Honestly: The pinnacle of FDM printing is happening RIGHT HERE ! Congratulations again and keep it up - it's so cool to watch the progress on your channel since the early beginnings, which was not long ago! Awesome guy, awesome tech, awesome printers 👍
My boyfriend brought his Ender 3 couple months ago and I have fully fallen down the rabbit hole. Now with Klipper and a LOT of mods I'm happily going at 120 mm/s with great quality. Videos like these are nuts! Really sweet content!
In life, you can count on the fingers of one hand the amount of people that you admire for their talent in a field for which you are yourself passionate about. Vez, you probably occupy 2 or 3 fingers on that hand of mine. Congratulations! A lot of work, sweat and raw talent goes into the long journey that led you to this result. Printer performance, knowledge, videography and let's be honest! That Acting!!! Beau bonhomme et vraiment plaisant à t'écouter mon chum! ♥️ lots of love from myself and the HevORT community to you and you VzBoT project. See ya, 😉livier.
@@highspeedpiTV I agree! those are the guys who made me rush into making my own 3d printer, wasted so much time trying to gain a bit of speed but.. thanks again the journey is awesome
@26:49 The pattern you are seeing is the extruder stepper extruding the plastic in pulses instead of smoothly. Once you accelerate to a high enough speed you end up seeing the individual pulses of plastic from the stepper resolution. You see exactly the same effect with extremely slow flow rates on extremely thin layer heights, where the stepper isn't able to extrude in small enough microsteps and it comes out in pulses also. Highly geared steppers help reduce that effect, it essentially increases the resolution of your extruder steps.
Im not so sure :) how can you explain that it was gone with minor adjustement settings...and how to explain it was only happening on the first pass and not on the way back?
I initially thought this was correct. But wouldn't that result in a similar pattern, but under-extruded? Looking closely at the print, things look over-extruded rather than under-extruded.
@@Mehecanogeesir oooooo very interesting conclusion, I'm just mystified as to what the cause could be, my original thinking was that it wasn't extruding enough plastic and because of the nature of how stepper motors work i figured that was the plausible cause, sure gives something to think about though
You are the reason I tune and work so hard to gain every bit of speed on my voron and my volcomosq at quality. Thanks for all of the inspiration and congratulations! Truly these are the moments where we can see the future of what’s to come with this awesome community. Cheers to more fun and more learning!
@@DrumaierI’ve hit the dunning Krueger wall on this comment. I can’t tell if their understanding of kinematics is way better or way worse than mine lol
You are a 3D printing mad scientist! I'm happy when I get 150 mm/s on my Prusa Mk3S+ hahaha. Keep up the awesome work, and thank you for contributing your awesomeness to the community :-)
@@srck4035 I just hit 100 on my custom steel mk3s+. Will definitely try for 150 after my reinforcements of my workbench and better place the printer and mount it to bench.
It's not stupid if it works! You just keep finding ways to improve this platform and it's almost to the point to where every household of makers will have a VzBot of their own. I'm happy to be a part of this community as well, awesome work Vez! Just got my 5160HVs in, can't wait to put them in and try something stupid myself 😁
Just awesome, guys like you are the driving force behind innovation in 3d printing. Thank you for that and keep going? I’m building a VZbot330.. by the way I will be able to print sls in March ‘23 and will be happy to be part of the community👍👍
Jesus. You are getting better surface quality at 2kmm/s than I am at 35mm/s. I love this, and I think it's a brilliant demonstration of how commitment and motivation can drive you to develop a 3D printer that does exactly what you want. Thanks for sharing!
Vez just adds another „0“ to - what I thougt - an already good print speed and that beast of a printer just simply does it. Respect, more than just impressive!
You have input shaping of some form, right? As someone that’s been using floating point math for 40 years, I suspect you’re running into that as one of the issues: changing parameters a little bit can make surprising differences. I’ve seen something as a bug not be able to be machined due to 9 billionths of an inch making it not zero, when I worked for Hurco on CNC controller software. I suspect you’ll run into weird boundary conditions and repeat patterns with combinations of flow rate and speeds, and changing things just a little bit will make huge differences in results. That you’ve managed to get such results (which can destroy printers quickly!) is bonkers!
Yeah Klipper has a really nice and easy input shaping program integrated into it. Can set the parameters either through a printed test part you take measurements off of or an accelerometer. I have used it on both my printers and have seen a world of difference from it. Especially being able to print at night and not have everything in my closet resonating from the vibrations
Wow, and I was delighted to print on an Ender 3 at 110mm/s! This is super impressive, I hope that these kinds of speeds become mainstream (and a little quieter!) soon
Wuuuhuuuu!!! You're back! Seeing and hearing you is great again and now you explain sooooo much! So great, man! I really appreciate it sooo much! Cheers for all your work!
