X2D opinions
I’ve owned my X2D now for a little under a month and have over 200 hours on it. I’ve been 3D printing for about 11+ years at this point and I’ve owned and interacted with a plethora of devices; acrylic i3’s, Crealities, Prusas, Vorons, etc. I’ve also owned a couple other BBL machines (A1 Mini and recently P2S).
The reason that I wanted to side-grade to the X2D is the secondary nozzle. Dissimilar supports are a feature that no other machine on the market has (had) at a relatively affordable price. Some can argue that the Snapmaker U1 is a solid alternative but it’s not enclosed by default and once you add the enclosure to it, you’re moving away from the price/performance offering of the X2D.
I’m going to be covering my thoughts of the machine itself, not the AMS 2 that I got as part of the combo. Mine had the feed issues and my filament hub replacement is en-route but that’s its own thing.
TL;DR for anyone who isn’t bothered to read the following stream of consciousness. The X2D is a well-engineered, feature-dense machine at a compelling price, currently held back by half-baked firmware and poor dual-extrusion UX in Bambu Studio. Both (hopefully) will improve with time. If you already know what you want from a printer and have a specific need for dissimilar supports or engineering-grade materials, it earns its price tag. If it’s your first machine, skip it for now. Get an A1/A2, a P2S, or build a Voron/Rolohaun/Prusa, if the Bambu ecosystem doesn’t matter instead.
The hardware
I feel that the hardware offered for the price(€630 + shipping solo printer) is insanely appealing; chamber heater, secondary nozzle, servo based extruder, filtration, whatever else I’m forgetting.
The AUX nozzle being bowden is (in my controversial opinion) an excellent choice. It’s marketed as an aux nozzle and its best application is dissimilar supports; PLA/PETG, PLA/PVA, TPU/PLA, ABS/HIPS, etc. It’s slower and can push less volume of material but that’s fine. You’re printing supports on the slower end anyway even with your main nozzle to avoid them breaking and turning into a spaghetti mess so you lose almost nothing due to the slower bowden extruder. Some people seem to misunderstand the point of the second nozzle but if you go into it knowing that this is primarily for dissimilar support and maybe at max 2-3 layer color changes (labels, etc) or even just 2 colors during the whole print, you’re going to have a great time. For all other multi color prints, the AMS experience is pretty much what you expect of it.
The belts are a nice change (I guess?) but they didn’t get rid of VFAs, they just shifted them to a different speed band. VFAs appear at speeds where the combined periodic disturbances of the system; belt tooth engagement, stepper motor torque ripple, and any other mechanical periodicity, produce visible banding on the print surface. Yes, belt pitch is a factor, but so is stepper motor torque ripple, which runs at its own frequency independent of the belts. By changing belt pitch you’re adjusting one source of periodic disturbance while the other remains. End result: you’ll still get VFAs, just at a different print speed than you’d see with 2mm belts. More of a marketing shtick than an actual fix.
The assembly and engineering of it all is also very impressive. It’s a complicated machine with complex systems packed into a tiny package. There’s a lot going on and a lot of care went into ensuring redundancy during the whole printing process. It’s trying to make the whole thing as fool proof as possible and for the most part I think it succeeds.
All this complexity however comes at the cost of repairability. The extruder disassembly looks horrendous and considering that Bambu markets itself as an “easy to use” machine, maintenance, while fantastically documented, should also be easy! Could some things be done better? Absolutely:
- I think the fan configuration could be set up to a push pull arrangement instead of the current push push. I don’t have any technical data to back this up so take it with a grain of salt. But from ambient nozzle temp readings, it’s as though a small hot air pocket gets formed between the two nozzles and there’s nowhere for it to go. This can cause issues (while bambu doesn’t recommend it, idgaf) when printing ABS/PETG. Excess heat of that enclosed space can cause the PETG to soften in the cool side of the hotend since it sits ambient at around 65-70C. My current workaround is to disable the chamber heater for those prints.
- The filament path for both the main nozzle and aux is over-constrained and under-constrained respectively. The main nozzle should be using a 3mm ID PTFE to make sure that the filament can easily travel to the hobbed gears (with the default cable chain arrangement) of the extruder and the aux would be best with a capricorn style 1.9mm PTFE tube to further minimise the flex of the filament through the long filament path (mine came stock with 2mm ID tube). You’re likely going to want to print a riser of some sort to free the main nozzle PTFE from the drag chain and smooth out the path a little. (I printed the P2S/X2D riser by 达达玩).
* This is a silly nitpick but ship the machine with two spool holders. Even if you’re going for the combo, the ability to throw on a spool of ABS as the external main nozzle filament, keep the PETG loaded into aux and not have to fiddle with an AMS loaded with PLA would be a big QoL improvement (yes I can print a spool holder but cmon…). A commentor on Reddit mentioned that their non combo machine shipped with two spool holders, so this point is a little moot. Where’s my second spool holder holder with he combo version Bambu?!
