I am looking to buy a 3D printer for my son (and for myself too). We want to print, not tinker, so it should be something that gives great results right from the start.
Can you guide me to a sensible choice?
My first choice would have to be the Prusa MK3S Plus but it is outside the price range I am shopping for, except if I buy used – would that be bad to do?
Realistic choices:
- €380 used Prusa MK3S+, with 10 days printing time
- €400 new Prusa Mini+
- €250 new Ender 3 V2 Neo
Criteria:
- High quality, no hassle. I want to print, not tinker.
- Preferably (semi)assembled.
- Auto bed leveling.
- Auto error detection (filament, power, etc.?).
- Budget up to 600 EUR/USD including extras, excluding filament.
- Speed is not important.
- Size is not important.
- Must not be cloud-based.
Questions:
- Surface?! Smooth, os satin, or textured? (Why) Should I have more than one kind?
- (Why) Do I need an enclosure?
Some notes:
Auto bed leveling is completely overrated. On a decently calibrated printer on a decently flat surface with decent spring tension you need to level the bed maybe once every few months. Auto bed level can also only correct small misalignments and it does so by skewing the one surface of the print. It’s at best a small gadget to make the first layer overly perfect, and it’s totally not necessary.
Powerloss recovery can be enabled on every printer by using (or compiling) the firmware with that setting enabled or by using an external print server like octoprint or repetier server. But: if enabled in firmware (not on an external print server) it causes bad print artifacts and wears down the SD card rather quickly. I have had a single time in 6 years of printing where I accidentally turned off the power mid-print and not a single time where I had powerloss due to things outside of my own stupidity. It’s not a feature that I’d place any value on, unless you live somewhere where power outages are very frequent. Also, if you recover after a power loss and the hotend has cooled, it needs to be heated up in-place to basically melt the hotend free from the print. You will have a massive gash at that point.
Filament runout sensor is a nice gimmick, but again only important in special cases, namely you intend to print really huge parts that take multiple rolls to finish. Otherwise you’ll just use the last bit of filament to print small things.
Size: I’d recommend you a bed that’s 220x220 to 240x240 as this is the standard size. This means, you will be able to print almost anything you find online without issues. Larger only matters if you have specific use cases in mind. Prints that huge will take very long to print though (longer than a week), so you might not want print anything larger than a standard board anyway.
Without enclosure you can easily print PLA, PETG, TPU/TPE and filaments based on these materials. You need an enclosure if you want to print ABS or Nylon or other specialized materials. An enclosure helps to keep the air warmer and draft-less to avoid warping for filaments that tend to warp. Also it allows you to use air filtration to avoid the toxic fumes that ABS or Nylon tend to create during printing.
Surfaces are a solved problem by now. They will all work fine. The only difference is cosmetic. I personally would go with a textured surface since all the other sides of a 3D print are slightly rough and it looks a little dumb if one side is super smooth.
One thing to really watch out for: Before your first print (and if your prints stop sticking) wipe it thoroughly with a cloth rag soaked in isopropyl alcohol (IPA). This is a step that you really shouldn’t skip, otherwise your prints won’t stick to the surface.
Used printers are a massive gamble. There are many things that can be subtilely broken or damaged that you will not notice if you don’t know exactly what you are looking for. It will just print really badly and you’ll never figure out why without the propper knowledge. If you don’t want to tinker a lot, don’t get a used printer.
Both the Mini+ and the Ender 3v2 Neo are really good printers. The Ender 3v2 Neo will get you a bit more for your money, but the Mini+ is really cute and comes with free Haribo.
Saying all surfaces are equal is wrong. Using PEI sheet with PETG or TPU (and no glue) is a nice way to destroy it.
I read about this so many times but I’ve printed with PETG a lot on both glass and PEI and never come close to experiencing this.
Lucky you, I was too lazy to read and ruined a brand new sheet the first time I was printing PETG.
Same with glass, I ruined a few with PETG.
Why would you use a glass bed in 2023?
Clean glass is best for pla imo. Also its flat. You need flat bed to avoid auto bed leveling and tooling plates are expensive. I know most peeps here disagree with me hehe
My stock Ender 5 print plate is super flat and I don’t need any auto bed leveling.
The print plate of my Tronxy X8, which I had before, was also super flat.
With spring steel PC, I need no surface treatment before printing PLA, PETG or TPU. It sticks perfectly every time and it comes off easily after cooling down.
No glue strick, hairspray, tape or other stuff that belongs in the 2010s needed.
It’s very flat, it works well with PLA, and it’s what my ultimaker 2 came with.
For a 10 year old machine that makes sense. But honestly, get your self a buildtak clone surface and stick it onto your glass board.
You won’t be going back to the glass, I promise you.
Costs maybe €10, takes 2 minutes to install and you won’t ever have to mess with glue stick, hairspray or any other surface treatment.
Largely agree, but two small counter points.
Active bed (or gantry) leveling, which actually squares the bed relative to the extruder, is very nice. I’m wrapping up a Voron 2.4 build and this was one of the features that motivated me to go this route. Sure it wasn’t the only one, but I was so tired of my Wanhao I3 clone maintaining bed level as you said then radically loosing it because one of the two z steppers randomly decided to misstep.
I haven’t built a filament runout sensor into my Voron, but probably will fairly soon. I tend to print larger prints and really disliked the constant game of “I wonder if I’ll have enough filament” as I got to the bottom of a spool.
Yeah, these are sure nice quality of life things, but also not super high priority for a beginner.
Regarding the active bed levelling, there are a few other options to get the same effect. First, if you have any of the TMC drivers with stall guard (and it setup correctly), random missteps just don’t happen. That’s maybe the best way to go since it fixes the issue at the source and prevents it from happening during the print.