It’s like a car in that past certain speeds you run into a completely different set of physics issues to deal with. Incredibly impressive, just don’t let it burst into flames
Got my son an Ender 3 V2 for Christmas and got myself hooked - upgraded bed, springs to spacers, sherpa mini, dragon hf, Manta MK2 w/2x5015s, 60W heater, now easily doing 200mm/s at over 4500mm/s^2 accel - still tweaking and learning, but definitely planning a VzBoT build in the near future. Thanks!
The ABS looks like that because it's inertia is greater than it's adhesion force at those speeds, meaning it's still in motion when being applied to the previous layer and potentially with enough force to shift the previous layer slightly as well. Requires additional data.
My guess: the ripples are a result of the corexy mechanism. Straight movements like that are still a superposition of two DOF, where any deviations in the movement will cause lateral movement of the head. The toothed belts might also add some vibrations at those speeds. Still really impressive that this works this well, those are some crazy speeds.
I am obsessed with this machine, working very hard to build mine as perfect as your, but let me tell you ,is not cheap. One thing is sure, is fast and beautiful
its not cheap, but its not THAT expensive. Unless you go all in with all the CNC parts etc.. and high end stuff on electronics. then it can bring you close to 2K$
I printed 650mm/sec with 10k acceleration on my solidoodle 3 with PLA at 310C with a 0.35mm nozzle. the surface finish was like satin. the walls were pretty close to perfect. I had to put a lot of weight in the bottom of the printer and reinforce the bed against wiggling.
Could the pattern be from the steps of the extruder motor. At this speed could the higher and lower pressure of each step cause increased and decreased extrusion?
Nice!!! Can't wait to see 2000mm/sec benchy. If you are able to, you could get someone with those really high speed camera to film the tip of the nozzle at that speeds and see exactly what goes on, maybe that will help you tune things better. Maybe you know another youtuber that owns one of those bad boys...
your subscribers count are well deserved, it's no surprise to me when you build up such excellent community here. recognize some of the shoutout names from speaking to those guys on the discord they're all really great people too. just always feel so far behind that i cannot afford parts (still yet) to start building my own VzBot. But your videos are so well done, to share what you are doing in a way which is so accessible, it's such a great substitute to not having a VzBot of my own. In the so until next time, always wishing you the best. And always learning so much from all you guys and this whole extended community (including also these neighboring communities of annex and hevort, and others). It really is a team effort, and your gratitude, it is also my gratitude.
Fist time seeing one of your videos. And my reaction was the same as yours. Just started giggling at the audacity every time it stepped up speed lmao. Good show sir
_"Why the hell would you do this?"_ I think it's a bit like Formula One racing: Push a technology to its limits in an unnecessary way, in order to make discoveries which can be used practically.
Wow this is just insane it's so fast probably too fast but this shows how much 3d printing has improved and to be able to print at this speed is just amazing and your passion shows in this video
I wonder how the speed would be affected if you just had one large circle, or maybe a square with rounded corners. Maybe a 220mm diameter circle or a 220mm square with maybe rounded corners with maybe a 25mm radius.
The most impressive thing imo here is how stable and quiet the printer is at these speeds, other printer approaching 500mm/s wobble like crazy and need ear protection to be in the same room.
I'm very interested in purchasing a pre built unit exactly like yours or at least a 100% complete kit. I don't really have the time to mess around sourcing all these parts on my own.
@@Vez3D Kit for the 235 or bigger as well? And another question. If you would build a 330 right now... would scale the 235 or stick to the old design (just with AWD)?
Since the pattern is so regular I would gess it has to do with pressure changes in the extruder that happens due to the steps in the filament feeder motor. Those would be invisible on slower speeds but at these extreme speeds each step of the feeder motor creates a visable preassure change and a unstable flow. Could be fixable with higher resolution stepper motor or higher gear reduction?
How's the table suspended or shock mounted. You know what happens when it travels that fast but only goes a distance of a mm, right? Back/forth/back/forth. The bed will start to shake and possibly lift off it's supports. Especially when the table is very high or very low in the vertical position. I see that on my markforged onyx machine. So, I put some very "soft/squishy" mounts at all 4 corners.
Are you using a possibly worn out CHT? I get that kind of artifact when the nozzle orifice has worn down to be wider than the line width and the layers are thin. Especially on overhangs but it happens on non overhang too.
