Great video. This little beast still blows my mind. From my years of printers at 500 acceleration and 50-60mm/s print time it is taking me some time to push them haha. Perhaps this will be the year I won’t take it so east on the printers. Looking forward to seeing your next version 😊
For the ones whose hearing Uzy for the first time; it has been more than a month I've been using. Never had any issues; had perfect prints so far. Give it a chance and you will never regret👍🏻👍🏻👍🏻
I honestly hope you analyze reworking the frame, even if it's just to add an extrusion in the front, FailFast has found that doing that alone helps a lot. My other take would be to go fully unsupported rail but really stiff AB joints, that would maybe prevent torsion better.
@@ferdinandhenkel4567 I agree, said it mostly as a joke. I do however think that maybe also adding more extrusions, at mid sections could help as well.
Seeing my printer of choice the FLSUN v400 as a runner up under this conditions is kind of inspiring. And so far it's been behaving flawlessly. Now I need to tune it properly to have it keep working not this fast but consistently and reliably fast. And I'm not is such a hurry, after all. Also it's comparably much less noisy than any of this.
Tolle Arbeit und es ist echt Klasse, dass es so Leute wie dich gibt, die die Grenzen der aktuellen Technik weiter ausloten. Ich denke in dem Design ist noch viel Potenzial, wie eine Carbon X-Achse oder CNC gefräste X-Achse (siehe VZ-Bot), hochsetzen (bzw. Mittig der X-Achse ausrichten) und abstützen des Hotends um die Hebelwirkung und Schwingungen weiter zu reduzieren, Versteifung des Rahmens mit einer vorderen Querstrebe (abnehmbar/Klappbar), Wandbefestigung des Druckers, große Radiallüfter statt der pabst Lüfter für mehr Druck bei niedrigerer Lärmbelästigung oder Ansaugschalldämpfer für die Pabst Lüfter, Abstützung oder Führung des Druckbetts an der Vorderseite, Doppelter Bowden Extruder für mehr Kontaktfläche. Es gibt noch andere Themen aber dann wird es schon recht komplex. Viel Erfolg noch :)
Best video yet. Personally printing fast benchie shaped objects isn't for me (but congrats on that silly fast time) so this one digging in to the how to take that and apply to real world printing is more up my street.
I'm convinced the next way to increase printing speed is to move both the nozzle and bed in opposing directions (dual corexy). Ideally the ratio of travel of both the nozzle and bed would be such that their momentums cancel out (i.e. if the bed was 2x the mass of the nozzle head then the bed would travel half the distance and thus the speed in the opposite direction of the nozzle head) and ideally the nozzle weight and bed weight would be the same (assuming they're already as light as possible). For the same effective printing speed you could halve the acceleration/speed the motors need to create while also cancelling out vibrations from the printer throwing itself around (equal and opposing momentum would meant the rig is stationary). For the same acceleration per motor you would double the speed. Just hope for good bed adhesion if the ratios work out that your bed needs to move more.
Wtf. Just put the printer on some bearings and mount it with soft rubber bands on all corners. Then it will do just that mass balanced moves and return to centre at moderate speed acceleration such as not to run away. But you don't want to introduce more movement axis than necessary into the actual mechanism since they'll all add tolerance play and distortion and these will all add up.
@@SianaGearz I feel like it would be difficult to prevent that from sometimes causing resonant oscillation given the highly dynamic nature of 3D printer acceleration and forces.
Nicely done mate, nice video. its hard these days to find proper scientific youtube channels that are not that sensationalised or drama driven just for views.
