I just watched a large part of Aurora Techs Review that you linked. What I take away from it:
On 4 color prints, the X2D is a bit better than all single nozzle printers.
It’s no comparison to multi nozzle or multi head printers like H2C or U1, which both produce almost no waste. At 23:51, she even says explicitly that “a dual extruder setup [e.g. X2D] is actually more useful for printing supports than for multi-color printing”.
I really don’t understand how this review motivated you to buy the X2D for multi color printing.
You have printed far more than 4 colors, so the advantage over single nozzle printers shrinks even more. With the 3D printing experience you already have, it should be clear to you that the actual difference in waste depends a lot on number of filaments and the actual distribution of colors.
Many of the reviews I watched compared the X2D against the U1. Many concluded that the X2D is useless when you can buy the U1 for little more. Only a few like CNyc kitchen clarified, that they have very different target use cases and the X2D is clearly aimed at printing engineering materials while U1 excells at classical multi color PLA prints.
In my perception, Bambu Lab have been very careful, even overly conservative not to advertise the X2D as multi color printer. It certainly can be used for that task at least as good as the previous single nozzle printers with AMS. In every presentation video, they pointed out that the aux extruder is not on par with the main. Bambu Lab also warned about reduced quality of the aux nozzle. I think they did all they can to warn people against seeing the X2D as multi color work horse. They even went so far as to show the aux extruder in worse light than it actually is. Have you ever seen any company doing that in their product presentation?
I can’t see how they generated expectations that don’t come to reality.
IDK seems to me that no matter in what orientation some rabbits will be printed, there are multiple issues with X2D and OP is within his full rights to complain about an expensive machine that did not meet his expectations.
I just can’t get over how janky this thing is. It should be called P2pro or some sh.it, most definitely not X series. I would bet that everyone who spent time with both X2d and X1C would agree.
Edit: The author of the thread has made it clear that he doesn’t want any help, but I’ll leave the final result visible in case anyone comes across this thread.
Separate the gun from the hands. That way, you only need to use the blue filament for the second build plate. That way, you can even fit 5 rabbits on it.
If you do that, though, I’d recommend using connectors; I just did this quickly without them. You can either connect them by pushing them together or use chemical welding - there’s a special invisible welding superglue.
If you want to do it in Blender yourself, you can, of course, use CAD, too: It takes one minutes.
Edit: And as a bonus, even if you don’t like the orientation, it’ll still be worth it… Of course, you can also use it for the eyes and nose to separate out the black filament.
I never claimed this printer was only built for multicolor prints. Or that multicolor prints is the only reason I bought this particular printer.
As I said in my post:
I originally intended to buy a P2S, but when the X2D dropped, it was enticing. It was only $100 more and featured many more sensors to reduce mid-print failures, a crucial factor given the duration and complexity of my jobs.
The difference between X1C and P1S was mostly improved sensors and ability to print more complex materials. So even if X2D was a single nozzle printer I likely would have bought it just for the enhanced ability to reduce/detect mid-print failures. I have no complaints here.
However, if you looks in both the Aurora tech review (and other reviews that address this) - the claim is significantly improved waste and print speeds. Look at the screenshots I took from the video:
X2D Compared to single nozzle printers:
vs. Centauri Carbon 2: 80% less purge waste, 25% faster print (6 hours saved)
vs. K2 Pro: 80% less purge waste, 11% faster print (2 hours saved)
vs. SparkX i7: 75% less purge waste, 16% faster print (3 hours saved)
How is that not significantly better results than a single nozzle printer?
I agree. U1 is indeed a less wasteful printer in multicolor prints. It does has 1 critical disadvantage (from my use-case/perspective) - they are limited in the number of colors.
I now have a 3-AMS system, that can print with up-to 12 colors. Even the rabbits example I posted has 10 different colors, and cannot be printed with less than 3 AMS units. So while U1 (and other alternatives) do have they merit, they still lack the ability to print with the number of colors I desire.
Even the H2S Vortek can only print with up to 7 colors without swapping. More than it - and it too will need to swap filaments and create poop.
Sure, but one of your main complaints is too much waste in your real life multi color prints.
