace 7.x?

    MS bought Caligari to use trueSpace as their response to Google’s Sketchup, with the intent of Virtual Earth users using trueSpace to populate it with 3D models of buildings, as Google Earth users were to do with Sketchup.

    Then not too long after, Microsoft shut down Terraserver and abandoned trueSpace.

    1. Oh wow!! I used to use trueSpace! I had forgotten the name of it. I’m not sure I’d go back to it now that I have OpenSCAD and do more part modeling than artsy stuff, but that brings back memories. It was quite simple to use if I remember correctly. This is closer to how 3D modeling software should be made user friendly.

  10. Currently 2 of my 3 printers are down. Neither is down for the count but they’re down. The FDM has a bad nozzle clog and I need to take it apart. The SLA stopped printing 3/4 of the way through a job and I need to test what happened.

    Neither is major but they mean getting my hands in there.

    They’re a lot of fun (when they’re working) but they also require the ability and willingness to tinker.

Leave a Reply

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