The Best Way to Reduce Filament Waste for Bambu Lab AMS
Вставка
- Опубліковано 31 тра 2024
- Turn your dream 3D Printing project into a reality at PCBWay:
pcbway.com/g/s49dwT
Check Leon Fisher Skipper’s Gcode and reduce your poop here: makerworld.com/en/models/91241
Umbreon model here: www.printables.com/model/1811...
Intro: 00:00
Sponsor Spot: 01:21
How it works: 02:00
How to do it: 05:53
Outro: 08:22 - Наука та технологія
"This Feature got implemented for P1 and X1 in BambuStudio Beta 1.9. "
This is per the creators linked page.
That’s incredible. I can’t believe they got it added in! I pinned this comment :)
how do i access it?
@@PunkRockBenWilsonthe betas are usually closed.
finally they added it i hope they give the original creator some kind of compensation as he fixed a huge issue in their ams
@@mavicboy Bambulab copied and pasted from all open source slicers and 3DP knowledge from the community…. I don’t think they will compensate anyone 😅
The primary reason why it cuts where it does is that pulling it up increases the chance of a jam. It's not necessarily a huge increase, but given the focus of he printer is reliability first, it makes sense why they did this.
Yeah I’ve mentioned that in a lot of other comments. I understand why they did it the way they did it. The creator actually told me BL is testing this method on their internal print farm now so it’s worth something :)
I was watching another video where someone changed to .5 flushing volumes with no issues, even without these additional steps. but I am still going to test this also!
I came here because of a UA-cam comment on the profile, and find your 1 day old video on it! What an amazing production man, very impressive and I hope bambu labs is able to test this and see if it's good enough as default
This is a great find! Thanks for sharing 🙂
great vid! going to pull the trigger on a x1c soon , doing alot of reading and watching videos so i can self educate myself into the world of 3d printing. being an auto tech i love building things so this is going to be a great hobby!
Hold off on getting the new x1c if you can wait 2 months or so. A newer X1 model is coming out this year so it could potentially be bigger and better.
Older gens will be cheaper.
Saw this a few days back and have P1S on order. Stoked to test this :)
That's awesome, you'll love the P1S!
I’m hoping that Bambu incorporates this. I’ve been using it a few weeks, but managing the changes to the profiles as updates come out is annoying.
Or Bambu/Orca slicer need nicer ways to handle g code deltas.
This has been SO annoying for me.
Maybe they could adjust the Gcode to use the leftover filament in the hotend as infill where the color doesnt matter further reducing the waste
You can easily change your purge volume to 0.5, works perfect
Can I add the geode before I add an ams ? Or will it work with or without an ams ?
Thank you so much i am saving over 50% more filiment
Thank you for your contribution to make X1 Multi Colour printing a breeze
Great job man!
What was the flushing volume that you use in the regular print?
How will this work with PA (nylon)? The stuff is tougher, and I notice that when it is pulled back it stretches and even the filament cutter may just squish it rather than cut it. Just completely took apart the extruder to remove some stuck PA and the stuff pulled back by the AMS just stretch after the cutter squished it sideways. Ran it with .6 nozzle and a second time it gets stuck in the extruder during the cutting and AMS pullback. Any ideas.....anybody? I do have a Kickstarter unit, so I may have missed some fix to this.
Thanks for the video, its an interesting option. I'm new to 3d printing, but from what I can see, by pulling back so much filament prior to cutting, you're opening the door clogs.
It’s possible yes, but after 465 for me it didn’t happen. You could always adjust it a mm or two if needed. It’s worth a shot though
Anyway to have it just print the cut part into an Object & not in the poop shoot at all?
Thanks for this
is it also saving time then basically right?
thanks for this, helped immediately! loading the profile directly wouldnt work, i did have to manually go through the gcode and add in the needed lines. thanks for the video =D
Hi learning a lot from your videos thank you do you make a video to show? How do you make a Bolt and nut in another words, how do you make treads and I can fit the screw into it thank you
thats nice. the only reason im not printing multicolors is that i do not like the waste. hope the next thing is that we print a little longer with the filament (pretty sure 1-2cm print is still safe), so we swap about 2 cm before the actual color change.
heads up to the guy that figured this out, thats already awesome :D
The only issue is that the print would need to abruptly stop where ever that 1-2cm is and it might be in infill or a wall. If the slicer had a way to account for that then I could see it but right now it’s just not possible :/
If only I could do 2 thumbs up. You're doing the Lords work. Fingers crossed this works ! =D
I figured this out a while ago when doing a print with 2000+ filament swaps, and it actually printed for days with no failure. The problem is that as soon as you have any contamination above the heat break (any cold filament stuck to the cooling zone), the nozzle is wasted. In theory this method is not ideal, but I simply haven't had any issue with it so far.