@@247printing ido t know if a annex engineering k3 would be faster tan the vzBot or a v0.2, since its built like a TANK , althoug its not the lightest printhead 😮 and its quite unknown bist du österreicher ?? der abspann verrät einiges 😂
From an enthusiast standpoint, this is so amazing, I love that the fans alone draw 150W ;D and you need hearing protection! But it must be so much FUN to design, test, and push beyond the limits :)
Thanks a lot and oh yes: It was a lot of fun work in 2021 - unfortunately I didn't have as much time as I wanted for the stuff, but it works anyways :-)
At a certain point the air pushed on the part will be the limiting factor ^^ , imagine you need so much cooling power your extruded line gets pushed away befor it cools and attach to the previous line Xp
With so much cooling and such printing speed, layer adhesion surely suffers. I'd love to see strength test comparisons between fast and slow prints! Keep it up 💪💪💪
It doesn't. If you have poor layer adhesion from speed/cooling it means your temperature is too low - either the base temp or from cooling hitting the nozzle and making a big temp gradient between it and the thermistor. Better shields/socks that insulate the nozzle almost all the way to the tip will solve the latter.
@@daliasprints9798It does matter. The faster you print, the worse is layer adhesion. CNC Kitchen did make a video about explaining how cooling affects it. Since printing faster requires more cooling, guess you can only print decorative toys at those speeds.
@@alejandroperez5368 You can keep citing the results of someone who's smart but knows virtually nothing about fast printing, or ask people who actually know this stuff. It's just not his area of expertise or interest.
From a quick search online some tests show better layer adhesion with speed and some show the opposite. Seems like as with most things this matter is situation dependent (your specific printer/filament/settings)... Another good reason to build a test rig and start doing my own research! 😉
now imagine you had two of these, one upside down, printing onto the same base from opposite sides to halve the print time, maybe with a thin filament sheet as a base plate to allow for a cohesive print.
Have you considered placing the hotend fan on an external frame above the printer and piping the air to the nozzle through some form of lightweight tube? It would get rid of a few extra grams and also allow you to use a more powerful fan. As for the tube: There might be lightweight fabrics available on the market that can be used to sew an airduct.
Brother, your channel is so legit. I'm really glad you started it. I have learned so much from you in a few videos, so keep spittin that fire! ❤🔥❤🔥🔥🔥
Just some thoughts on your upgrades. For the air flow, you could use some pneumatic tubes coming in the top like you're wires. You can blow air in it to get the airflow, with less power and noise, and about the same weight. Triangle Labs makes the Rapido Hot end which they say will go up to 75mm3/s. This seems like BS, but if it's not, there you go. You can make the frame more stiff with carbon fiber parts, they make race cars that can do similar things to what your looking for all the time, or you could use a very interesting material: Boron fiber. It's stronger, stiffer, and cheaper, but more delicate than CF. Boron Fiber is twice the stiffness of steel and light weight too. It would be interesting to see. Either way I think you'll get the stiffness you want.
Very cool, you should check out the solid state cooling they've showed off at CES, literally no moving parts , would bring down printing head weight a ton and with its super fast jetstream of air offer way better cooling, which could I bet almost double printing speeds of todays printers while maintaing the same quality. Just guessing but these coolers seem next gen ,and like every video suggests, cooling is still a huge obstacle for printing speed, its the main one in computing speed, hence the motivation for these coolers , they can apparently give 15-25w laptop cpus another 5-10w of heard room which is pretty remarkable, doing with less power and less form factor too. Someone needs to put one of these on an Ender asap, Frore Systems is the company and i seen the video from PCWorld's channel covering CES from a few days ago. It looks like a flat lithium battery that sucks air in from one side and blow it at high speeds out the other. Many "vibrating membranes" at supersonic speeds are responsible for this when voltage is applied is basically how it works.
I have an idea for you. Instead of reducing your mass - triple it. But triple it with counterweights you accelerate in the opposite direction of the tool head to cancel out the X and Y forces.
It's like he's trying to R&D the ultimate FDM printer. Hope that when he finally get's it all sorted out that he provides a kit, or something, for the rest of us who aren't as diehard as he is.
you know, if you had two steppers at each point, and they spun in opposite directions (like with a gear or something to compensate for direction), you would have a lot less resonance. The opposing motions would cancel forces and keep them from going into the frame.