Congratulations Simon! So impressed with your results! Hope you will never stop with your project and just push it forward. Best DIY printer from first day of development. Soon starting to build my. Is it possible to ask from you model which you are printing in this video please?
have you considered chipset waterblocks, that is WBs meant for computer chipsets and mounting those to the bottom/back of those XY motors. Though water cooling stepper motors with PC water cooling parts might sound insane. does that X linear rail heat up at these kind of speeds?
I would bet you can tune that salmon skin pattern out of it. I had this problem one time, I think acceleration and jerk setting corrected the salmon skin. I don’t quite remember so have a look.
Could an extruder moving that fast cause cooling issue with the print? (due to it push air around as it moves so fast) Anyway, what an amazing project! Good work!
It really is amazing, on another note I knew it's ABS, but if you try 'tough PLA" even at 1000mm/s and you get 10 like that it's 10,000mm/s it's like you can 3d print a car within a couple of days, it's a game changer for everything.
With the graphical model at 13:50 I think you can just y shift the graph do that the 0 is located at your corner speed. I may be wrong but I think that would work
Congratulations! I really like the answer at 1:22 😀 Thank you for the joyful laughter in this video! Finally, may I ask which slicer did you use to generate the testing print files? The fact that using 0.55mm layer width and 120% extrusion ratio can get better print is interesting! Thanks!
I think that ripple could be due the rheological properties of the molten plastic, it might behave like a non Newtonian fluid or it might be such a fast surface speed that it forms a small circling current that makes the plastic build up behind the nozzle until it detaches and the cycle repeats. Kinda like how air vortexes form behind a truck.
Hi there, maybe one day I'll test of of these in my chanel LOL!!! I think the pattern is due to DRAG on the fillament as it lays on the print, maybe try to cheat the slicer changing the E-steps/mm making it extrude 70mm3 thinking it is extruding 60mm3... and as you said without cooling look better and it just keeps going to that direction of drag in my head!!! Another thing that leads me to that is that the second pass looks better because instead of air it is printing near a hotter surface(plastic itself)... just throwing ideas here!!! Love your content!!! Congrats on your 20K mark!!! Best regards from Brazil!!!
34:04 That's insane quality considering the speed and the high res camera. Great work. You should consider finding a manufacturer or becoming a parts manufacturer. Perhaps you already are ?
hello my friend from ukraine. I'm interested in the work of a hot end with an ultra-flexible TPU with a hardness of 75 А Shore.The fact is that this is a very difficult TPU for printing at high speeds, even directs with a short feed cannot develop high speeds. Thanks for the answer.
At 2000mm/s you're moving at nearly 3 times the speed of ASML's EUV Lithography machine, which is pretty freakin incredible. Granted I think with their precision measured in nanometers, they might have you beat accuracy wise ;-). For real though dude, this is awesome. I was stoked to see 1000mm/s before, but 2000 is even more ludicrous. Can't wait to see how much faster you're able to get it moving.
You never got a good close up from the left front of the part, but from what i could see it was smooth there, or at least had less ripple. You need some sort of active vibration compensation for those speeds. Even 300mm/s requires it, theres just too much mass here thats causing the ripples. Edit: rotate the part 45 degrees so its only running one stepper on the straights and run it again, and see what the pattern does. If its smoother in the middle and worse near the corners you've confirmed that its vibration/resonance.
@@Vez3D Well the reason you even slow down currently is because the print head needs to take a fairly steep curve. As you make that curve larger in radius, you can afford to slow down less... and I am proposing that at a large enough radius, you don't have to slow down below your target speed... The question really is, if the radius of curvature for 2000mm+ speeds is smaller or larger than half the shortest dimension of your print area.
@@BirdbrainEngineer im well aware of that :) that small radius is still going very fast .. at least it will not go under 200mm/s as set per SCV, so the straight lines are probably keeping full 2000mms speed for 200mm long. having a large circle with big radius still probably wont be good. I could be wrong though... unless we set SCV to 2000? A circle, even with big radius will be made out of multiple tiny straight lines, so a lot accel/decel, and always calculated based on SVC and accel/decel setting.. and lots and lots of processing. So I think it will be very hard to keep full speed... unless SVC is to set to max speed. But again.. i can be wrong here :)
@@Vez3D On a circle approximated with 10° turns, the junction velocity limit is around 10x SCV. With a smoother approximation it gets better but the returns are diminishing. Maybe around 20x SCV is the best you can hope for. There's also a centripetal velocity limit of something like √(accel×radius)
Sorry if I missed this in all the comments below: I think the reason your flow rate differed from what was expected is that the nova flow rate calculator appears to use a line width that is equal to the nozzle width. There doesn't appear to be a way to change this. Are you using the 20t motor pulleys in the BOM or something else? I ask because *theoretically* (according to the motor torque simulator database spreadsheet from @eddietheengineer), a single motor should be capable of this, but with a 20t pulley, would skip/fail around 1400. Halving that torque requirement (because 2 motors = ~2x the torque, would still have failure at around 1750). The gearing required to hit 2100 (the 2000 here with a small safety margin) for 2 motors is around 26t. Your achievement is impressive. I'm just trying to figure out where things need to change as I'm modeling a slightly modded vzbot around this spreadsheet. Thanks!
