Let me first say that I am a biologist by education & experience (a plant ecologist to be exact), not a hardware designer, engineer or general egghead. So if the following idea is off the charts and downright silly, be gentle in telling me so.
When printing multiple colors, the time spent changing filaments is quite long, as I think most/all of us would agree. And that time is due to the fact that the AMS is pulling the filament all the way back to the spool, then pushing the next filament all the way out to the hotend. The AMS has a mechanism which feeds filaments from the four different spools into a single bowden tube - a “funnel” if you will. Why is that mechanism located inside the AMS and not at the back of the printer right where the single tube goes into the printer to feed the hotend? If that mechanism was located near the hotend, the AMS would only have to pull the filament a few cm from hotend to basically the mouth of that “funnel”, then feed the next filament a few cm from mouth of the “funnel” to the hotend. 3 extra bowden tubes from the spools to that “funnel” would be the cost (more or less). Whatever the cost, it would be comparatively minor relative to the cost of the printer, and I think the increase in print job speeds would be well worth it for many users.
Think about how much time that would save for a print job utilizing four different filaments and, say, 100 filament changes.
You are describing a similar system to the A1, or closer the CC or AD5X.
The short reason, in my understanding, is that it cuts down on expanability and having it all included in the AMS is cleaner/simpler).
It’s not as big of a limitation in the whole picture. Filament feeds very quickly. I much rather see them improve the changing cycle by using stronger heaters and fans.
The cynical will say it’s to sell you more dongles And you still wont even be happy cuz it does not solve speed problems.
Check out how the kobra x does it - the toolhead basically has the hub portion of the AMS built into it. It holds all 4 filaments, dispenses the correct one when needed, and thus only has to move filament 1cm instead of 1 meter. They also optimized for the cutter to be as low as possible so that waste [time AND material] is minimized for a single-nozzle printer.
Because not everyone has an AMS? The basic functions must be provided even without AMS.
In fact, the AMS doesn’t retract the filament all the way back to the spool, but only as far as Bambu considers appropriate. So that it works on as many devices and setups as possible.
The time taken to rewind the filament is put to good use, as the nozzle is cleaned at the same time. Therefore, getting the filament to the nozzle faster is not helpful, as cleaning is the limiting factor.
It’s like a traffic jam at a tunnel: Even if the cars reach the tunnel faster, they don’t get through any faster. Whether the car is going fast or slow, drives 10 meters or 1 meter, doesn’t matter - as long as the tunnel isn’t cleaned.
Otherwise, I agree with what Bullocks and CincinnatiDesigns said; I don’t need to repeat that.
There are many different iterations of how filament can be fed to the extruder, but one big issue is keeping many filaments all fed right to the extruder makes it extremely likely that the inactive filaments will break over time - hence the need to retract at least outside of the printer housing itself.
There’s also a need to keep moving mass as absolutely low as possible.
I would say that overall Bambu’s solutions are the best “on average” across all use cases.
This is a very good point. AMS units are a safe haven for filament from humidity and help keep filament dry if properly maintained. The farther the filament is pushed out of that dry environment, the more water it can pick up and get brittle or have other water issues. Some unprotected filament will get purged out during filament changes but could possibly cause problems there too.
As Carbon notes, less used filament may pick up enough water to become brittle and fracture.
It could work well for some that print a lot where filaments don’t go long unused, but others might get blindsided and wonder why their filaments don’t feed well or cause problems. It’s those somewhat “hidden” gotchas that cause problems for folks who might not be aware of the issues.
That, and if there were 4 then someone who wants 5 filaments would ask why stopping at an arbitrary number of 4 and not 5
Aother source of significant waste of time (and materials) is the need for the prime tower. BL CEO did mention that at the Formnext last Nov. Not sure if he was just talking about the well-recognized pain point or if it was an indication that BL had been working on something to eliminate the need for the prime tower. When Vortek was introduced, they did say it was their first attempt at reducing waste.
It’s simple really, this is the State of the Art as it is today. But in the future, it may change to where multiple filaments can be extruded and ejected onto what is being printed.
Now, to print a field of wildflowers, color changes have to happen as layer progress. But imagine if there were 20 nozzles all hot and feeding different colors, our printers could do in a single fast pass what takes hours now.
Hold on, the future is coming like a runaway train… My best solution is to get multiple AMS’ and let them do their thing, however frustrating that may seem. And multi-headed printers that can spray a ton of filament in one pass.
Meanwhile, we’ll just have to wait. But I like how you are thinking “Outside the Box”.
Sometimes when I have a long print, or a very involved print, I will get it going then go to bed and let it run overnight. I wake to a finished part ready to remove from plate and enjoy. Often, I will use the Silent Mode at night, so I don’t wake up to a cranky wife because the “Damned Printer kept her awake”.
A good chunk of time is not so much reeling the filaments but in a single nozzle printer you have to purge color #1 out of the nozzle and then feed color #2or#3, or more each time. Some manufacturers including Bambu have varying numbers of nozzles in an attempt to resolve that. Certainly a personal preference item but I try to stay away from a model that has color changes within the model for that reasonand save multiple colors to when I have a group of models that combine as an assembly
I printed a multicolored figure overnight once. The next day I had a stopped machine and warning my poop chute was blocked.
I have a large poop can on my P2S, and it had filled to the point the filament change waste had backed up into my printer. Nice that the printer saved itself, but an eye opener about color changes in a print.
I, too, prefer prints that can be done and assembled. Cuts down on a great deal of waste.
And you still wont even be happy cuz it does not solve speed problems.
