Alright, so want a 3D printer, but - which
one to choose? In this video, I want to help you plot out
what your use case is, what that requires in a machine and what to generally look for
and what to avoid in a 3D printer. This video is brought to you by Slice Engineering. Their Copperhead hotend system makes it easy
to switch to a high-performance, all metal hotend. With a bimetal heatbreak, it reduces heat
creep and clogging and it comes in versions that directly fit into a huge variety of 3D
printer toolheads. With a maximum temperature of 450°C, you
can print virtually any material, and the highly compatible design lets you use your
favorite nozzles, temperature sensors and heaters. Check it out at the link below. So first of all, there is no “best” or
“perfect” 3D printer. Even when you find someone that claims that
this one machine is the best 3D printer ever, it might be, but just for them, and it’s
just for that one use case, which might just as well be totally different to what you actually
want to do with the machine. And before we get into the meat and potatoes,
I want to be very clear about why I chose Prusa for the filament and filament printer
sponsor for this series, it’s because they are the machines that I personally use the
most, they are a well “balanced” printers, and, most importantly, the MK3 and the Mini
work well for demonstrating what I’m trying to show here on the series. On the same hand, just like any other printer,
by no means are these machines the perfect choice for everyone, but they happen to work
well for what I’m most often using 3D printers for. So, what should you be looking for. Let’s take a quick look at resin vs. filament
printers first. The next video in this series is actually
going to be entirely about the ins and outs of resin printers, but let’s go over the
brief version real quick: Resin printers are great for making small, but high-detail parts,
but to use them, you need to be able to safely handle the resin itself and the solvents involved
and then dispose of them properly when you’re done. The process of getting a usable part also
takes more effort because you can’t just use resin parts straight off the printer,
you’ll need to wash and then cure the parts before they’re actually done. If you need the best possible surface finish
and detail reproduction, that extra effort may be worth it, but in general, filament
printers are easier to use and still give you pretty good detail on the parts you print
- and you can choose if you want your parts to finish more quickly or have finer details,
too. Filament printers are also cheaper to run,
and, what I think is actually much more important for getting started with them, you can watch
them work, and when something starts going wrong, you can see how it’s happening. In a resin printer, you don’t even get a
peek at your printed part until at least an hour into the print. But like I said, if you think want a resin
printer, we’re going to look at that in the next video, for now, let’s stick with
filament printers. Or FDM, or FFF, or whatever non-trademarked
abbreviation you want to use. A filament printer’s basic job is to heat
up and melt the plastic filament and then squeeze it out of its nozzle in the right
spots to make the exact model you told it to create. That really is the same, no matter which filament
printer you have. They all make parts, but which parts and how
well they make them, that depends on how the printer is built and set up. The first thing you’ll usually see on the
specs is how large the build volume is, so what size part the machine can print. Most printers are around the size of being
able to print a 20 cm or 8 inch cube, that size is a sweet spot for a lot of reasons:
At that size, the printer itself isn’t too big to handle yet, and the components it uses
are still light and short enough to keep the motion of the printer quick and snappy and
accurate. When you have for example a larger moving
bed, because that will also be heavier, the printer will either need a more powerful motor,
driver, and belt combination to maintain the same snappiness as a smaller one, or, to not
compromise on accuracy, it will need to limit how fast it positions itself if it’s using
the same common NEMA17 motor size and 6mm belts. That size is also where a “normal” toolhead
will finish prints within reasonable timeframes. Printers like these will still take close
to week to finish a print that fills up its entire build volume, but if you now double
that 20 cm cube to a 40 cm cube and fill that up, you’re now talking about print time
closer to a month. There are specialty toolheads for those larger
printers that can feed more filament and heat it up more quickly, so that’s going to make
prints complete faster, but usually, you do lose out in detail reproduction because those
toolheads are also laying down a thicker bead of filament. But typically, you’re not going to fill
up your printer’s build space entirely, and most prints will take two or three hours,
with larger prints that use, for example the entirety of the printer’s height, those
will usually take a few days at most. If you want to find out how long a specific
printer you’re looking at will take for the parts you want to print, you can set up