I didn’t see any promises made by Aurora Tech. What I see, are a bunch of test prints she selected and presented. None of the tests she made are close to the number of colors you are printing. And I nowhere see 50% time improvement either.
To me it was very clear that at even higher numbers of colors, the savings in time and waste would shrink rapidly. It also was clear to me, that actual waste highly depends on the model and the color distribution, e.g. an 8 color model where most layers only contain 2 colors benefits much more from the second nozzle than 5 color print with all colors on every layer. That becomes very clear in the kitty print where white and pink are the predominant colors and there are just a few spots in more colors. Naturally this benefits a lot from the second nozzle.
I’m very sorry for you that the results she showed don’t transfer 1:1 to the number of colors you are using, but this really isn’t her fault. And as I pointed out, she even explicitly mentioned that it isn’t the best printer for multi color printing.
Apparently I would have drawn completely different conclusions from her test than you did.
And it certainly isn’t Bambu Labs fault what some independent YouTubers put into their review.
I fully understand that you are frustrated. But I don’t think this is anybody else’s fault as you put it.
We’ve spent most time discussing this. But I wouldn’t say this is my main complaint about the printer.
And I wouldn’t get BambuLab off the hook that easily. True, the claims were made by Youtubers, not Bambu itself. But that’s only half the picture. I was watching the videos on the machine’s release - these Youtubers didn’t get their printers at best buy or order them from Bambu.
All of them received advanced prereview copies of the printers weeks in advance (even before they were announced), in order to have time to print and record their review videos. It’s all made in cooperation with Bambu.
And it’s not some random Youtuber with 100 subscribers. These are some of the biggest 3d printing review channels, with hundreds of thousands of subscribers each.
The point I want to carry over is: I can’t see any of the claims that you see in at least the review by Aurora Tech. What I see: Some prints benefit considerably, others do less. Your prints seem to contain far more colors than the most complex ones she tested. I can’t see any claims at all in her review. If at all, it is the explicit quote I gave above, that the printer is not ideal for multicolor. Nowhere is said that those time and waste numbers carry over to prints with far more colors.
Before purchasing, out of curiosity, I downloaded the latest version of Bambu Studio, installed an X2D and sliced some files to see the time benefits, although that was not my main incentive (it was: no more contamination between different materials, heated chamber, the finer belts, better cooling & overhangs). That would have been a great option for you too, as the tests didn’t cover the type of prints that you apparently want to make.
asking again: Where did read or hear that promise? I neither see it from BambuLab, nor the Youtubers.
It was the one complaint taking the most space But here you go:
I haven’t printed bigger objects or supports yet with the aux extruder, so I kept my mouth shut about failures further up as you reported. Well possible, that you have a point. I could also imagine that your unit has a defect? Have you contacted support about it? Did you calibrate flow & flow dynamics for the used filaments on the aux extruder? Those are independent from the main extruder and not automatic. Maybe flow calibration solves your adhesion issues? Otherwise, I would contact BBL to look into that. I think, CNC kitchen also printed some bigger parts with the aux extruder without any issues. That’s why I believe more in a unit problem than a fundamentally flawed design.
As a test for the aux extruder, I printed two different tiny dual color versions of the famous tree frog with 0,2mm nozzles at 0,1mm layer height. I didn’t expect much, but the aux extruder surprised me with far better quality than I expected. The tree frog contains tiny islands of the secondary color (< 0,4 diameter), that came out far better than a bowden extruder has any right of.
Apart from that, I have mainly used the aux extruder for PLA support interfaces for PETG prints, which the aux extruder did inconspicuously. I still have problems sometimes that PETG doesn’t adhere sufficiently to the PLA interface. I think that is mainly due to cooling and too high speed of the first layer on top of the interface. Little time at the moment to investigate that more thoroughly or contact Bambu about. I have already seen some solutions posted by users. That is something that BBL should definitely improve.
I still have a replacement of the left heating assembly in front of me. Print came loose and caused an enormous blob of death. I read not only from you, that disassembly is no fun compared to the old printers. So yes, BBL probably didn’t make that any better. I still take the print head as an improvement over the gen 1 printers with the easy to to swap nozzles. Unplugging and replugging those connectors was really inadequate. I’m very happy about the nozzles and accept the more complex disassembly of the head if it is necessary for the new nozzles.