Love your dedication to insane print times AND quality. Weird questions have you experimented with something like air assist on laser cutters? A pump instead of fans?
PC's usually use larger diameter fans to quite them down which could be a solution for static fans. As for warping: when pushing things to the extreme, probably a thorough air flow simulation and optimization is required to better target the printed layer without cooling a big chung of the part(which is hard work but could have high benefits). Also the gantry could be made from 2 carbon fiber tubes with linear rail attached to them to increase stiffness (i don't think it will increase the weight of the gantry)
That is impressive as fuck! Also you telling the viewer you are not telling them that subscribing liking and commenting will help 247 brought me here. Cause I thought it's a funny way to tell your viewers to do those things.
That's a very important open question, yes. I don't have equipment in order to test this properly atm. We'll see if Stefan is going to do this some day - he hast he printers and machine for that.
Hello, first time on your channel. You have very well edited videos and especially good information on your videos. I have to say, this is the good stuff concerning 3d printing. Thumbs up !
Just got my first 3d printer this week. Made the mistake of following google search results, rather than deep diving forums and reddits. So I bought my Prusa Mk3S. Who doesnt love spending $800 for something that is half the speed of printers for half its price? lol
This is great - [stealth edit] - nay, this is AWESOME. I'd like to see another very simple measures to compare quality which you can do here even with this "dataset" of prints that you already have in the video. What 3D printer users want is faster speeds but only at comparable quality (generally speaking, outside of draft prints, etc). Here's a trick that we use in the wine world: the triangle test. Three 3D prints, two that are the same and one that is different. The only question that is asked (of a small but representative sample of people) is, "is one of these different?" If the majority of tasters/inspectors cannot tell the difference or choose the wrong answer, then the prints are comparable. So to compare printers and settings I would humbly suggest this methodology - then the other features of the printer/print can be compared apples to apples, since for your 247zero print at the same quality level (indistinguishable in a triangle test) the print time will be much much lower. What speed can you print these objects at with truly similar quality results to the others?
Inhaltlich hab ich genau das gesucht! Danke Dir. Hab grad zwei last Gen Drucker ,,geklippert" und mit tmc2208/9 bestückt. Beide laufen leise und gut, jedoch gedeckelt durch die Creality Hot Ends auf 10mm³/s. Wenn ich den Flaschenhals entferne, hat ein CoreXY noch immer einen rund 30% Geschwindigkeitsvorteil.
it would be interesting to think of a printer with the speed of this one but also with the table moving in the opposite direction so as to add inverse speed
Awesome work! I love to see people pushing the boundaries of 3d printing. Although it is fast, I think you will always run into the limitations of physics and the fluid dynamics of the material. This is a great example of quality vs quantity. If you are looking for precise prints that require engineering quality and need to have tight tolerances and integrity, fast printing just doesnt seem to work. Even if you tightened the frame up to be as solid as possible, I dont think the quality would ever compare to something printed slow. I have yet to see the quality of a fast print compare to a slow print.
1.What about using a compressed air line instead of the small fans on the print head. This would allow for the redesign of the ducting as you could more precisely direct a higher air flow where necessary. Make sure you use a water separator. 2.You could also chill the air for increased performance. Perhaps, chilled air would not need to be at a high pressure. This should also reduce the necessity for the noisy side fans. This should also reduce your vibrations, thus increasing your print quality. 3.You might consider using a carbon fiber square tube for your cross axis. They are very rigid and light weight. I have seen them for sale on Ali Express. Hopefully this will help reduce weight and increase rigidity, thus reducing your waviness.