the slicer, the software for your printer, load in your model and just see what sort
of print time estimate you get. That takes me to the next, and I think most
glossed-over difference between machines, which, especially for getting started, I think
is equally or maybe even more important than how exactly the printer is built - and that’s
what the manufacturer provides for you for software, specifically for the slicer. Again, you can normally give this a try before
you even buy a printer. The best case scenario is that you get a slicer,
that, after you install it, just asks you “hey, is this the printer you have” and
then sets itself up to perfectly prepare your files for that very machine. That means that the manufacturer has set up
and tuned the settings to a level of quality that they are happy with for that machine
and it means that you don’t have to go through any debugging and tuning - and printer tuning
is anything but self-explanatory, especially when you’re still wrapping your head around
how all these settings and components work together.. What you might also find from the manufacturer
are recommendations for which slicer to use that has profiles already built-in for their
machine or something like a downloadable profile and instructions on how to use those. That’s also great because you get a solid
starting point. What would be red flag for me is when there
is just nothing provided at all by the manufacturer. Facebook groups and forums often do share
profiles that work for that specific printer, but you never know if those profiles come
with quirks that create weirdness in how some parts come out, and also it’s also not very
reassuring, I think, when the manufacturer can’t even be bothered to tell you how to
use their machine. If you don’t find anything on the manufacturer’s
website, it certainly doesn’t hurt to ask their customer service what slicer setup you
should be using. If they have a customer service. Alright, so that’s the “soft skill”
portion, let’s move on to the “hard” skills - actual hardware, features and how
the printer is built. These things will determine how capable the
printer is, what materials it can print, how convenient it is to use and how long it’s
going to last. What makes recommending one thing over another
so hard with a lot of these details is that, when there are two common approaches, either
one is usually going to work if it’s done well. Just like the first obvious thing, other than
size, and that’s kinematics. For the sake of keeping this video short-ish,
I’ll generalize all printers that have their movement axes perpendicular as “cartesian”
machines, where the other category is “delta” printers. There are some more obscure designs, too,
but I’ll just ignore those for now. I would recommend starting with a cartesian
machine of some sort, deltas can be built incredibly well, too, and they’re fascinating
to watch and listen to, but any flaw in a Delta get amplified in ways that can make
them really hard to track down and fix. I’ve had a delta machine produce distorted
parts for years and only noticed it when I was wondering why the printed parts weren’t
fitting together the way they were designed. Again, there are exceptions and really well-built
deltas, but in general, I find cartesian machines easier and more predictable to work with. Next, the one thing you’ll often see discussions
about is what sort of hotend you should be using, the thing on your printer that actually
melts filament, and there’s two large groups being all-metal and Teflon- or PTFE-lined. Teflon and PTFE are the same material, it’s
just that Teflon is a brand name. The difference between those and all-metal
hotends is that the lined hotends will use a bit of PTFE tube to keep the filament from
sticking to the inside of the hotend where it transitions from the low-temperature, cooled
side, to the hot side, where the filament melts. Lined hotends use PTFE because it’s naturally
slippery, all-metal hotends have a special surface finish on the inside of that transition
where it touches the half-molten filament and they keep that transition zone as short
as possible. Both lined and all-metal hotends can cause
some headache if they aren’t well-made. From my experience, if you have a decent all-metal
hotend, it’s going to keep working with basically no maintenance until you find a
way to break it, like bending it; but with a lined hotend, because the PTFE is still
a plastic, it does get soft and deforms over time, so it’s most likely going to require
periodic replacement of that liner. On the other hand, a badly-made lined hotend
has a better chance of working at all compared a bad all-metal one. So which one to get kind of depends what budget
range of printer you’re looking at, too. One thing the all-metal hotend will allow
you to do, though, is to work with higher-temperature materials, but to print those successfully,
you will also need a few other things like a printer enclosure and a higher-temperature
heated bed. If you’re planning on printing with ABS,
ASA, polycarbonate or some of the Nylons, to have a good time, you should be looking
to buy or build an enclosure, preferably get a printer with an all-metal hotend and a capable