Great video! Thanks! Happy that this method is spread 😊
What surprised me though, was that you did not demonstrate with black and white filament. The critical filament changes are dark to white, and there is whether you clearly see how successful this is (or not).
I'm a believer though! I will try 👍☺️
I didn’t show it because I didn’t want to specifically cover tuning your flushing ratios, but rather talk about the filament pull back technique that Leon Fisher Skipper presents and let you tune it all up yourselves!
@@ButterPocketsu could do that while printing the same model with mostly white and black eyes and rings, this just shows that the method doesnt work well for everything, maybe it isnt, i dont have a bambu, but no reason to not show a better print.
@@tired5925 obviously the method doesn’t save filament in EVERY scenario. The point of this video, and most of my videos, is to show a technique that you can use and apply to your own prints. It doesn’t mean “follow exactly what I’m doing and it will always work”. It’s for you to try and experiment with yourself!
when I download the file mine doesn't say save filament instead it only shows reduce purge. Is that the same thing or did I somehow do something wrong?
Just to be clear, you changed the Prime Tower volume from 45mm cubed to 5mm cubed? If so how did you come to this number?
I just randomly picked it honestly. It doesn’t need to prime all that much. It just happened to work so I stuck with it.
Excellent video. Thank you for sharing. I just bought an A1 with the AMS light and I’m looking forward to saving filament with these g code changes.
So your prime tower segestion its prime volume set to 5 mm correct insdead of the default 45mm?
That’s just what I did. Less of a suggestion on the volume and more of a suggestion to mess around with it yourself!
Did you also use the option to use the „mixed material“ for infill structure?
I did, it’s called “flush into infill”. I did it for both so it didn’t contribute to the savings. It’s something I’d recommend to have on for most multicolor prints.
@@ButterPockets didn’t know how to call in English (not my native language) but learned it some time ago.
i may have missed something, but after i changed to this profile my print went from 12 hours to 20?
I have tried flush to object instead of the purge tower and now Im going to try to purge to stacked objects, of course this is for parts whose colour is no issue.
It’s a prime tower not a purge tower. I really recommend keeping it because it ensures the pressure in the nozzle is back to where it needs to be.
A couple things:
You have to cut the filament. In the infographic at 2:51 it shows the “blob” that you don’t want to pull back into the PTFE tube. That means there needs to be some amount of purge or poop before the color swap.
Second, you have to have a prime tower because the pressure in the nozzle needs to equalized before it starts working on your model.
So yes it needs to do both!
They could eliminate a prime tower by starting the next sequence inside a model. This is just something that slicers will eventually handle well.
@@boggisthecatI’d prefer not to have those imperfections in my print at all! Even when I was pause prints I do a prime tower.
@@ButterPockets
You wouldn’t see it, unless you are printing with translucent materials.
I mean since you need to do both you can also finish the priming on the tower too? Reducing poop even further than 0.5, assuming you're using a single plastic type it won't matter if the color change finishes on the tower.
@@SquintyGears that’s possible sure, just hard to calculate. You’d have to intentionally lower your purge volumes and calculate that it would switch over enough during the prime. Worth exploring tho
This is great. I wonder if bambu will adopt this idea so we don't need to have modified gcode or concern ourselves with updates from BL overwriting our customizations.
They are, its in Beta now.
Depends on the colors used, Try white on black or black on yellow
That’s why it still requires tuning on your end.
Has this been implemented for the A1 series? This is pretty cool
Yes it has! His maker world post has it for all the BL printers!
so when its putting the filament back in he has it put in 18mm at 200 and 2mm at 20, is that just to account for the blob thats in there? I guess I don't see why its not just G1 E20 F200
My best guess would be it’s either similar to what BL did by default or he found that slowing down before jamming it all in helped!
Duke Power? What state are you in?
Can you test that mode when you are switching from black to white filament?
I believe fisher skipper has that on his page. You’ll see that a black to white did need more purge than other transitions which is why I mention tuning your flush ratio more! But regardless, you’re always going to save the 50mm3 every change even if the flush was the same so that’s really nice!