Super video. Maybe some keys to make this process mature coule be to work on water cooling stepmotors or hydraulic actuators. Another interresting thing, maybe easier than other ideas, is to try to make slowmotion videos to see plastic layer behavior. At these speeds it would be necesseary to perform modal analysis by FEA calculations.
my cell can do 960fps.. and there is short clip of that in the video.. but its not super good quality. It needs TONS of light to have good quality on the image since the shutter speed is soooo fast... and the lights I have are all 60pfs because that is the Electricity cycle we have here in my country.. so its not fast enough... so it just relies on my DC LED on the printer for now. I would love to invest more money into a nice speed cam and speed lights... but money money. hehehe. Maybe if camera companny could sponsor a video and rent me a camera ? ill see
I'm sitting here having troubles printing at 50mm/s and this guy is doing 2000... Kudos and good job!
u are not only one i broke my nozzel on 80mms and damaged bad 10 cm in middle.. and this guy fliyng on 2000
Honestly: The pinnacle of FDM printing is happening RIGHT HERE !
Congratulations again and keep it up - it's so cool to watch the progress on your channel since the early beginnings, which was not long ago!
Awesome guy, awesome tech, awesome printers 👍
It means a lot to me brother. thanks a lot
@@Vez3D Musk of 3dp
My boyfriend brought his Ender 3 couple months ago and I have fully fallen down the rabbit hole. Now with Klipper and a LOT of mods I'm happily going at 120 mm/s with great quality. Videos like these are nuts! Really sweet content!
I had my stock ender 3 printing at 110mm/s with excellent quality.
In life, you can count on the fingers of one hand the amount of people that you admire for their talent in a field for which you are yourself passionate about. Vez, you probably occupy 2 or 3 fingers on that hand of mine. Congratulations! A lot of work, sweat and raw talent goes into the long journey that led you to this result. Printer performance, knowledge, videography and let's be honest! That Acting!!! Beau bonhomme et vraiment plaisant à t'écouter mon chum! ♥️ lots of love from myself and the HevORT community to you and you VzBoT project. See ya, 😉livier.
Ahhhh you are so nice :) its the same for me. Lets rent a room asap!!!!
Vez and MirageC u guys rock!
@@highspeedpiTV I agree! those are the guys who made me rush into making my own 3d printer, wasted so much time trying to gain a bit of speed but.. thanks again the journey is awesome
@@Vez3D loooolllll youguys rock
(Oh, un québécois! 😄)
Vez, you have no idea how inspiring your videos are!
2000mm/s on a DYI printer - a you freaking kidding me?!!
That is insane, dude! Love it!
@26:49 The pattern you are seeing is the extruder stepper extruding the plastic in pulses instead of smoothly. Once you accelerate to a high enough speed you end up seeing the individual pulses of plastic from the stepper resolution. You see exactly the same effect with extremely slow flow rates on extremely thin layer heights, where the stepper isn't able to extrude in small enough microsteps and it comes out in pulses also. Highly geared steppers help reduce that effect, it essentially increases the resolution of your extruder steps.
Im not so sure :) how can you explain that it was gone with minor adjustement settings...and how to explain it was only happening on the first pass and not on the way back?
while that does sound logical, im not sure these pulses would make it trough that huge melt zone.
I initially thought this was correct. But wouldn't that result in a similar pattern, but under-extruded?
Looking closely at the print, things look over-extruded rather than under-extruded.
@@2mD not if the extruder wasn't extruding enough plastic, if it wasn't extruding enough plastic then the effect would be amplified
@@Mehecanogeesir oooooo very interesting conclusion, I'm just mystified as to what the cause could be, my original thinking was that it wasn't extruding enough plastic and because of the nature of how stepper motors work i figured that was the plausible cause, sure gives something to think about though
You are the reason I tune and work so hard to gain every bit of speed on my voron and my volcomosq at quality. Thanks for all of the inspiration and congratulations! Truly these are the moments where we can see the future of what’s to come with this awesome community. Cheers to more fun and more learning!
"It's starting very slow, at 300 mm/s" - You made my CR-10 cry.
Insane layer stacking quality at that speed. Shows that the printer has excellent Z Kinematics.