heated bed, but if you’re ok with the two common materials PLA and PETG, then pretty
much every well-made setup is going to work for you. Some manufactures also specifically validate
some materials on their machines, so for example, even though the Prusa Mini uses a lined hotend,
they’re providing settings that let you print their Prusament ASA on this machine. More info what those materials are and what
you’d use them for in a later video. Usually, the next logical thing to talk about
now would be the extruder, the thingy that pushes filament into the hotend, but because
even simpler designs are usually good enough to at least get you printing and with a good
printing profile, the difference between the two main ways of building them, where you
either have the extruder sitting directly on to of the hotend in a “direct” setup
or coupled to it with a long PTFE tube in a “bowden” setup, the difference between
those in practice is so small that you really shouldn’t worry about which one your printer
has unless you’re planning to print a lot with flexible filaments - in that case, a
direct or non-bowden extruder is a much better choice, but for everything else, it kinda
doesn’t matter all that much. So I’ll move on to the heated bed that we
were just talking about, it’s a standard feature on almost any printer now, it heats
up before a print and then cools down when it’s done, and that heat not just helps
with keeping parts stuck during a print, but it also helps with releasing them afterwards,
because as everything cools down, the part and the bed will shrink at different rates
and in the best case that just pops the part right off. The one feature to look for here is a flex
bed. Printers without one are still usable, but
it’s incredibly convenient when you can just pull off the bed surface, slightly spring
it up and the parts will just pop off. On printers that don’t have a flex bed,
you’re often going to need to pry off larger prints with a sharpened spatula or a chisel
or something, and you need to be careful not to damage the bed surface. Looking for a machine that comes with a flex
bed is something that I would definitely recommend, they are also available as an upgrade that
you can fairly easily put on almost any machine, but if you choose an aftermarket option like
that, it does add to your pricetag. Now, if I had to choose between a flex bed
or auto bed leveling, I’d go for the flex bed for sure. But sensor-based auto bed leveling or tramming
or however you want to call it has become another useful feature that allows you to
only worry about setting the nozzle height of the printer once instead of having to periodically
adjust the height for each corner of the bed by hand. A level or square and trammed bed is essential
for good adhesion of your part while it’s printing, especially of larger prints, and
as a printer bed itself becomes larger, it becomes almost impossible to get every single
spot on the bed to the same height, especially in between your leveling points. Auto bed leveling can adjust for that and
for any warp or bow in a bed, so I think for machines with a 30cm bed and up, auto bed
leveling almost becomes a necessity, for smaller machines, it’s still a great convenience
feature, but manually adjusting the bed now and then is totally doable, too. I’m covering how to do that later in the
series. Before we move on with what else you should
be looking for, let me cover a few things that some manufacturers love to use in their
marketing, but actually say nothing about how a filament printer is going to perform
at all - or at the very least you shouldn’t be comparing machines based on who is claiming
the best numbers here. So those things would be something like “maximum
print speed”, “resolution”, “finest layer height” and anything else like that
is really just a theoretical number. For speed, yes, the printer may somehow be
able to move that fast, but typically you limit your print speed way below what the
absolute limits of the machine are because if you actually tried to print that fast,
your prints would look absolutely horrible. For resolution, the theoretical numbers are
way less important than how smoothly and accurately the printer can move. You’re not going to notice a difference
in a bit of extra “positioning resolution” when, for example, the belts that have to
transfer that motion are made from jello. Having good belts and bearing and all that
does make a difference. But something that, as of today, is not really
a good indicator of how well a printer is going to perform is the choice of an 8-bit
vs. 32-bit controller. The Marlin firmware that normally runs on
the 8-bit chips is extremely well-optimized these days and a printer like the MK3 with
that truly ancient 8-bit processor in there, at least for printing, still works perfectly. The more modern 32-bit chips are actually
cheaper and they allow for more functionality like easily having a nicer screen on a machine
or adding network connectivity. But when these 32-bit chips run a hastily
thrown-together, proprietary interface on the screen or unoptimized code to control
the printer, they’re actually going to be inferior to a well-oiled 8-bit system, I think. There are good and bad examples for either