A second option is to tie your Z axies together with a belt. That way they all do the same, and you can just replace all these motors with a single big one.
Or you go the creality route and just use a single Z axis, but that requires a really stable gantry/bed, depending on what hangs off your z axis.
Also, skipped steps (especially on Z) aren’t really a normal thing and point either to a mechanical issue (check bearings, belts, pully screws and obstacles), too weak motors (rare if you are using stock motors) or too little stepper motor current. Especially the last point. Turn the current up a little and see if that resolves skipped steps.
I haven’t had a single skipped step on my current printer, which I have had since 1.5 years. Did a motor current tuning when I got it (necessary because I swapped the mainboard) and that’s it.
I disagree on the “beginner” part. Yes, I am a beginner, but that does not mean I want an entry level device, nor that I want to replace this device soon.
I want one solid machine that I will be content with for years. So any QoL details would definitely be useful, even or especially to a beginner.
More systems means more that can go wrong and more difficult trouble shooting. No matter what printer you get, stuff needs to be tuned, stuff needs to be maintained and stuff breaks.
Getting the biggest best do-it-all device with all the bells and whistles (like a fully speced Voron) means not only that you spend a massive amount of money for a machine that does the same thing just a bit faster, but also that you have tons of things you need to watch out for.
Auto bedlevel, for example, is by far not a fire-and-forget solution.
Upgrades are also a thing. Once you get into printing and understand what it’s all about you will learn what you want and need. This allows you to upgrade the machine and make it better. Especially the Ender 3 series is built with upgradability in mind. They have a lot of drop-in upgrades that are as simple to integrate as the (very simple) initial setup of the machine was.
If you buy your first car you also don’t start out with an 800 PS super car or a semitrailer.
Sure it’s fun to upgrade but there are things you should just want to have on your 2023 machine from the start. It’s more like you’re suggesting a starting driver that they don’t need electric windows, cruise control or whatever because they have to learn why they’d want it first…
That’s why I got the v2 Neo and am very happy with it, as it’s got the stuff you’ll want after one week anyway but pre-installed and very cheap as a bonus: ABL, all metal extruder, improved hotend, better springs. There’s no reason to pain yourself and your wallet by waiting with these ‘until you get more experienced’ IMO
Better mechanics are never bad. But the points we where talking about weren’t that at all. We were talking about automatic alignment of multiple Z motors. That doesn’t apply to any stock Ender 3, because they all only have a single Z motor. No alignment needed, and actually no alignment possible.
Regarding the other points:
But of course the recommendation would be for the newest version of Ender 3, because there’s no point in starting off with an outdated machine.
Again, the upgrades the other guy mentioned where not that.
The problem with dual Z is when you turn off the printer, motors lose power and can missalign. You can connect motors with belts or just print some brake-like part that makes enough friction to stop motors from moving under weight of printer itself. Idealy you will have 1 motor per axis
I had a dual Z printer before (Tronxy X8), so I know the issue.
I already mentioned tying the motors together with belts.
But if you want to not do any modifications, you can also move the print head off the bed, disable software endstops and move the Z axis down until both motors start to skip. That sounds bad but doesn’t hurt the mechanics or engines at all, they are built with skips in mind.
That also aligns both Z axies.
As someone who’s started printing a few weeks ago (E3v2 Neo) i’m very happy with the QoL upgrades on it. Regarding auto bed levelling: I think you misunderstand what it does, as it’s not about maintaining Z stability/accuracy at all. It corrects slight bed level problems from the springs but also imperfections in the bed material itself.
Believe me, I know what it does. At this point I would respectfully point to the fact that I’ve been printing for 6 years, have been running self-compiled Marlin for over 5 of that and have been fixing up the printers of 6 people in my friend group.
If your printer runs Marlin, your printer contains code I wrote.
I’ve been doing this for more than a few weeeks.
Also, you misunderstood what that guy before meant. He wasn’t talking about simple Autobedleveling, but rather about auto-aligning multiple Z axies. See, if you have multiple powered Z-axies (e.g. if you have a bedslinger with two Z axies or a Cube-style printer that moves the bed along multiple Z axies), these Z-axies can become misaligned if one of them skips a step or you power the printer off and they become misaligned. There are multiple solutions for fixing this, and the guy before went for the nice but expensive route of controlling each Z axis with a separate stepper and homing each of them separately. That is what I said was a nice gadget, but not a must-have feature for a beginner.
Now regarding classic auto bedlevel: It’s meant to correct slight misleveling and bent beds. It does so by purpously warping the print to follow the misalignment of the bed. This means, you’ll end up with a print that is not straight. The reason why ABL exists is that 5 or 10 years ago, springs and beds were utter crap and thus people had to workaround in software.
In 2023, if your bed loses leveling all the time, you have the whole bed leveled too high so that the springs aren’t tensioned correctly. On my printer I have to re-level the bed maybe 2-3 times a year and that’s usually related to modifications like using a different nozzle.
Also, in 2023, if your bed is so bent that you’d need to use ABL to compensate, that’s a warranty case.
If you actually don’t RMA such a board but seriously try to compensate it’s failings with ABL, you can choose between a fast 9-point ABL, which does nothing, a 16-point ABL which doesn’t measure the center point, a 25-point ABL which does a bit more but takes forever or you go even higher and spend more time leveling than printing if you do small prints. Also, you need to re-level every time you print with a different bed temperature.
All in all: don’t compensate mechanical issues in software. Fix your mechanical issues.
Oh man am I so happy to hear someone say this.
Go over to the clipper forums and watch people spend days calibrating resonance compensation instead of just installing a brace on the tower or some shit.
Software compensation is good for stuff you really can’t fix.
But people use it to compensate stuff they totally could fix, e.g. a badly levelled bed. Almost as good as those guys who level the bed using a spirit level.