Isn't z the easiest thing to do for this print? I'm impressed by the extruder and hotend firstly, then the xy speed is mind blowing, just a blur 🚀
@@DrumaierI’ve hit the dunning Krueger wall on this comment. I can’t tell if their understanding of kinematics is way better or way worse than mine lol
You are a 3D printing mad scientist! I'm happy when I get 150 mm/s on my Prusa Mk3S+ hahaha. Keep up the awesome work, and thank you for contributing your awesomeness to the community :-)
150 is super fast for a bed slinger
@@srck4035 I just hit 100 on my custom steel mk3s+. Will definitely try for 150 after my reinforcements of my workbench and better place the printer and mount it to bench.
Im happy with 60mm/s haha
@@jakobwest4811 nothing wrong with that. I just want to see how far it can go.
impressive progress! 👏😎
thx
It's not stupid if it works! You just keep finding ways to improve this platform and it's almost to the point to where every household of makers will have a VzBot of their own. I'm happy to be a part of this community as well, awesome work Vez! Just got my 5160HVs in, can't wait to put them in and try something stupid myself 😁
Just awesome, guys like you are the driving force behind innovation in 3d printing. Thank you for that and keep going? I’m building a VZbot330.. by the way I will be able to print sls in March ‘23 and will be happy to be part of the community👍👍
definately keep me posted on this :)
@@Vez3D Is your printer related to the Voron printers?
@@termgrecos3451 no. But some parts were inspired by voron yeah
Jesus. You are getting better surface quality at 2kmm/s than I am at 35mm/s. I love this, and I think it's a brilliant demonstration of how commitment and motivation can drive you to develop a 3D printer that does exactly what you want.
Thanks for sharing!
there is no such thing as "kmm". the prefixes are made in such way, that you use only one of them. in this case you need no prefix as it's 2 m/s
Thank you for the shout out brother. I'm proud to be apart of this.
Vez just adds another „0“ to - what I thougt - an already good print speed and that beast of a printer just simply does it. Respect, more than just impressive!
You have input shaping of some form, right?
As someone that’s been using floating point math for 40 years, I suspect you’re running into that as one of the issues: changing parameters a little bit can make surprising differences. I’ve seen something as a bug not be able to be machined due to 9 billionths of an inch making it not zero, when I worked for Hurco on CNC controller software.
I suspect you’ll run into weird boundary conditions and repeat patterns with combinations of flow rate and speeds, and changing things just a little bit will make huge differences in results.
That you’ve managed to get such results (which can destroy printers quickly!) is bonkers!
Yeah Klipper has a really nice and easy input shaping program integrated into it. Can set the parameters either through a printed test part you take measurements off of or an accelerometer. I have used it on both my printers and have seen a world of difference from it. Especially being able to print at night and not have everything in my closet resonating from the vibrations
Wow, and I was delighted to print on an Ender 3 at 110mm/s! This is super impressive, I hope that these kinds of speeds become mainstream (and a little quieter!) soon
Wuuuhuuuu!!!
You're back! Seeing and hearing you is great again and now you explain sooooo much! So great, man! I really appreciate it sooo much! Cheers for all your work!
Dude freaking removed the mili from milimeters/s. That is amazing,how it can actually print something coherent.
It’s like a car in that past certain speeds you run into a completely different set of physics issues to deal with.
Incredibly impressive, just don’t let it burst into flames
love it bro keep it up im blown away with your quality and attention to detail in every part and every step !
Got my son an Ender 3 V2 for Christmas and got myself hooked - upgraded bed, springs to spacers, sherpa mini, dragon hf, Manta MK2 w/2x5015s, 60W heater, now easily doing 200mm/s at over 4500mm/s^2 accel - still tweaking and learning, but definitely planning a VzBoT build in the near future. Thanks!
This is very cool print video my Friend😃🙂🤝🤝🤝
Congratulations Vez, that is an impressive achievement. You are pushing the limits. Keep up the amazing discovery you trailblazer you ; )
Glad to be a part of the journey!
Wow. This is a crazy printer. First time I’ve encountered it. I need to learn more about it. Should I build this instead of a voron?
The ABS looks like that because it's inertia is greater than it's adhesion force at those speeds, meaning it's still in motion when being applied to the previous layer and potentially with enough force to shift the previous layer slightly as well. Requires additional data.
My guess: the ripples are a result of the corexy mechanism. Straight movements like that are still a superposition of two DOF, where any deviations in the movement will cause lateral movement of the head. The toothed belts might also add some vibrations at those speeds.
Still really impressive that this works this well, those are some crazy speeds.
Woohoo fun stuff!🥳
That pattern is the interference of frequency of feed stepper motor and ABS feed stepper motor.