one, but honestly, though, 32bit vs 8bit is such a marginal difference in what your printer
is going to enable you to do that I would just not care. It depends more on how well everything is
implemented anyways, and that’s something that’s hard to judge from afar. Ok, but what will make a real difference in
the electronics is the choice of stepper motor driver, and I know we’re kind of wading
into the deep end here, but trust me, it’s going to make quite a difference. Better stepper drivers will get you a smoother
surface on your parts and will make the printer much, much quieter. The difference between a manufacturer paying
no attention to them and another one engineering the crap out of a stepper driver system can
be going from “holy crap I can hear this printer through two concrete walls” to not
noticing that the printer is still working when you walk into the room. I’m not exaggerating here, I’ve had printers
that landed on both extremes of that scale. Better machines will use Trinamic drivers,
there are some off-brand drivers out there as well that are labeled as “silent”,
those aren’t always great, but they will still be better than the classic, non-silent
drivers. For the rest of the electronics, it’s hard
to judge them based on a shop site and the manufacturer's materials, so have a look at
what other people and reviews are saying about issues that might pop up. A few printers used to ship with control boards
that would sometimes set the entire printer on fire, others have bad wire management that
gets caught or pulled on or bends in the wrong spots and eventually fails more or less spectacularly
only after you’ve printed few spools of filaments. Loose wires hanging about the printer and
spots where cables kink every time the printer moves should at least make you suspicious. I’ve got a whole video about that here. For the last aspect that I want to cover in
this video, you should think about why you’re getting a printer. Is it a tool for you, or is a hobby because
you want to learn about the technology. It’s totally legitimate to buy a printer
with the intent to immediately start modding and upgrading it. You will learn a lot about these machines
by working on them and hopefully, you’ll have mastered and understood your own machine
once you’re done. But you should be realistic and allow for
some snags and hiccups along the way and for being without a working printer if something
doesn’t go perfectly smoothly. I’ve seen people get frustrated and give
up on 3D printing entirely because they went head-first into a modding mania that they
weren’t prepared for. Getting a printer that is does not require
mods to be at least usable at a basic level is something that I would highly recommend. Most popular printers manage to do that these
days, and if you want to upgrade components like the hotend to print new materials, or
add a flex bed yourself, you can usually still go in and do that. Like I mentioned in the last video, open-source
machines really make that easy. Of course, if you want maximum involvement,
you can also source a machine from scratch, but that’s not something I’d quite recommend
as a beginner project. If you just want a tool you can rely on, it’s
probably a smart move to get a machine that has everything you’re looking for, with
a warranty from the manufacturer, instead of tinkering with it yourself. Most printers come almost completely assembled
these days, but if you are considering getting a kit with a proper assembly manual, that’s
a great learning experience or family projectl. So, ok I think that should cover everything,
pretty much. I know it’s a lot to take in, but hopefully,
considering some of these basic now is going to save you some frustration later on. Let’s do a quick recap! To start out, a filament printer is easier
and cheaper to use than a resin printer, but doesn’t quite get you the same level of
detail and surface finish. Pick a printer that is big enough for the
parts that you want to make with it but don’t just go for the largest one you can find,
and keep in mind that large prints will take significantly longer than smaller one - sometimes
even weeks. Check what software, if any, you can use with
your printer before you buy it and try running some of your designs through it. For printer hardware, I’d recommend a cartesian
machine instead of a delta, and whether that should have an all-metal hotend or a PTFE-lined
one depends on what materials you want to print and what price point you’re shopping
at. Good comfort features to have are a flex bed
and maybe a bed leveling sensor, but you can safely ignore the manufacturer’s claims
of “resolution” or “maximum print speed”, and instead try and find out if they are using
sub-par mechanical components. Look for silent stepper drivers and the community’s
experience with the electronics overall and don’t plan on relying on mods and upgrades
to get a particular printer to work for you. You can still mod them to expand the functionality,
but a printer should at least be usable out of the box, I think. So that’s it for this one, in the next video,
we’ll look at resin printers, until then, thank you for watching, make sure to get subscribed,
keep on making, and I’ll see you later.
Pretty based way to debunk a silly article. :)