So I just picked up an X1C and went to do this, and when I pulled up the flush volume, it's already set to a .3 multiplier. Do I need to do anything special with that, or am I just already in a good spot with it?
Wow! That is very interesting. I was going to change mine to ..25 ... now I will go with .3. A good nod from Bambu Lab I'd say.
@@pjtodd6756This was in orca slicer, so I don't know if it was bambu or orcaslicer setting it that low out of the box
@@FlesHBoX Thank you!
I'm suprised that you don't do variable layer height (or I cant see it in the vid - possible) to reduce the jaggedness of domed objects i.e. top of deer's head. you are a very knowledgeable fella so mthere must be a reason for that?
I didn’t particularly care about the actual quality of the print just that it did a LOT of color changes. I ended up giving these to a friend who likes that Pokemon and he was fine with the quality :)
can u teach how to make a small tower waste
Anyone tried this on A1 Combo? Is it useful?
That’s such a smart way to save filament
What I don't get is why the change is not done a bit earlier and the remaining filament is used for printing. e.g. if there is a colorswap coming up in 1-2 grams, why not already du the cut, change the color, and use the remainder in the nozzle for those 1-2 grams. That way no hot filament would have to be retracted. I'm totally new to 3D printing, so I must be missing something since that would be the most obvious way.
I am wondering this as well. I'm new to printing and currently researching but this seems like an obvious way to handle material excess.
Turning off tangle detection had a big improvement for me. My fault though. Some spools were overheated when drying, causing filament to stick together a little, Apparently enough to make the printer think the filament was tangled.
I did this to myself accidentally too!
Is there a way to save the flushing filament ratio?
It would be saved in the 3mf file when you save it. I don’t think it’s attached to a particular filament or print profile. I hadn’t actually thought of that till now. I’ll test that when I get home
Looks like butter never got home.
How were you able too resume print after a power outage? Thanks
The X1 has it built in. It happened to do it during a color change so it really didn’t lose its spot.
@@ButterPockets that's really cool, interest to see if the A1 has similar functionality, something I'd have to check..
The P1s started back up where it left off during 3 power outages in a row. I was impressed
I don't quite get why the filament is cut anyways?
It has to be. Check out on the maker world post he has a picture of the blob that would pull up into the PTFE tube if it wasn’t cut. In order to make the system automatic and smoothly operate it needs to be cut.
Was it reported to Bambu Studio github issues? What's the issue number? Bambu probably didn't do this to be on safe side and don't potentially pull back half melted filament (and there is ton of materials to be tested) and cause clog.
I fully get that it probably is to be on the safe side and it makes a lot of sense for them to do it that way!
It’s the same reason gpu and cpu manufacturers technically underclock their processors. It’s where they found it most reliable to run but not maxing out the potential.
"how do you save on poop, thats the most important thing right" lol i love the 3d printing community
Does this effect the print times?
Not really
Is there any way to turn off the cutter or the purge function? Just use the purge tower? Or does this system let you turn off the purge tower and just use the purge function? Still bugs me that bambu has its machines do both.
It has to do both. The “poops” are to purge out what is left in the hot end from cutting the filament. In order to retract the filament cleanly into the AMS this really is the best way to do it. Like fisher skipper shows on his maker world page if you just try and pull the filament back without cutting it there will be a blob of filament that will cause clogs and stuck filament in the PTFE tube. So it has to be cut. Second, you absolutely have to have a prime tower. That serves a completely different role in that the pressure in the nozzle needs to be equalized before you start printing. So it shouldn’t bother you that it does both because both are 100% necessary. It’s bothersome that neither are fully optimized. BUT you can optimize them on your own which is nice that it’s not locked down.
Why not cut and switch before getting to the end of a color? Everything is calculated volumetrically. The slicer knows how much filament it will need. Why not cut early, feed in the new color, continue printing the old color timed to that you get to the end before the purge. Then purge a very small amount to get the color changed and continue printing. I don't understand why it wait until a color is finished to cut it. It knows how much filament it needs to finish the color, cut it there.
That's a fix Bambu would have to implement though I think.