I am obsessed with this machine, working very hard to build mine as perfect as your, but let me tell you ,is not cheap. One thing is sure, is fast and beautiful
its not cheap, but its not THAT expensive. Unless you go all in with all the CNC parts etc.. and high end stuff on electronics. then it can bring you close to 2K$
I printed 650mm/sec with 10k acceleration on my solidoodle 3 with PLA at 310C with a 0.35mm nozzle. the surface finish was like satin. the walls were pretty close to perfect. I had to put a lot of weight in the bottom of the printer and reinforce the bed against wiggling.
Could the pattern be from the steps of the extruder motor. At this speed could the higher and lower pressure of each step cause increased and decreased extrusion?
could be yes
Nice!!! Can't wait to see 2000mm/sec benchy. If you are able to, you could get someone with those really high speed camera to film the tip of the nozzle at that speeds and see exactly what goes on, maybe that will help you tune things better. Maybe you know another youtuber that owns one of those bad boys...
Congrats on the channel and the amazing print speed!! You Rock!!!
do we finally start to measure the speed in meters per second instead of millimeters?
your subscribers count are well deserved, it's no surprise to me when you build up such excellent community here. recognize some of the shoutout names from speaking to those guys on the discord they're all really great people too. just always feel so far behind that i cannot afford parts (still yet) to start building my own VzBot. But your videos are so well done, to share what you are doing in a way which is so accessible, it's such a great substitute to not having a VzBot of my own. In the so until next time, always wishing you the best. And always learning so much from all you guys and this whole extended community (including also these neighboring communities of annex and hevort, and others). It really is a team effort, and your gratitude, it is also my gratitude.
That is an incredible printer. I wish I could build something HALF that badass. Well done. Abbsolutely gorgeous technical piece of artwork.
You may have to build a solution to water cool the motors... but HOLY SNAP DRAGONS
*standing ovation*
Fist time seeing one of your videos. And my reaction was the same as yours. Just started giggling at the audacity every time it stepped up speed lmao. Good show sir
_"Why the hell would you do this?"_ I think it's a bit like Formula One racing: Push a technology to its limits in an unnecessary way, in order to make discoveries which can be used practically.
Holy shiz dude if I hadn’t been following this channel for awhile I never would have believed it.
Shoot, 500mm/sec still blows my mind
Wow this is just insane it's so fast probably too fast but this shows how much 3d printing has improved and to be able to print at this speed is just amazing and your passion shows in this video
بتعمل اي يا ابراهيم
I wonder how the speed would be affected if you just had one large circle, or maybe a square with rounded corners. Maybe a 220mm diameter circle or a 220mm square with maybe rounded corners with maybe a 25mm radius.
I can see flow went from 57mms on 1900 to 60mms on 200 so lets say you can go up to 2200 mms ?
The most impressive thing imo here is how stable and quiet the printer is at these speeds, other printer approaching 500mm/s wobble like crazy and need ear protection to be in the same room.
I am adoring your machine!!! Solid AF! 😱😄 AND THAT VIDEO QUALITY!
Congrats on the 20k subs mate. Very well deserved 👍🇦🇺😊
I'm very interested in purchasing a pre built unit exactly like yours or at least a 100% complete kit. I don't really have the time to mess around sourcing all these parts on my own.
we are working on it, and we are very close to releasing it :)
@@Vez3D i can just visit France to buy one, so you don't have to deal with patents if any.
@@Vez3D Kit for the 235 or bigger as well? And another question. If you would build a 330 right now... would scale the 235 or stick to the old design (just with AWD)?
@@ResistanceLion Vez is Canadian (Im proud to say!)
@@pmcquay1 Wow, then i need shipping
That is f****ng insane!! Btw love that CR3D PEI sheet!!!
If something is stupid and works - it's not stupid
Man now when I look at my printer working it looks like a turtle LOL
this is the coolest printer I've ever seen bro, nice work. This is what I think enders will look like in 10 years
Watercooling the four motors next?
I'm absolutely blown away!
So awesome Vez!! Great job ;)
Now do that print rotated 90 degrees! I wanna see that gantry being thrown around at 2000 mm/s like that!! :D
Perhaps the pattern on the walls is extruder stepper cogging just from being driven so fast?
I'm first time on this channel. This is nuts!
This is also really interesting from an economical point of view. Having fast printers can bring down the cost for print-on-demands.
Since the pattern is so regular I would gess it has to do with pressure changes in the extruder that happens due to the steps in the filament feeder motor. Those would be invisible on slower speeds but at these extreme speeds each step of the feeder motor creates a visable preassure change and a unstable flow. Could be fixable with higher resolution stepper motor or higher gear reduction?
