Should You Buy a 3D Printer?

Back in 2016 I bought a 3D printer. It was mostly on the spur of the moment although I did have some concrete ideas for a few things I could repair with 3D printed parts. Now almost 5 years later, I haven't regretted it at all and I have created and repaired many things using this printer, most of which are my own designs. Not too surprising because if you browse around on this website, it is obvious that I am a tinkerer and having this new means to produce parts with high precision allows to tinker in ways I never could before. If you fall in the same category as I, then you can stop reading and just buy that 3D printer, you won't regret it. But what if it is not that clear-cut?

This article only focuses on the type of printers that are affordable for the typical consumer or small businesses. If you plan on starting a business with a fancy metal SLS printer, most of the advice on this page is probably of little use.

Possible Motivations

Roughly spoken these are the main possible motivations to spontaneously buy a 3D printer:

1. You are a tinkerer and want to print parts for projects: like I said before, this is a good motivation. Have fun!
2. You want to sell prints: this is a decent motivation, but be aware of the many pitfalls. If you have no technical proficiency at all, you may want to rethink this. Even the most advanced 3D printers still are not push-the-button-and-it-all-works. Some skills need to be acquired to get reliable good-quality results. Don't start selling prints until you have built up enough experience to know what your printer can and cannot do. When selling prints of models made by others, make sure to honour licences. Also be careful with supposedly “free” CAD software that usually is no longer free when used in a commercial context. (More details in the next section.)
3. You just want a 3D printer because it seems cool and someone you know has one, and you also want to print those cool articulated gadgets but have no idea what to do with it afterwards. This is a dubious motivation. It could turn out OK if by printing gadgets you get inspired to start making your own designs. If printing pre-made models is all you are going to do however, in the long term you risk ending up with both a machine that takes up space and gathers dust, as well as a big pile of plastic junk, and regret.

In Detail

In this section I dive a little deeper into reasons why a 3D printer may be a good investment or not, as well as some more general advice.

• Only buy a 3D printer if you have time to spare. It takes time to learn to know your printer and what it can and cannot do, and how much time it takes to finish a certain task. This is especially important if you want to sell printed objects. You need to have some confidence in whether you will be able to fulfil the requirements from customers, and in what time span you will be able to deliver the product.
• Printing useless gadgets is useful to learn to know your printer and its limitations, but after a while you will want to think twice before printing something. Unless your only goal is to print and sell existing designs, if you plan on buying a 3D printer, you should have a concrete idea for at least one project to print that will be useful in some way. Ideally this would not just be an existing 3D model but something you will have to design yourself. If you can reach that point, your 3D printing future has a decent chance of being bright. Otherwise it risks becoming a waste of time and money.
• The usefulness of a 3D printer grows considerably if you can design your own 3D models, even if they are only basic. If you are only able to print ready-made things downloaded from someone else, your printer risks being a poor investment and will likely start gathering dust after you have printed all the cool demos and gadgets. If you are not willing to learn to use some basic CAD program, I would advise not to buy a 3D printer.
• Even if your intention is to only print readily available models as a business, it still would be unwise not to learn basic skills for fixing and editing models, because a lot of downloadable models have flaws or can be made much easier to print through minor modifications. Your customers may also request for the printed model to be modified in some way. For instance you should be able to cut models into multiple segments, extract parts, or add simple structures like a flat base.
• If you plan on going commercial with your 3D printing business, be wary of supposedly free CAD program subscriptions. They usually are only free for personal and educational use, and you are left to the mercy of the software publisher's whims. If the company behind the software gets wind of you breaking their EULA, it could become expensive. There are some truly free and open source alternatives that might not be as fancy as the popular commercial programs, but they could suffice for your needs.
• Also if you plan to sell prints, please honour the licences associated with designs made by others. Many designs are released under certain Creative Commons licences. Although some dispute the legal backbone of those licences, they are gaining acceptance and sufficiently motivated authors can make your life very difficult if you violate their licences. Often the requirements are simple anyhow, for instance to honour the very common Attribution licence, it suffices to provide a visible reference to the source material on your store page and on a paper note included with the physical sold object. There is not even any need to explicitly ask permission of the author if you do this.
• A printer that allows to tinker may require more time to set up, but is a better long-term investment than something with high degree of vendor lock-in (a bit like Apple products versus building your own PC). For instance in 2016 I bought a Flashforge printer because at that time it was a pretty open platform that could be easily customised. This is no longer the case and I would not recommend a Flashforge printer today if you do not want to be locked into their ecosystem.
• Buy a good calliper and learn to use it. About 50% of all the things I print, involve measuring accurate dimensions on something for which the printed part is intended. There are pretty cheap digital callipers these days that work well. Because I dislike anything that quickly eats through batteries, I personally prefer a mechanical calliper with a clock-style dial accurate down to 0.01 mm.
  