Hmmm... The geek in me is jumping up and down in excitement and has a few thoughts: 1. Adding laminar flow to the fan output would concentrate the air vertically. Adding a lightweight shield that has a gap to line up with the nozzle--linked to the belt(s)--located between the fans and the air nozzles would focus the air horizontally, requiring less air pressure for the same cooling. 2. The fans don't have to be attached to the printer. If they are placed in a heavy enclosure with flexible hose & feet, their vibration will be dampened, which should help with the noise. 3. Actively cooling the air itself before it is blown around the nozzle would *DRASTICALLY* reduce the amount of air needed, and there are *MANY* commercial and DIY solutions that would work. 4. Have you considered using counterweights to combat vibration? That would mean throwing around twice the mass, but should (almost?) completely cancel the shaking cause by unbalanced forces. Cheers!
I had to look it up it was over a year ago I suggested the 247 Zero name… time flies. Did you officially get the name yet? Or is it still a nickname. Either way I am stoked you have been so successful!❤❤
@@247printing Yeah, it is probably not worth your time, they have a whole board of people to review and decide that stuff. I just thought maybe they were recognizing your contributions to the space and reached out to you… a long shot but not impossible.
Impressive, i like the real calculation behind a 3D printer. I love your design and the test of Stephane. I'm looking forward to your next upgrades and designs. M4 voron extruder is not good enough ? just wondering. cheers!
MORE SOUND, The micro equipment used for the printer was the same for the voice. frequencies clash. it's tinny quality/thin sound micro preset. Just adding bass to the voice and not letting reverb from walls to the mic will make it 200% easier for me to hear the words. Audio is difficult to focus with ears: 1/ I can hear reverberations from a small room and a microphone that isn't near enough or is too omni-cardiod 2/ The audio frequency is the same as the 3d printer, the two soundracks are clashing. the bass cut could go above and below for one or the other, there sounds like there is bass cut on the micro that is too high and loses some of the voice gravity. Other tips working with difficult audio is: practice putting silence between words and practice speaking at 50% speed, because slow motion speaking YT are generally the highest views.
I double your statements and you are absolutely right: My recording room is around 20m² and from an acoustic point of view it's hell on earth concerning reverb. Fighting with that on all of my videos... It's a RØDE NT1A microphone I use, but my software settings I do (which I tune last every time) don't help much... I'll do my best for the next video as also for me atm: The voice over can be annoying atm for some people (including myself)!
The frame mounted part cooling is a very cool idea. Can save weight for fans and wirings, even mounting bolts. You could find a way to receive supplied air more effectively and to control air speed based on the head acceleration. Air speed may have to be controled too according to required cooling for varying melt volume. Keep it up!!!
@@247printing Have you considered also replacing your delta fans with Artic S4028-6K fans in that configuration? If you can get them in a 5 pack, they're about 40 USD on amazon(which is far less than those deltas cost). They're the same dimensions and type of fan as those deltas, but far far far lower RPM(and therefore lower noise, amperage, air flow volume/static pressure), Since they're that same dimensions and they're normal 4 pin/12V fans, they should be a drop in replacement to lower everything down to "sufficient and not requiring hearing protection nor a second mortgage."
The makers of the 3D Benchy made this neat print called “Smartphone Photo Studio for #3DBenchy and tiny stuff”. I think this would be a great way to get documentation of how each benchy printed in a nice objective manner. Any thoughts on this?
like in motorsports and almost all areas of technical performance, it all ends up basically in the materials chosen to make the parts and how(shape and design) used to build those essential components, so they are lighter/stronger/more precise/better thermal efficiency. if you use steel screws to hold moving parts, that are built to support 3kg(in a certain kind of mechanical work), but you need it to bear only 600 grams of mass at peak effort, you should exchange them for titanium screws that are rated to 133%(of your total mass load of %100 or 600g), they should be 1/3rd the weight of the steel ones, if the titanium screws would need to carry the same stress or mass/speed work , they would be 30% lighter at least(consider all numbers here not precise values, but approximated enough for the comment session eheheh).