26:42 the pattern might be extrusion-pressure pulsing from the stepper motor in your extruder
HOOOOW bruh HOW can the print head not break, and the belts, and the arms, and everything this blows my mind
So glad you went with milled aluminum.
How's the table suspended or shock mounted. You know what happens when it travels that fast but only goes a distance of a mm, right? Back/forth/back/forth. The bed will start to shake and possibly lift off it's supports. Especially when the table is very high or very low in the vertical position. I see that on my markforged onyx machine. So, I put some very "soft/squishy" mounts at all 4 corners.
Printer is bolted to concrete wall ;)
Are you using a possibly worn out CHT? I get that kind of artifact when the nozzle orifice has worn down to be wider than the line width and the layers are thin. Especially on overhangs but it happens on non overhang too.
Brand new cht
can you make a video zoomed in, to check the shakiness and the mechanical limitation?
Congratulations Simon! So impressed with your results! Hope you will never stop with your project and just push it forward. Best DIY printer from first day of development. Soon starting to build my.
Is it possible to ask from you model which you are printing in this video please?
here: github.com/VzBoT3D/VzBoT-Vz235/tree/main/Assemblies%20%26%20STL/FlowTest
@@Vez3D Thanks a lot! tested it,max what i can get now from my printer is 1600mm/s on 30k.
have you considered chipset waterblocks, that is WBs meant for computer chipsets and mounting those to the bottom/back of those XY motors. Though water cooling stepper motors with PC water cooling parts might sound insane.
does that X linear rail heat up at these kind of speeds?
I have my 330 on AWD. It works great. With Tronxy motors and 24v will still hit 800mm/s
I would bet you can tune that salmon skin pattern out of it. I had this problem one time, I think acceleration and jerk setting corrected the salmon skin. I don’t quite remember so have a look.
Could an extruder moving that fast cause cooling issue with the print? (due to it push air around as it moves so fast) Anyway, what an amazing project! Good work!
It really is amazing, on another note I knew it's ABS, but if you try 'tough PLA" even at 1000mm/s and you get 10 like that it's 10,000mm/s it's like you can 3d print a car within a couple of days, it's a game changer for everything.
With the graphical model at 13:50 I think you can just y shift the graph do that the 0 is located at your corner speed. I may be wrong but I think that would work
Congratulations! I really like the answer at 1:22 😀
Thank you for the joyful laughter in this video!
Finally, may I ask which slicer did you use to generate the testing print files?
The fact that using 0.55mm layer width and 120% extrusion ratio can get better print is interesting! Thanks!
I'm happy with you! 🥳 🥰
I think that ripple could be due the rheological properties of the molten plastic, it might behave like a non Newtonian fluid or it might be such a fast surface speed that it forms a small circling current that makes the plastic build up behind the nozzle until it detaches and the cycle repeats. Kinda like how air vortexes form behind a truck.
can you borrow a highspeed camera to see where the ripple comes from exactly?
That is pretty rad. Slight underextrusion. Maybe try PLA? Less heat needed. Which software is that?
😄😄😄absolutly crazy! nice job.
Hi there, maybe one day I'll test of of these in my chanel LOL!!! I think the pattern is due to DRAG on the fillament as it lays on the print, maybe try to cheat the slicer changing the E-steps/mm making it extrude 70mm3 thinking it is extruding 60mm3... and as you said without cooling look better and it just keeps going to that direction of drag in my head!!! Another thing that leads me to that is that the second pass looks better because instead of air it is printing near a hotter surface(plastic itself)... just throwing ideas here!!! Love your content!!! Congrats on your 20K mark!!! Best regards from Brazil!!!
PS: your 20K will pass my 30K in no time!!! Next year I bet you will be close to 100K while I'll be chearing my 40s LOL
34:04 That's insane quality considering the speed and the high res camera. Great work. You should consider finding a manufacturer or becoming a parts manufacturer. Perhaps you already are ?
I like your cam setup, what's the lens/camera pls?
This is so clean, just beautiful, wont ask how much you spent on this 🤑🤑🤑🤑🤑🤑
If you want to build it it will be arround 1500$
hello my friend from ukraine. I'm interested in the work of a hot end with an ultra-flexible TPU with a hardness of 75 А Shore.The fact is that this is a very difficult TPU for printing at high speeds, even directs with a short feed cannot develop high speeds. Thanks for the answer.
Of course printing speed is very important as it is also an indicator for capacity of better good printing quality as well.
29:00 Could be a shock wave effect.