So, in order to switch material automatically you need to cut and pull the filament out from the cut. There’s no getting around that part of an automated material switching system. With that established, yes you could write an algorithm into the slicer that would say “you have 50mm3 of filament left in the hotend, stop when there is 40mm3 of this color left, go cut and retract, then go back and keep printing, then go switch colors”. I think the problem with that solution is that because it would be different on every model, every single slicer setting would change it, walls, infill, width, pressure advance, flow ratio, e steps, etc it would be really hard to do. It would first off be hard to calculate in the slicer. You’d also never want it to stop and then restart in the middle of a wall because that would probably lead to an artifact in your print. You’d also still have to have it “poop” some and still do a prime tower because you can’t just transition to a new color without one. It would also add a lot of time because it would have print some part of the layer, go cut, go back to print, stop to change colors again, then go prime. Instead of just one action of cut/switch/prime. So overall I understand what you’re saying, but I don’t think it’s as feasible as it sounds. It would be a lot more unpredictable and like others have commented it seems BL went with the most fool proof way of switching colors to run into the least amount of problems.
@@ButterPockets I hear you and I'm not saying it would eliminate waste and purging. But really it sounds like you just solved most of the things. I'm not saying it would completely eliminate waste. But the slicer is already doing those calculations. It could plan to switch in infill. I do it manually just by cutting filament when I think it's time and pushing a new one in to the extruder behind the old one. It's the same amount of changes, it's just different timing.
When you do G1 E-20 F200, does the AMS also pull the filament back 20mm, or is the extruder just retracting it into the buffer on the back? If it's retracting into the buffer, I think it's possible that the buffer is fully extended at that moment (like AMS had JUST fed more filament in), and there might not be 20mm of space there to retract.
As an aside, on my Vorons I retract 25mm in PRINT_END at a much higher speed (F3600), which then allows me to change filament later without heating up the hotend, because the filament end is completely out of the hot zone.
I can see the filament move up in the AMS. I didn’t specifically check the buffer, but mine is very noisy and it’s clear when it’s operating. I never heard it do anything out of the ordinary.
I wish the software / firmware had a checkbox not to poop a bunch of filament. If I haven't changed the filament since last time then there is no reason to. Maybe it should remember the last filament & the current and intelligently check/uncheck the box while leaving the user the choice to override it.
The reason it does that is because it always retracts the filament. So it can’t do it “smartly” unless you always used BL filament with RFID and it *knew* you didn’t change what was in the AMS slot. Because you can use filament that’s not RFID it can’t know for certain what is in there or not. I get that they could have a switch that says like “same filament, poop less”. BUT there is going to be a disconnect between the cut and the new filament always. That would lead to a small pressure change somewhere in your print. The best option would be “leave the filament in and don’t cut it”. An option like that might be helpful!
I don’t get why they don’t just cut before the color swap needs to happen in the print? Just cut, keep printing the same color until you reach the transition point. Purge, then print the new color. In theory you could manage prints with 0 waste this way, if you can purge enough to infill/inner walls.
Another thing you could do to get 0 waste is to ”purge to into new filament”. That is, use the ”purge into object” and have the object be new strands of filament which you can later use for printing test pieces. You might need a way to attach the strands to each other though if you can’t spare the space on the plate to print a spiral.
I would think the reason they don't do this is because it can't be 100% controlled. Its similar to this method as well that other commenters have pointed out - that the reason BL hasn't implemented this is because its not 100% controlled and *could* lead to a clog or failure. If you know exactly how much filament is left in the hotend you could flush that into infill, but what about when there isn't enough infill to do that? I think a lot of this *could* be implemented into a very smart slicer, but because all these slicers just take from Prusaslicer and Slic3r until someone does something custom (a la Superslicer) it kind of has to be this way.
@@ButterPockets Yeah, this is probably a limitation in Slic3r that makes it difficult… Guess I’ll just have to make my own slicer then ;D Might aswell make it non-planar too, and sprinkle a bit of magic on top.
@@beaconofwierd1883 hey I’ll contribute to that!
Saves filament but, extended my print A LOT.
Does this save you time as well?
I'm a little confused why these printers make towers AND poop.
The poop is because it cuts the filament and retracts it. This is a better option than JUST retracting the filament because the end will have a blob that will clog up the PTFE tube. So the it “poops” (purges) the old filament out. This is why it has to poop at the beginning of a print as well. The tower is a prime tower because the nozzle pressure needs to be equalized before you start printing. If the nozzle pressure isn’t right it will sputter and ruin your print.
so basically, its as easy as downloading the 3mf file and once you choose your printer you save the presets, and change the flush voulume to 0.50 and that should be it? cause i just tested it on a sliced file with the normal settings and then with the less poop settings and after changing the flush volume, my waste literally got cut it half. which is amazing!!!!