The software
This is where my opinion takes a turn. I think the firmware and slicer usability is half baked at best and hopefully will improve with updates.
Firmware
This feels half baked. Almost there but just playing it a “bit too safe”. Works in internal testing but then collapses a little in the hands of end users.
- Filament feed issues seem to be the most common problem I’ve seen. Just up the threshold in the firmware. More often than not, the AMS will feed the filament 1-2mm before the hobbed gear and get into a loop. Just push it in there, you’re so damned close. The machine will check for a clog when it asks you to confirm that filament is coming out in the next step anyways and surely if it’s automated during printing and you’re doing color swaps on the AMS, either the servo motor overload error or the AI detection system will be able to tell that something’s up. You’re taking it apart anyways at that point so why add this seemingly pointless redundancy?
- The whole filament adjustment (for PA) being done through filament presets on the machine (or to the machine through Bambu studio) is clunky. Do it on the filament profile in the slicer. This whole feature is just an added complexity for the sake of complexity and the user experience would improve not having to remember to set it on the machine when changing filaments.
- The constant baby sitting of “this might not work”, “use bambu filaments for best experience”, and even the “you might have quality issues with Bambu PETG Basic in aux…” is useful for sure, the first time, maybe even the second. But I get it, I also know what I’m doing and happy to deal with the problems created by my unsavoury to Bambu decisions. Let me “Don’t Show Again” this warning and move on from it. Not a big issue but a bit of a QoL negative in my opinion.
A lot of people seem to be complaining about general print quality and I think that’s something that will get ironed out (no pun intended) with time. Comparisons to the X1C are pointless in my opinion because that’s a machine that’s had at least 30 firmware updates in the 4+ years that it’s been around. OFCOURSE it will print better than something that’s just hit the mass market. No matter how much Bambu polish the firmware internally, users will always find ways to break it, make it work worse than the “ideal testing conditions” and add another bunch of variables that they could not have conceived being relevant. It will get better as time goes on, they will collect real world data and update their firmware. Hopefully the hand holding will also go away but…they’re catering to a specific market.
Bambu Studio
This is where I want to say the UX is just not good. It might be nit-picky but I think it could be executed better. This section is going to be shorter because it’s potentially a lot more subjective.
Having 4 tabs for each nozzle type and each extruder, within the already huge ocean of settings, to change variables that are perfectly fine being global is just not good UX. The worst part is that the tabs are locked on the top of the screen and are relatively small so you might spend a few minutes working on some settings only to realise that you’ve been changing them onto the wrong extruder/nozzle.
Making some variables global would be a good simplification. Temperature can be global for all extruders by default and then having a separate temperature for the aux nozzle under that would simplify set-up. Max volumetric speed can also have an extra “aux max volumetric” option under the default one. Having all the settings again in tabs just makes it clunky.
Hide the high flow tab unless you have a high flow hotend chosen under the machine config, and then just automatically adjust the tab to display the right nozzle type when it’s needed.
Again, as before, not being able to adjust PA under the filament settings also feels like a step back.
Conclusion
Overall, when this machine works, and for me it has, with some tinkering, it’s a damned solid piece of engineering. Well-made, feature-dense for the price, and the dumbed down Bambu “press print and printer goes brrrr” experience is there. The firmware will mature, profiles will improve, and most of my gripes will likely shrink with time. My skepticism about BambuStudio’s UX getting meaningfully better remains however.
That said: can I recommend this to a beginner? No. Get an A1/A2 or a P2S* if you want the Bambu ecosystem, or build a Voron/Rolohaun/Prusa if you don’t. You don’t need a chamber heater, aux nozzle, or filtration system while you’re still figuring out why your first layer isn’t sticking.
The X2D is a machine for people who already know what they want from it. If you have a specific need for dissimilar supports, 2 color printing, want to run “engineering-grade” materials, and aren’t going to panic when something needs adjusting, it earns its price tag. If you’re buying it because the spec sheet looks impressive, you’ll probably be frustrated within a week.
*Yes the price difference is small but the learning curve feels like a decent jump with more variables to consider and more things that can go wrong. You’re better off spending that ~€100 difference on filament.
Prints off the X2D
Poke Ball Model by Torikami, originally remixed from MrFozzie
Printed with Bambulabs PLA Basic and Elegoo PETG HF for dissimilar supports.
Uno case by Happy Day Fun Merch, originally remixed from Hugo
Printed with Bambulabs PLS basic.
Gridfinity Storage box by Pred
Printed in Nobufil Black ABSx
Nobufil ABSx Benchy with Elegoo PETG HF Supports
This print failed interestingly. It was a bad idea to be printing the supports FULLY in PETG in a heated enclosure and it basically clogged simultaneously in the hot end and in the aux extruder.
Nobufil ABSx Junction box with Elegoo PETG HF Support interface layers
Spiderman Helmet by javiles
Printed with Bambulab PLA Basic with Elegoo PETG HF Dissimilar supports.