At 2000mm/s you're moving at nearly 3 times the speed of ASML's EUV Lithography machine, which is pretty freakin incredible. Granted I think with their precision measured in nanometers, they might have you beat accuracy wise ;-). For real though dude, this is awesome. I was stoked to see 1000mm/s before, but 2000 is even more ludicrous. Can't wait to see how much faster you're able to get it moving.
Do you think you would do a even bigger printer maybe 500mm m3
Pla probably will work better and probably running temp on extruder bit higher than normal to keep up flow?
No..pla is worse from my tests
as someone stepping back into the hobby after like a 4 year hiatus, that is fucking incredible and my monoprice mini is feeling very self-conscious xD
How about trying brushless motors with feedback loop?
You never got a good close up from the left front of the part, but from what i could see it was smooth there, or at least had less ripple. You need some sort of active vibration compensation for those speeds. Even 300mm/s requires it, theres just too much mass here thats causing the ripples.
Edit: rotate the part 45 degrees so its only running one stepper on the straights and run it again, and see what the pattern does. If its smoother in the middle and worse near the corners you've confirmed that its vibration/resonance.
A tungsten carbide CHT nozzle? That's fascinating! Where can I get one?
No..its a bozzle nozzle
Wouldn't a large circular path be better for high speeds? Especially if you would like to have the speed be closer to constant.
not sure.. if it would be able to reach full speed with circular
@@Vez3D Well the reason you even slow down currently is because the print head needs to take a fairly steep curve. As you make that curve larger in radius, you can afford to slow down less... and I am proposing that at a large enough radius, you don't have to slow down below your target speed... The question really is, if the radius of curvature for 2000mm+ speeds is smaller or larger than half the shortest dimension of your print area.
@@BirdbrainEngineer im well aware of that :) that small radius is still going very fast .. at least it will not go under 200mm/s as set per SCV, so the straight lines are probably keeping full 2000mms speed for 200mm long. having a large circle with big radius still probably wont be good. I could be wrong though... unless we set SCV to 2000? A circle, even with big radius will be made out of multiple tiny straight lines, so a lot accel/decel, and always calculated based on SVC and accel/decel setting.. and lots and lots of processing. So I think it will be very hard to keep full speed... unless SVC is to set to max speed. But again.. i can be wrong here :)
@@Vez3D On a circle approximated with 10° turns, the junction velocity limit is around 10x SCV. With a smoother approximation it gets better but the returns are diminishing. Maybe around 20x SCV is the best you can hope for. There's also a centripetal velocity limit of something like √(accel×radius)
@@daliasprints9798 ill look into.. thakns guys for the feedback :)
I wanna see a video on those drivers, there doesn't appear to be a good one anywhere
they should use nozzle angle speed compensation, so that material lbasically falls to place at zero speed being ejected at certain speed..
Great work! What gcode viewer did you use at 19:05 ? It looks great
Have you thought of trying the Mg95 ABS I think is what it's called made for high speed
Sorry if I missed this in all the comments below: I think the reason your flow rate differed from what was expected is that the nova flow rate calculator appears to use a line width that is equal to the nozzle width. There doesn't appear to be a way to change this.
Are you using the 20t motor pulleys in the BOM or something else?
I ask because *theoretically* (according to the motor torque simulator database spreadsheet from @eddietheengineer), a single motor should be capable of this, but with a 20t pulley, would skip/fail around 1400. Halving that torque requirement (because 2 motors = ~2x the torque, would still have failure at around 1750). The gearing required to hit 2100 (the 2000 here with a small safety margin) for 2 motors is around 26t. Your achievement is impressive. I'm just trying to figure out where things need to change as I'm modeling a slightly modded vzbot around this spreadsheet. Thanks!
Yes the calculator use width = nozzle
20t gear
Super video. Maybe some keys to make this process mature coule be to work on water cooling stepmotors or hydraulic actuators. Another interresting thing, maybe easier than other ideas, is to try to make slowmotion videos to see plastic layer behavior. At these speeds it would be necesseary to perform modal analysis by FEA calculations.
Great idea, maybe he should get a Chronos high-speed camera, even a 1.4 should work great for this...
my cell can do 960fps.. and there is short clip of that in the video.. but its not super good quality. It needs TONS of light to have good quality on the image since the shutter speed is soooo fast... and the lights I have are all 60pfs because that is the Electricity cycle we have here in my country.. so its not fast enough... so it just relies on my DC LED on the printer for now. I would love to invest more money into a nice speed cam and speed lights... but money money. hehehe. Maybe if camera companny could sponsor a video and rent me a camera ? ill see
oh and dont worry, water cooling is the next step on the motors.. all parts were orderd 2 weeks ago.. :)
@@Vez3D you impress me a lot!