• Multi-material printing is cool but of very limited use. A possible application is printing nice figurines without the need for painting, but mind that even with the better multi-material implementations, print time can be a multiple of a single-material print and there will typically also be a lot of waste material due to purging and priming. You will need a separate filament spool for every different colour you want to use.
  Another application is combining different materials like rigid parts with flexible joints, or regular plastic with soluble support material. Be aware though that not every pair of materials can be combined and not every multi-material printer is capable of combining any pair of materials.
• If you do want a multi-material printer, avoid the models that have two extruders side-by-side on a single carriage like the clones of the Replicator Dual. This is a rather poor design that exhibits many problems. A single extruder that swaps filaments like the Prusa multi-material design is better, but I believe the best multi-material designs are the ones with independent extruders (IDEX).
• Flexible filament really allows to make things you otherwise would have a very hard time making, so I would recommend to look for a printer that can handle TPU and similar soft filaments. This pretty much rules out printers with a Bowden type extruder, a direct drive extruder is required for truly soft filaments.

Printer Types

When considering consumer-grade printers, you are basically limited to two main types:

1. FFF or Fused Filament Fabrication, also known as FDM or Fused Deposition Modelling, is the kind of printer that uses a spool of plastic strand that is melted and extruded into thin lines that make up the printed shape. It is the most convenient type of 3D printer because there are no liquid reservoirs or hazardous powders to handle. This doesn't mean this printer cannot pose health hazards because the plastics may emit fumes when heated, and even those that don't may still emit tiny particles. Good ventilation is always important.
   This type of printer is pretty easy to get started with, although getting a good first layer with proper print bed adhesion remains a challenge even after all these years of evolution. The advantages are a wide range of materials to print with and which are generally cheap, and relative ease of use. Parts that come out of the printer are generally ready or need only minimal post-processing. Disadvantages are that these printers are not very suitable to print fine details or truly transparent objects, the parts are generally not very strong, and the printing process is inherently slow because the entire model basically has to be squeezed from a single tiny extrusion nozzle that has to move around.
2. SLA or stereolithography was actually the first type of 3D printer to be developed. It uses a light source (a laser or projector) to solidify material inside a tank of liquid resin. This has the advantage of allowing to create very detailed models, as well as a potentially higher speed (especially when using a projector which can do an entire layer all at once). The printed objects can also be truly transparent and are generally pretty strong. Disadvantages are the liquid resin which can be nasty to handle, may emit noxious fumes, and is generally expensive. Parts need post-processing when coming out of the printer: washing, drying, and curing for final strength. From a consumer point-of-view, this type of printer is not very suitable to print large objects, but if making tiny figurines is your thing, this is probably the printer type you're looking for.

Manage Your Expectations

Things one can and cannot do with a consumer-grade printer:

Can

• Print pre-assembled things that would be very difficult to manufacture otherwise: there are some classic examples like the Flexi Rex.
• Print with flexible and elastic materials: again, this allows to make parts that are difficult, cumbersome, or impossible to manufacture with other methods.
• Print with other exotic materials like conductive filament. Do not expect too much of the latter however: the conductivity is low and you are basically always printing resistors, but combined with flexible filament it is possible to make some really cool things.
• Print composite designs: with a multi-material printer one can print a shape that combines different materials. For instance you can directly print something that has rigid parts connected through flexible joints, which would be difficult or impossible through other means.
• Print very accurately: this does require knowledge, skill, and often patience as well. I can achieve tolerances approaching 0.02 mm, but this requires controlling every part of the process like room temperature and cooling fan speed, and often also modifying the model to compensate for certain inherent limitations of the printer.
  

  An adapter to mount a mini drill chuck meant for a Dremel-like tool in a larger drill. The thread is M8 with finest standard pitch (0.75 mm). Printed in PETG with 100% infill, this is a very strong part.
• Print strong parts: contrary to popular belief, not every 3D printed part is weak or will crumble under minor stress although that is indeed likely what one gets when just going the plug-and-play route. With the right materials and printing process, it is possible to print parts that approach or sometimes even exceed the strength of injection-moulded parts. But again, this requires experience and skill.
• Repair things you could otherwise almost never repair: when combining some of the above bullet points, it proves possible to print replacement parts that would otherwise be impossible to obtain due to age of the device, or general unpleasantness of the manufacturer. I have replaced many a device using 3D printed parts and in some cases, I expect the replacements to be better than the original parts. It is also possible to create adaptors to make old devices work with new components.