I don’t agree. Many use the X2D without it. I have installed a manual switch: The „Four In Two“ PTFE Adapter - Free 3D Print Model - MakerWorld
That serves me much better than the FTS could. The FTS is really a terrible product in my opinion that BambuLab should revise. It needs more than 2 inputs, so that it can switch at least 4 AMS individually instead of in two groups. And they should improve the resistance that many users suffer from.
The golden solution would be a new AMS, that includes a 4-in-2-out FTS. That would allow to assign each filament swap individually to a nozzle.
Actually the FTS increases filament swap times in many cases. Especially if you have to print filament from one AMS on both nozzles, that also creates some extra waste. Maybe that explains some of your disappointing print times and waste?
You have to configure once how your AMS are connected to the Filament Track Switch / the printer. That can also be done automatically by the printer. It will feed one filament from each AMS and look out, in which extruder it appears
After that, the printer / bambu studio decides how it has to guide the filament to the nozzles.
You just select, which filament should be printed in which nozzle. Alternatively you let Bambu Studio examine the best assignment and it will actuate the switch accordingly.
That might be with some penalties in certain cases. Or in other words: You still might save some time and waste by rearranging the rolls in your AMS. That is, whenever the switch has to change the assignment of AMS to nozzle within the print, you loose some time and have a bit of extra waste compared to a static assignment.
That is, why I choose the manual switch, which allows me to assign each AMS statically to one nozzle. I don’t need dynamic switching. And the FTS can only assign two groups of AMS to each nozzle. The manual switch allows assigning 4 AMS completely independently.
So if two AMS are on the same FTS input but should be connected to different nozzles, for each filament swap between those two AMS, the printer still has to retract filament from one AMS, switch and then load the other nozzle from the second AMS.
X2D compared to X1C: almost 40% less purge waste, More than 40% time economy.
X2D compared to A1: More than 50% less purge waste.
X2D Compared to single nozzle printers:
vs. Centauri Carbon 2: 80% less purge waste, 25% faster print (6 hours saved)
vs. K2 Pro: 80% less purge waste, 11% faster print (2 hours saved)
vs. SparkX i7: 75% less purge waste, 16% faster print (3 hours saved)
How is that not significantly better results than a single nozzle printer?
EDIT: Funny enough I just found this in the official Bambu documentation of the FTS:
Dual-extrusion 3D printers significantly accelerate multi-color printing by reducing the flushing time required by single-extrusion machines. Traditionally, ensuring filaments reached the correct nozzle required manually swapping spools between AMS units. With the Filament Track Switch, those days are over. It intelligently routes any filament from your AMS units to whichever nozzle the slicer assigns, giving you total hardware flexibility and faster prints.
I actually did a similar test shortly after ordering the printer. And the results were not good. On a 2-color print the X2D marginally less purge than P1S, while showing slight increase in print time. So in real world use cases, depending on the specific print, even 2-color prints are no better than 1 single nozzle printer.
That might be. I have not done extensive testing on the issue. At least not recently.
And checking my printer I see it already worked 1,200 hours - so maybe it’s due for a re-calibration. (not sure how many print-hours ago I last calibrated it)
That’s a very interesting solution. And free to boot. I wasn’t familiar with it. (hopefully Bambu won’t decide to remove it because it competes with a product they sell)
The FTS does have one advantage over it - it’s controlled electronically. Which means the sliced file can use the same filament on different nozzle, in accordance to need. One of the features of the Bambu Slicer is the “Filament-saving mode” (it’s the default mode when slicing). Which tried to minimize waste by routing the colors to the relevant nozzle.
Without the FTS, it can only assign specific colors to specific nozzles, and you need to manually place your filament in the correct AMS.
But when you have FTS, it can assign the same color to different nozzles at different times, thus supposedly reducing filament purge.
The filament track switch unbinds the AMS from specific nozzles, allowing any filament to feed into any nozzle regardless of configuration. This enables dynamic filament mapping, permitting mid-print color reassignments that significantly boost multi-color efficiency.