Yep! You can fine tune the flushing from there but that’s basically it
0:27
lets talk about some ,,,,,,😂
I did this and somehow my print time increased lol
I just don’t understand how this can be so full proof but not a part of the firmware or default profiles
Because, in full honesty, it probably isn’t 100% fool proof. Nothing really is. But I would bet BL went with whatever in their testing yielded the best results for the most possible cases. This might not work for every printer, every print, every filament, every use case, etc but I think it’s worth a try and can always be adjusted if you run into issues!
That makes sense. They probably decided it was better for people to waste filament than complain to customer support that their prints didn't have clean color changes...Thanks!@@ButterPockets
The only reason BambuLab don’t use this method is it increases the chance of clog as the hot blob pulled back into cool zone of the hotend.
You can fiddle with the amount of retraction to minimise the risk.
I haven’t done this, yet. The purge settings can be dialled *way* down from the automatic results, which saves a huge amount of waste. (From e.g. 420 mm^3 to 120 mm^3. Depends on the colours, of course.)
Yeah I understand that! I like that we are able to try this out tho. It’s worth a shot for sure
not reducing anything for me
So still 2.5 dollars per print in waste.
F200 increases the print time, if you have multiple color changes, is not worth it.
Yea now try doing it with white filament and see if it bleeds.
Congrats, you just introduced one of prusa mmu issue that was fixed with the filament cutter 👏
What do you mean? Can you explain please? 🙏
Oh, I think you are referring to what is explained at 2:51 in that center text there.
According to that, your comment is incorrect
@@oljobo I mean the fact of not cutting filament introduce blob and 'hair' at the end of filament, and it gives many issues. On prusa you have to adjust ramming settings for each brand of filament because of that behavior. The cutter is cleary here to fix that, bambu lab didn't add it for fun, I'm not sure it is smart to bypass it 🤷♂️ my guess is you can have issues depending the filament brand
@@superpatate1234 ok. But this method does not bypass the flush cutting at all. It just does it more smart and precise, without wasting more filament than •necessary•. No blob or hair will be retracted into the PTFE as I understand it. Again.. see text at 2:51
@@oljobo oh ok, I was thinking the gcode modification remove the cutting step, misunderstood, my bad
This wasnt an "issue". Bambu obviously makes money by selling filaments so they were draining users with this waste.
Thanks for the vid
3:31 im not. as a designer you schould know that you design something thats a shit design but works....and some day on the toilet you get a heureka moment.
The best was to reduce „poop“ is not to use that shit.
Yep. You called it. No multi material any more!
Your pricing is way off, why can’t someone just go to Chat GPT and do everything one there own? Most of the people watching are doing it because they don’t have the kind of money your asking for. If your so proud then give it all for free!
What the heck are you yappin about?? I’m not selling anything?
@@ButterPockets sorry I’m not sure what happened I was watching a webinar and then you answered.
Senior moment
Yeah... but still, the amount of poop is still pretty horrible. Bambu needs to optimize the software.
Well thankfully you can fully optimize the flushing volumes yourself. There’s no way Bambu Lab could know what color, brand, material, etc you were going to use so they had to make it unoptimized in order for it to remain repeatable. You could run flushing volume tests and get it to the point where it only flushes exactly what it needs to. It’s going to have to purge and prime for a swap so at that point there would be no unnecessary waste.
Filament purge being called poop is the cringiest thing Ive heard
Well, you heard it here folks, we gotta stop calling it that.
This must be your first day on the internet.
The machine literally spirals it like poop to prevent it being stuck on the nozzle. Every maker doing videos on Bambulab printers call it poop, and have since their first release last year.
So congrats on your first internet day and discovering 3d printing, bambulab, and ButterPockets all in one swoop
The best way to reduce waste with an AMS is not using one!
Ha ha duh!
It's incredible how wasteful it is. I love 3d printing and feel wasteful already with just single color making useful thing. Lots of videos I see are people making stuff one could buy at gamestop like a Luigi, not trying to knock it cause maybe I don't get it but it's just a speculation
@@redherring5532how much of ALL 3D printing is waste? How many millions of benchies and other useless stuff gets printed? It’s inherently a wasteful hobby. If we can reduce it we should, hence this video, but unless we all stopped printing then there’s going to be waste. Amazon throws thousands of pounds of electronics into landfills every month. I care about waste too, but at the scale of a couple grams of plastic filament, it’s really not much.
Both prints look terrible.
If you watch the video that’s not the point and I talked about that :) The point was to make sure that the filament retraction method was reliable.