Very nice work! What about channeling air inside a tube and blow it directly where its needed (like the bowdentubes for the filament but maybe bigger) instead of blowing air all over the plane? This makes the head maybe even lighter and the air could be cooled down with a liquid cooling system and a radiator or with peltier elements... just a thought
For the torsion aspect I would recomand printing it with channels and adding materials later on such as carbon fiber sheets. This could have drastic efeect on stiffness while still being relatively cheap to do.
One problem I see with cooling is there are two airstreams blowing against each other which could cause stagnation when they hit each other. I think a better solution could be to have a windtunnel like concept, where the air is blown from one side and sucked from the other in a closed casing. Just an idea
I blame Sophie from Noctus3D :-) Yes the Hemera gave me a lot of annoyances, but I didn't want spend another 100 bucks... at first... Not using a Hemera shouldn't be an obstacle for a serial!
Rather than cooling by moving more air faster (and risking deformation), have you considered using *colder* air? A cheap A/C or refrigeration unit or even a peltier stack feeding the fans could probably let you drop your fan speed by 50% and still provide better cooling. Heck, even a block of dry ice in front of the fans might cut it.
I understand that this run for speed essentially speeds up the process of makin "nextgen" printers, but the real usability can in my opinion only be reached for larger printers. The small bed printers cannot have prints that will take really long time, because there is not enough place for them either way. Any future plans to really go for the proper "hardening" of the setups for the bigger bed size? The real benefits of fast prints only shows then, as 24h or more prints are really the problem.
Hopefully many of the learnings from printing small things very fast can be replicated on large format printers. Getting a 5-10x speedup on a multi-day print would be huge.
With that amount of air flow coming from the sides, you can remove the nozzle fan and just redirect some of the side airflow into the nozzle. Should save you some weight.
@@247printing Hi Albert, ja ich habe in letzter Zeit viel zu tun gehabt, auch mit der Schule ect... Deine Videos habe ich natürlich die ganze Zeit angeschaut! 😁 Ich habe ja auch mal gedacht einen Voron 0.0 zu bauen, war mir da immer unschlüssig. Aber als du deine Dateien veröffentlicht hast war das anders! Ich habe mir die STLs schon runter geladen und dir auch Trinkgeld da gelassen. Mach weiter so mit deinem Content 🙋♂️ Viele Grüße/ Luca 🤗
Another way to reduce weight, take a look at Frore Systems AirJets. You can probably look their CES demo up here on youtube. Crazy small, and way quieter, while still putting out some serious CFM.
Great content! I love your approach and to topics discussed. Have you looked into hybrid extruders? I have a Nimble Giddy on order that I am excited to give a go and is looking like a viable option from the herera!
I have a dream build, making a voron 0 style printer with an epoxy granite frame and awd servo setup. Then to make it really spiffy I would make the gantry and supporting pieces out of carbon fiber 👍
I'm not sure this would be completely workable , potential idea though , because you typically print very fast you always use the part cooling fan and because of that you should be able to get rid of the heatbreak fan and just redirect a bit of the part cooling flow up through the heat break. 😉 You will have to completely redesign the tool head but it should save you another 20 grams or maybe more ...
Heck yea! Great job so far! How can you go faster? Custom extrusion nozzle with integrated compressed air cooling powered by a vortex cooler, and of course linear motors instead of rotary motors, with sprockets and belts.
And here I was sitting there wondering. You're printing ABS, you're running open frame, your throwing mad cold air over the whole print. How is the print not turning into a Pringle? And if it's shape stable, that's still no guarantee that it's actually cohesive and fully bonded. But I see that you have something to show us in the future. I hope.
I think it would be great to see what exactly you are doing to find the bottlenecks of you're printers. Measuring flow rate is pretty simple, but I'm not sure how to tune the speeds and acceleration limits for going fast
It's a lot of testing and observing for each feature which has a speed setting. Quite iterative print -> adjust -> print -> adjust per setting thing. Not very exciting