The original strap and cradle for this detachable smartwatch had disintegrated after a few years. I 3D printed replacement parts in TPU and PETG. These are now lasting longer than the originals.

Cannot

• Print fast at high quality: this is a trade-off and even at the fastest speeds that produce parts that are not total garbage, it still takes long. I always prefer quality over speed whenever possible, hence I usually print slowly.
• Print any possible shape: there are constraints to what shapes can be printed because aside from the limited ability to bridge gaps, a printed layer must always have sufficient support from something below it. Some of these constraints can be circumvented by printing support material, but even then there are still constraints. Don't expect to be able to print anything you want. Resin printers offer slightly more freedom than FFF but in general the same constraints apply. Sintering-based printers (SLS) have more freedom when it comes to printable shapes, but they have the same drawbacks as metal printers and you won't find a true consumer-grade SLS printer anytime soon.
• Print metal as if it is plastic: the process for printing with true metal is complicated and expensive, and requires a huge machine and dealing with hazardous powder. Metal 3D printing at home will not become mainstream anytime soon until someone figures out a way to simplify the process, but even then it will probably still remain expensive. The closest you can get to metal printing with the typical FDM printer, is printing with metal-infused plastics. The result is a part that has some of the properties of a true metal part, but will not stand up to the same demands as a true solid metal part. You can indirectly make solid metal parts with an FDM printer through the method of investment casting, but this requires working with molten metal hence is also far from trivial.

Where to Find and Share 3D Models

Long ago the obvious answer was Thingiverse. Today it is much harder to recommend a single website and I won't give a strong recommendation. Look around and do your own evaluation, the landscape continually changes anyway.

I do can say one thing: do not waste your time on Thingiverse anymore, especially not if you are looking for a reliable place to distribute your own models and run some kind of 3D-printing related business. Thingiverse's golden days when it had a vibrant community are long gone. I explain this in more detail on my former attempt at an issue tracker, but in a nutshell it is as follows.
Thingiverse is now owned by a company that just happened to inherit the website as part of its acquisition of Makerbot, and this company does not care about Thingiverse but they also don't dare to simply ditch it. All the employees who acted as true community moderators have left or were fired, and have not been replaced. This leaves the site in a state of palliative care. There are tons of bugs and problems that make the site a chore to work with, and it is very hard to have them fixed because communication with the site maintainers is mostly a game of shooting in the dark. The seldom moments when the site gets a significant update, are either because they want to try to milk some ad revenue from it, or make it appear superficially trendy. Each time they do fix an issue, they will likely introduce 2 new others.

If you have already uploaded models to Thingiverse under the delusion that they are safe there, I strongly recommend to ensure you have a backup of those files elsewhere, including any descriptions and photos that you would want to use when uploading the models to another site. I deem the risk pretty high that the whole site will go belly-up in the near future and even if that does not happen, it is still possible that your files will be randomly deleted.

Arguably the only reason why Thingiverse still exists is due to its legacy, its cobwebbed good reputation from the early days, and its huge archive of models—although there are indications that the owners of the website have started to randomly delete less popular models, maybe to avoid having to invest in more storage space. Thingiverse is still worth a look if you are searching for something, but I consider uploading models there a frustrating endeavour and I stopped doing it. Some like to defend the site with the “but it's free” argument, but there is no such thing as a free lunch. At Thingiverse, you pay with frustration and anguish, and having to find silly workarounds to get things done that would be trivial on any well-maintained website. If time is money, then Thingiverse is expensive.
Speaking of searching: do not bother using Thingiverse's own search because it is often broken. Use Google instead and add site:thingiverse.com/thing to your query.
Also do not try to use the site's Customizer because you'd be lucky if it would work. Just run customizer on your own computer instead.

If you really want some kind of recommendation for a Thingiverse alternative, at this time I would point to Printables.com, which seems to be the most similar to how Thingiverse was before it all went south. Given that Printables is also backed by a single manufacturer (Prusa) however, there is no guarantee that it will not develop similar problems as Thingiverse at some point in the future.

Unfortunately every major 3D printer vendor now believes it is a good idea to launch their own Thingiverse clone. When Printables launched, it was mostly a good thing because it was actively maintained while Thingiverse was starting to rot away. Now however, we have multiple new repository sites, and this makes no sense. The 3D printing community will become scattered and authors of 3D models will have to create and maintain profiles on each of these sites to be maximally visible, as well as avoid that someone else impersonates them. Users will need to scan each of these different sites when they are looking for a particular model, because some sites will have models that others don't have. This is a silly and annoying situation. It would be better if an independent repository would be created that is supported by all the major printer manufacturers, and that consolidates all the 3D models in one place. This would benefit everyone.