The switch is controlled electronically by the printer, thus the slicer can assign any filament to any nozzle during print time. Regardless of the actual AMS the filament resides on.
And it can even recommend moving the filaments to different AMS units, to optimize/minimize filament swaps. For example:
What’s the point of keeping this thread alive if OP already shown that they have zero to none interest in getting help? Seems like a bunch of people suggested things for OP just to say “nooooooooop, you are wrong”.
It’s a feature of the filament switch. It’s printer independent; it works the same on the X2D as it does the H2D.
Having two nozzles will reduce waste, but that only applies when using two filaments. As the number of filaments used for a given print increases, that advantage diminishes, I believe exponentially.
For me, filament waste is a tiny, tiny fraction of the overall cost of printing when you factor in the costs of the printer, its accessories, and the electricity used to run it. The same goes for print time. My printer spends most of the day in the off position. If I was operating a print farm and pinching every penny it would be a different matter.
However, I returned my switch because it just added complexity to an already overly complex back end of the X2D and its constraints to routing PTFE tubing to manage friction. I just mount the rolls I want for each nozzle in the appropriate AMS and print. Manually swapping filaments around takes only a minute.
Not quite. The filament switch is a feature of dual-nozzle Bambu printers (X2D & H2D). If it was a printer-agnostic feature - you would be able to install the FTS on any Bambu printer.
I guess it depends on your printing habbits. I personally mostly print video game character (which are multicolor non-functional prints) - so for me waste constitutes a large portion of where my filament goes. This obviously affects costs. But also the need to constantly buy new filaments (as filaments keep running out), and the need to store large quantities of said filaments at home. And also the print times (the more time spent purging, the more overall time the print takes).
If someone for example only prints functional parts - purge waste would be neglible for them, and they wouldn’t care about it much. As I, for example, don’t care about max print volume - as I never reach it in my prints.
I completely agree about the complexity. FTS make the already complex X2D even more complex. And adds even more place for failure. However, for me personally, FTS constitutes a huge QOL improvement. Both in the convince of not needing to swap filament rolls, and the reduced purge waste.
I guess what I meant was the filament switch is the module that swaps filaments, not the printer itself, and therefore the “feature”. Just a matter of semantics.
The filament switch should have been included in the X2D buffer as a single unit. That would have reduced a lot of external plumbing and friction.
It is absolutely, but it just are individual examples, not a general rule. Actual waste and time heavily depend on the individual parts, including number of colors, distribution over layers, size of the parts, etc.
Eg, a color change always takes the same time and purge amount per layer, no matter how much part you then print with that color.
So a print with 4 colors on each layer and 100 layers will always create the same amount of purge, no matter if you print a very thin tower or a huge bulky part filling the build plate (assuming, both are 100 layers high).
That is true for every fdm printer, be it single nozzle, dual nozzle or multi nozzle.
Therefore, the absolute saving when going from a single nozzle to a dual nozzle will be always be the same.
If you print a very slim part, that percentage of waste compared to the part will improve dramatically. If you print a very big part, that same saving is just a small improvement of the total filament used.
For print time, it behaves similarly. If the single nozzle printer is busy most of the time with printing the part, then improving filament swap times doesn’t help much to speed up the print. If printing time is mainly governed by filament swaps, then prints can be dramatically faster with more nozzles. As soon as nr of filaments exceeds nr of nozzles, only very few swaps become faster and most take the same time as on the single nozzle printer, so time saved becomes marginal. Reducing
All that is general mathematics, that does not only apply to the X2D but to all fdm printers.
A very striking example are those tiny frogs I posted above: When adding more frogs ( 4 instead of 1), both filament usage and print time stayed almost identical. One could say that I saved 75% waste and time per frog by quadrupling the frogs. I don’t find that view very useful. It was still the same amount of waste.
I remember when the X1C was released, most youtubers mocked about the amount of poop and compared weight of poop to weight of the actual parts. Their parts though were individual figurines in like 6 colors. They could have printed 20 more figurines on the same plate and total amount of poop would have stayed exactly the same. The ratio then would have been oh so much better. Already then I found those reviews very deceptive. Why would anybody print the way they did? Even with an INDX, I would probably maximise the nr of parts per plate.
But here you go:
