Making cheap components accessible for hobbyists while fixing their downsides... this is video is everything what is RIGHT about 'youtube blogging' and restores my faith in humanity!
@@UncleF115 let's look at the content of the comment: the comment shows adoration for a yt creator for showing how to take a cheap component into a more capable version while saving a significant amount of money, the original comment also alludes to the yt community as a whole in contrast to this video. I think even the most fervent fan of many yt channels would admit that the majority of yt content could be considered frivolous, pedestrian, non helpful, etc. Nothing against that, even within the 3d printing community engineering videos with humor and appreciable understanding and skill is not the majority. Most content in this space consists of simple reviews (to the point of barely being more then a spec sheet read or outright advertising-again not hating just stating the environment. Creators have to eat) or "look at my thingiverse model I printed" (again no judgement, it has its place especially for children or people new to the scene-it's how I introduced my wife to the hobby). The original comment could be perhaps energetic but people get excited (I know I am) at what possibilities open up on the diy scene. The person was far more concise than I have been for sure. If you have any videos that you feel perhaps are more worthy of praise please share especially if they are aimed at the making/engineering scene. I assume from your username and previous comment you may see yourself as a person with more discerning taste. Please do share any links or channels worthy of general praise.
This is what engineering is all about, see a limitation and find the best solution to delimit within the working constraints. This is a nice solution that allows one to save a ton of money and print parts that are good enough for most jobs, I'm impressed.
As my dad says "what you don't know will kill you". I had been planning on making a printer with the $20 ball screws, but I didn't understand the issues I could encounter from them! Thanks for this very interesting and informative video!
I wanted to build a cnc with them. Sheeesh the decouplers are a must. Good old Oldham could work, but ball bearing wing thing like in this video is even better!
I need to say, that I am absolutely super impressed with the solution to deal with using cheap parts, also the fact you share this with the hobby community it is just outstanding. I am about to embark on a high temperature version of your DIY for carbon fibre printing. So excited. Once again ingenious solution to the problem.
@@TeaObvious - I agree, totallyObviously a great channel which deserves all the best (And it seems to be working - RUclips promoted this to me on the frontpage :ø)
OH MY thats high quality content and i thought about the same but after i looked up the price i threw the project in the thrash bin... im amazed that u solved the problem soo nicely Congrats!
This approach is quite smart! I worked on an approach to solve this when still using home depot threaded rods during the reprap times and it never really worked well. This approach is brilliant in comparison! And I even like the name :D
You gave me a perfect idea to what to use as a lubricant, hmm, I thing that turbojet hybrid ceramic bearings are the best thing to test this kind of application for, hmhmhm, MaPle SYroUp.
Excellent WORK Thank you for the clear, detailed explanations and their implementation in Fusion 360. As usual, you have made an excellent analysis and found the right solutions for the use of inexpensive ball screwdrivers. Bravo et merci pour ce partage. un cousin des alpes Françaises ...
Ok, I'm no engineer, but that was freaking awesome as far as the design and the implementation is really straight forward to implement. You have a new subscriber.
Thank you David 😊 I do have a new solution that is a lot more performant than this one. you can see a preview here on my channel dedicated to HevORT : ruclips.net/video/kpUDHc5eknM/видео.html
Great video, great analytics, and innovation to solve a problem! About ball and lead screws: They are more accurate and last much longer if they don't support weight, such as the Z axis. This should be done with a counterweight or counter spring that reduces the force on the screw to a few ounces. Even massive milling machines with massive ball screws use big counterweights or their screws wouldn't last a month. It could introduce backlash, but that's another problem for you to defeat!
I like the counter weight idea... make the counterweight big enough and now the ball bearing is pulling down the bed, instead of a ball bearing and magnet to support the plate, a simple set of strings might be enough. Or invert the whole thing, instead of having the bed pushed up, have it pulled up (suspended on the ballscrews.... no counter weight, same results
Very cool solution. What I did with my z-wobbly screws was set them free. They spun in ball bearings at the top. I simply removed those bearings. Now my cheap screws end in the larger hole that used to hold the bearings. This solution only works because my build plate travels on four hardened locating rods which your hevort does not have and which would kill your planar printing concept.
Very nice, clever, and fun approach! I would also consider simply removing the support bearing at the top of the ball screw, and replacing it with a relatively loose fitting plastic/bronze bushing. The clearance (looseness) should be large enough to enable the imperfect ball screw to freely 'wobble' radially, without transmitting that wobble to the platform. It's not a fun, elegant solution, but it may work very well. Thank you -- I watched your entire video :)
Yet an another amazing video, please upload them more often. As a fellow engineer I love the designs, personally I would have just gone with a good quality lead screw if price is an issue. Cheapo ball screw additional to the wobble might have deviations in pitch as well. Also experience showed me simple always wins at long term use.
All screws have pitch errors. It just depends on your accuracy requirement. When the screws are rolled short term deviation is probably quite small. It will depend on the last couple of turns of the tooling, which one imagines is smoothly channelled. Absolute error is something else. That depends on how accurately the rolls are made. For short lengths it is probably irrelevant but for 2 mtr lengths a cheapy could be interesting. If you can accurately measure those distances any error can be accounted for in the software. The real warthog in the workshop is temperature variation. With a daily domestic shed variation of 30 degrees c. here in Oz the 2 mt shaft varies by some 0.4 mm. or 15 thou. Because of this and I am interested in absolute accuracy, just because I can be, I am looking at measuring displacement independently of the drive system.
Exceptionally Brilliant. I could not figure out what was causing the periodic wall deviations, but 12:10 explained exactly what is happening, as I held a sample up to my Z-axis and sure enough, an exact match to my thread pitch; also, I could see the upper unsupported section of the leadscrew wobble in an orbit. Thank you.
Excellent engineering design and solution! Reminds me of the experiments of earth quake on a tall building. The horizontal movement can be damped by putting a roller underneath the tall object.
@@MirageC Quelle surprise ! Merci pour votre réponse :) Je suis attentif devant votre travail qui m'inspire pour un prototype d'imprimante innovante. Pour moi, la voie que vous avez choisi à travers les duets/servo moteur/X Y Core, etc... et une recherche de performances maximales d'une FDM. Je pense que vous pouvez encore améliorer vos performances en allégeant vos Axes X et Y avec du Composite de type "Igus" par exemple. Il permet aussi de diminuer les frictions, le bruit à haute vitesse et d'utiliser des lubrifiants secs (silicone ou PTFE). Bref... La limite est dans le "Flow Nozzle" ? C'est sur ce point que je travail... Merci beaucoup pour vos partages. Je ne m'étendrai pas plus sur YT mais si il était possible d'échanger avec vous sur une messagerie plus personnel, ou par email je serais ravis de faire votre connaissance :) Salutations, Claude
@@claudedicesare951 Bonjour Claude et @MirageC : bravo pour cette super invention et conseils technologiques. En tant qu'ancien automaticien (presque qu'à la retraite) j'aimerais beaucoup contribuer, tester et suivre vos avancées.
@@maconly34 Bonjour Maconly34, merci pour cette proposition. De mon côté, je recherche et développe pour ma petite startup qui fabrique des outils numériques. Si vous avez une solution pour communiquer en M.P. je suis partant. Salutations :)
Another great innovative discussion!! One thing I would like to have seen is a mention of the fact that lead screws (maybe ball screws too, but I don't know much about them) can vary as to the number of starts for the screw, commonly anywhere from 1 to 4. The more starts the fewer rotations needed for a given vertical displacement, but I'm guessing the easier they are to back-drive. I'm curious if you know any other limitations or advantages to the number of starts, such as wear, or runout of the screws? (Oh yeah, obviously you must match the nut to the screw in number of starts.)
You are correct, the more starts on a lead screw with same thread pitch will provide more travel distance. This allows to have low angle of threads will maintaining good travel distance. It helps avoiding the back driving. spool3d.ca/news/lead-vs-pitch/
I guess an alternative to the extra 2 wings is to do only 2 of the ball bearing magnet setups, and have a V shape channel at the third location. The V will prevent rotation and still allow rolling in 1 direction. The other two ball bearings will be able to take up the remainder of the wobble.
Only 1 is a V shape. Other 2 are the ball bearing on magnet sliding joint. Rotation of the unit will be constrained but it will still be able to slide/wobble in the plane perpendicular to Z
@@benharris4436 Yeah, my thought was that 1 V shape channel on each of the 3 Z axis will form a kinematic mount (across all 3 Z axis). But the points are not rigidly connected to the ball screw or the print bed due to the other 2 magnetic ball joints on each Z axis. So it probably won't be a kinematic mount. But through the V shape channel, you create a "harder" pivot point compared to the magnet alignment. This could be worse if the ball nut hangs compared to the magnet coupling. I don't know, I'm not convinced by the concept. I'd rather use belts.
I have though of that at first, but as @phibel mentions here above, that V grooved ball bearing will act as an off axis pivot point potentially inducing rotation deviation in case of wobbling.
@@MirageC I had a similar idea to your approach, also using ball bearings. I was going to use two cylindrical pins embedded into the print for the bearing to run in. Those units can be arranged 180° across from each other and will also constrain rotation. However you need a second stage offset 90° , so it may be a little less compact. I might try it if I have the time to build a printer at some point.
If you're brave enough (and I definitely don't encourage it) you can do like I did and hand bend them very carefully back, they're sufficiently soft to be bent by hand at least on 400mm size, it just has to be veeeery slight bends at a time, always checking against a known very flat surface to see the results. It's easy to make it worse tho, specially applying the force on the wrong spot.
Awsome video! Informative and also entertaining, that‘s how it should be. Which 3d printing material have you used, there you have glued in the magnets?
There is nothing wrong with belts. Use them if you like. However screws being a metall part, they can widthshand high temperatures much better. For example the best belts at the moment are made by Gates - they are rated for 135c, but they do tend to wear out faster at elevated temperatures nevertheless. Also i wonder what happens if your z belt ever fails? Does one side of the bed just fall down until it jams on it's rail's thrust bearing?
@@dreamcat4 You can pretty easily calculate the amount of gearing required for the motors to hold the weight of the bed when turned off so your bed doesn't droop if you loose power or something. It is my opinion as well that if you've got this many motors you might as well go with belts since I don't really see many benefits for the ball screws.
@@dreamcat4 It is possible to isolate the belts from the heat (in some cases). And yes, belt failure would have your heatbed drop. However since belts are so dirt cheap and easy to replace, and dont have any of the problems associated with ball screws, I believe a well-designed Z-axis-system using belts (maybe more than one?) is superior in all aspects to a ball screw. It has to be overengineered so flexibility-related issues don't come into play, but it all pays off in my opinion. Z-wobble is just fucked man.
@@youngtschakaloff Thanks you your replies here. I think in response to your points... it seem like a difficult problem to thermally isolate z axis belts AND seal the chamber for higher temperatures. Compared to (for example) sealing away belts for the other x-y axis. Your other point about belts being cheap to replace: well that is only really true for regular belts. However the hi temp belts are twice the price. To you can immediately double that. And of course then the more belts you need to replace, that difference in cost is going to eventually become more significant at some point. Doubling the belt width also going to double the price of replacing them too. So it all stacks up. And depends entirely on the specific configuration. But more so the MTBF for belt failure become exponential past a certain point when increasing the temperature. And getting to around or above the rated spec. Therefore it's more difficult to know ahead of time designing the printer how durable the belts will actually be in practice. I.e. for a given target temperature that you are trying to hit (that is above 100c). Well... by comparison ball screws don't seem to have the same set of problems / trade offs. I am willing to accept the fact there are other possible issues with ball screws. Just that belts are not always so perfect either. It just remains very unclear to me what the actual other dealbreakers still are with ball screws, after seeing this wobble joint solution video. And the performance / results of it as demonstrated.
@@dreamcat4 yes, great points you made there. You see, I talked to many people about the best kinetic system (cartesian, delta or coreXY) and it seems as if coreXY is the best, except for if you have Z-wobble issues. So I decided to construct and build my own printer. It is still an unfinished project but it will be a CoreXY with linear rails on all axis as well as worm gears and belts for the Z-movement. After putting the Z-system together and making everything work, I will isolate the whole print area and maybe make a custom watercooled hotend. That way no components, including belts are facing any heat. I am sure that this will give the best printing results for the money, all things considered (heated chamber too). I am planning to make a video when its all done but until then, there is some months of work left. Now, you got me thinking. Would it maybe make sense to use chains instead of belts? It's just that I don't think that noise ever was an issue for the Z-axis. Just an idea.
This is a video I didnt know I needed to see. I had an issue with my cnc router where the Z axis was off by up to .020 TIR along the entire work area. I have bought linear rails to replace the rods. But now I will check the ball screws as well.
by the very nature of publishing it, it can't be patented - its now "prior art" that will also prevent other patenting the same thing (but ofc someone will change the font and get a questionable patent)
patents are an hinderance to progress. IF reprap was patented, what would have happened to the 3D printing community you think? Many stupid patents exists, for instance, the heated enclosure is patented... I could be partially fine with a patent offering a manufacturing method, or a non-trivial innovation, but many patents are simply preventing concepts from being used by others. IE : heated enclosure. Did they invent the oven? no. But using it for 3D printing is enough to be granted a patent it seems... Open source is a constant source of innovation. See how Arduino changed stuff. Reprap too. Openbuilds. Or more recently how risc V is changing the micro controller ecosystem.... So thank you to the creator for his video, and documenting it. Indeed, open source is nothing without documentation!
This isn't actually new: Roll-in-Cage (RNC) isolation has been used in civil engineering to decouple the lateral movement of foundations from the structures they support in quake-prone areas for decades. The first ever suggestion (on record) of supporting structures with spherical bodies dates all the way back to 1906.
@@AudreyRobinel I believe the FDM 3d printing process was already patented long ago. I'm sure a google search or two would turn up some patent documents. I think reprap was only able move forward with things because the patent ran out but that is just speculation on my part.
@@K9Megahertz that was my point! When the patent was active, not much happened in the consumer 3D printing space. When the patent expired, a new industry emerged. 3D printers went from enormous expensive machines to small, cheap and easy to use devices... Slicers got amazing, with plenty of amazing free software to back things up, such as marlin, octopring, prusaslicer, cura, etc...
What a great idea ! I can feel all the fun you had creating these magnetic sliders. By the way, ton accent Québécois est à peine perceptible. Excellente présentation. J'aime bien. Je m'abonne.
That was an elegant solution to a well explained problem. In a dusty environment (mine) I would mount the lower disk's surface proud of its plate and then extrude a shroud from around the upper disk to the lower disk to keep the system clean. A plastic oldham coupler would probably fail through plastic deformation of the surfaces because of the static load, apart from work wear. Cross drilling 3 steel disks clamped together should work. Multiple balls per channel would ease the static load. The lost nut motion due to it partial rotating that you illustrated was worth several thou of linear error. I am using a double hinged linkage from the nut to the track truck to allow for all possible deviations of the screw to the track distance and stopping nut backlash .
so now i know why the leadscrew nuts on my printer are meant to be somewhat loose. Always tempted to tighten them down. I know much better now. Thank you. The animations were really good and helped me understand.
This was actually right in my wheelhouse. Very enjoyable. Really like the presentation of a problem, and then the demonstration of a solution. This is excellent. I might share it with a friend of mine, and am for sure sharing it with my kids.
Dude - Amazing video! One of my favourite TV shows is "How it's made" - I have watched countless of episodes and it surprises me that I have never heard about a ball screw before. Watching that animation of how it works is so satisfying that I think I will have to buy one for myself just to fidget with 😂
Also. Can I say excellent video. After a career in mechanism and mechanical design, I can say, based on graduates that I saw, universities just don't include this sort of detail consideration. Keep it up.
Clever and inexpensive workaround! Once the steel balls wear through the plating on the magnets, the exposed magnet will quickly oxidize and MAY give a rough rolling surface and could change the dimensional spacing. I look forward to a follow-up video!
Great job! I didn't think of magnets immediatly, I think I would have but the ball in the middle definetly not 😂 I will try this in future designs of 3D printers!
Just discovered this video and immediately subscribed to tour channel; I am a hobbyist aiming at building a precision heavy cnc, so as costs are closely looked at since it is a hobby activity, your shared explanation is very valuable for peoples like me. Many thanks!!
Amazing work and equally amazing job on making the video. Its pretty rare these days that Ill sit through an entire video like this without skipping ahead but you nailed the content and kept my ADHD focused lol.
Second time watching this video. This is very cool. I don't have a 3D printer but I like learning about them. The problem solving method you used here is something I can take note of for possible use on different types of machines I may build in the future. Thanks for sharing this really cool solution!
great video. Another option to eliminate back driving is to use a counterbalance system with pulleys and weights. You can run higher accelerations/speeds and reduce friction. It will also make Z more responsive overall
Making cheap components accessible for hobbyists while fixing their downsides... this is video is everything what is RIGHT about 'youtube blogging' and restores my faith in humanity!
Video is good agree but this comment is a bit too much of praise ,🤣
@@UncleF115 let's look at the content of the comment: the comment shows adoration for a yt creator for showing how to take a cheap component into a more capable version while saving a significant amount of money, the original comment also alludes to the yt community as a whole in contrast to this video. I think even the most fervent fan of many yt channels would admit that the majority of yt content could be considered frivolous, pedestrian, non helpful, etc. Nothing against that, even within the 3d printing community engineering videos with humor and appreciable understanding and skill is not the majority.
Most content in this space consists of simple reviews (to the point of barely being more then a spec sheet read or outright advertising-again not hating just stating the environment. Creators have to eat) or "look at my thingiverse model I printed" (again no judgement, it has its place especially for children or people new to the scene-it's how I introduced my wife to the hobby).
The original comment could be perhaps energetic but people get excited (I know I am) at what possibilities open up on the diy scene. The person was far more concise than I have been for sure.
If you have any videos that you feel perhaps are more worthy of praise please share especially if they are aimed at the making/engineering scene. I assume from your username and previous comment you may see yourself as a person with more discerning taste. Please do share any links or channels worthy of general praise.
How "accessible" are they, though?
@@lloydrmc literally 20 bucks he explained this in the first minute of the vid
@@lloydrmc ’סעה🍓🥬🍒🥬🍒🥦🍒🥦🥕🥑🥑🍒🍒🍎🧄🍆🧄🍆🍑🍑🍏🧁🍼🎂🥛🍯🍯🍮🍮🍩🍨🍦🍦🍦🥮🍦🍦🍮🍮🍭🍮🍮🥡
This is an elegant workaround fix : fair amount of thinking, efficient, compact...and it seems pretty cheap. Impressive ! Congratulations !!!
In very few occasions I'm grateful for the all mighty RUclips algorithm. Today was one of the them!
This is what engineering is all about, see a limitation and find the best solution to delimit within the working constraints.
This is a nice solution that allows one to save a ton of money and print parts that are good enough for most jobs, I'm impressed.
As my dad says "what you don't know will kill you". I had been planning on making a printer with the $20 ball screws, but I didn't understand the issues I could encounter from them! Thanks for this very interesting and informative video!
I wanted to build a cnc with them. Sheeesh the decouplers are a must. Good old Oldham could work, but ball bearing wing thing like in this video is even better!
You've made a "flexure", very cool to see someone arrive at this idea all on their own!
I need to say, that I am absolutely super impressed with the solution to deal with using cheap parts, also the fact you share this with the hobby community it is just outstanding. I am about to embark on a high temperature version of your DIY for carbon fibre printing. So excited. Once again ingenious solution to the problem.
As an Electrical engineer, I have never admired mechanical engineering as much as I did here. Great video
Pure quality content! Thank you!
yeah, totally underrated channel
@@TeaObvious - I agree, totallyObviously a great channel which deserves all the best
(And it seems to be working - RUclips promoted this to me on the frontpage :ø)
I’m an engineer from the U.K. You have become my favourite channel! Beautiful solution! Ur rite, magnets are cool!
that was a great watch. the cad alone was impressive. Let me know if you need some parts machined lol
Where are you located?
@@andinbriwel1092 toledo ohio
OH MY thats high quality content and i thought about the same but after i looked up the price i threw the project in the thrash bin... im amazed that u solved the problem soo nicely Congrats!
This approach is quite smart! I worked on an approach to solve this when still using home depot threaded rods during the reprap times and it never really worked well. This approach is brilliant in comparison! And I even like the name :D
Wow, you just killed National Geographic! The quality of this video is better than the most professional documentaries i have seen before.
Thanks!
Thank you :)
A serious piece of engineering. Awesome project.
You gave me a perfect idea to what to use as a lubricant, hmm, I thing that turbojet hybrid ceramic bearings are the best thing to test this kind of application for, hmhmhm, MaPle SYroUp.
excellent insights! thanks a lot! 👏😎
Mars!! Nice to see you here :)
This is INSANE! Thanks man! Learned a lot here!
Thank you for this!
I ordered lead screws just 2 days before this video and I just cancelled them immediately upon seeing this video👍
Wow. A new standard in RUclips vids related to 3D Printing!
Interesting implementation of the Oldham coupler. Pretty nice machine and build, following.
This is madness. With closed loop control we can ignore all this high precision. But this is also a really smart solution, nice work!
Wow, fantastic work, an elegant outside-the-box design solution! You won my subscription with this video!
Excellent WORK
Thank you for the clear, detailed explanations and their implementation in Fusion 360.
As usual, you have made an excellent analysis and found the right solutions for the use of inexpensive ball screwdrivers.
Bravo et merci pour ce partage.
un cousin des alpes Françaises ...
Merci cousin! :) Bien le bonjour à toi et à ta famille :)
What a genius resolution! I’m am always I’m awe of the work that you produce, you are incredibly inspirational. Thank you!
Ok, I'm no engineer, but that was freaking awesome as far as the design and the implementation is really straight forward to implement. You have a new subscriber.
Thank you David 😊 I do have a new solution that is a lot more performant than this one. you can see a preview here on my channel dedicated to HevORT : ruclips.net/video/kpUDHc5eknM/видео.html
Great video, great analytics, and innovation to solve a problem! About ball and lead screws: They are more accurate and last much longer if they don't support weight, such as the Z axis. This should be done with a counterweight or counter spring that reduces the force on the screw to a few ounces. Even massive milling machines with massive ball screws use big counterweights or their screws wouldn't last a month. It could introduce backlash, but that's another problem for you to defeat!
I like the counter weight idea... make the counterweight big enough and now the ball bearing is pulling down the bed, instead of a ball bearing and magnet to support the plate, a simple set of strings might be enough. Or invert the whole thing, instead of having the bed pushed up, have it pulled up (suspended on the ballscrews.... no counter weight, same results
Very cool solution. What I did with my z-wobbly screws was set them free. They spun in ball bearings at the top. I simply removed those bearings. Now my cheap screws end in the larger hole that used to hold the bearings. This solution only works because my build plate travels on four hardened locating rods which your hevort does not have and which would kill your planar printing concept.
Very nice, clever, and fun approach! I would also consider simply removing the support bearing at the top of the ball screw, and replacing it with a relatively loose fitting plastic/bronze bushing. The clearance (looseness) should be large enough to enable the imperfect ball screw to freely 'wobble' radially, without transmitting that wobble to the platform. It's not a fun, elegant solution, but it may work very well. Thank you -- I watched your entire video :)
I don't always comment on RUclips videos but this is probably the best 3D-printing related video I have ever seen. Cheers!
Wow! thank you, your comment means a lot to me! Best Regards
Man, this is an incredible video. Thanks for your work, it was a pleasure to watch.
Thank you for sharing results of your hard work, trials and errors! This will make hobbyist world greater place
Yet an another amazing video, please upload them more often. As a fellow engineer I love the designs, personally I would have just gone with a good quality lead screw if price is an issue. Cheapo ball screw additional to the wobble might have deviations in pitch as well. Also experience showed me simple always wins at long term use.
All screws have pitch errors. It just depends on your accuracy requirement. When the screws are rolled short term deviation is probably quite small. It will depend on the last couple of turns of the tooling, which one imagines is smoothly channelled.
Absolute error is something else. That depends on how accurately the rolls are made. For short lengths it is probably irrelevant but for 2 mtr lengths a cheapy could be interesting. If you can accurately measure those distances any error can be accounted for in the software.
The real warthog in the workshop is temperature variation. With a daily domestic shed variation of 30 degrees c. here in Oz the 2 mt shaft varies by some 0.4 mm. or 15 thou.
Because of this and I am interested in absolute accuracy, just because I can be, I am looking at measuring displacement independently of the drive system.
Exceptionally Brilliant. I could not figure out what was causing the periodic wall deviations, but 12:10 explained exactly what is happening, as I held a sample up to my Z-axis and sure enough, an exact match to my thread pitch; also, I could see the upper unsupported section of the leadscrew wobble in an orbit. Thank you.
It's a good stop gap, using magnets to alleviate the z-wobble associated with inexpensive rolled ball screws.
This's what engeneering means!! Congratulations from your new subscriber, from Uruguay.
Excellent Idea, I'll add to my design, it's amazing!
YAY for people who think outside of boxes!
Kudos to you 👍🏼
revolutionary! Youre on another level with this stuff.
You are great inventor. Thank you for being such a active part of community. May your work help humanity achieve great things.
Elegant solution, and great explanation. Thanks!
Excellent engineering design and solution! Reminds me of the experiments of earth quake on a tall building. The horizontal movement can be damped by putting a roller underneath the tall object.
Excellent, i know why i subscribe. I like your work and your big level ! Big thank from France
Bien le bonjour à toi cousin Français :)
@@MirageC Quelle surprise ! Merci pour votre réponse :) Je suis attentif devant votre travail qui m'inspire pour un
prototype d'imprimante innovante.
Pour moi, la voie que vous avez choisi à travers les duets/servo moteur/X Y Core, etc... et une recherche de performances maximales d'une FDM.
Je pense que vous pouvez encore améliorer vos performances en allégeant vos Axes X et Y avec du Composite de type "Igus" par exemple. Il permet aussi de diminuer les frictions, le bruit à haute vitesse et d'utiliser des lubrifiants secs (silicone ou PTFE). Bref...
La limite est dans le "Flow Nozzle" ?
C'est sur ce point que je travail...
Merci beaucoup pour vos partages.
Je ne m'étendrai pas plus sur YT mais si il était possible d'échanger avec vous sur une messagerie plus personnel, ou par email je serais ravis de faire votre connaissance :)
Salutations,
Claude
@@claudedicesare951 Bonjour Claude et @MirageC : bravo pour cette super invention et conseils technologiques. En tant qu'ancien automaticien (presque qu'à la retraite) j'aimerais beaucoup contribuer, tester et suivre vos avancées.
@@maconly34 Bonjour Maconly34, merci pour cette proposition. De mon côté, je recherche et développe pour ma petite startup qui fabrique des outils numériques.
Si vous avez une solution pour communiquer en M.P. je suis partant.
Salutations :)
wow, as a retired machinist I know ball screws very well but this way past my pay grade. awesome work sir.
Thank you! This means a lot to me!
(@12:50) ha loved it, C- 20 bucks good buy lol! Great video ;)
Wow, incredible "maker" engineering! Super inexpensive solution to a complex problem. Bravo.
This is unbelievable content dude. Keep it up.
Amazing engineering around constraints. Subbed.
Another great innovative discussion!! One thing I would like to have seen is a mention of the fact that lead screws (maybe ball screws too, but I don't know much about them) can vary as to the number of starts for the screw, commonly anywhere from 1 to 4. The more starts the fewer rotations needed for a given vertical displacement, but I'm guessing the easier they are to back-drive. I'm curious if you know any other limitations or advantages to the number of starts, such as wear, or runout of the screws? (Oh yeah, obviously you must match the nut to the screw in number of starts.)
You are correct, the more starts on a lead screw with same thread pitch will provide more travel distance. This allows to have low angle of threads will maintaining good travel distance. It helps avoiding the back driving. spool3d.ca/news/lead-vs-pitch/
Excellent and ingenious solution. Good sense of humor.
I guess an alternative to the extra 2 wings is to do only 2 of the ball bearing magnet setups, and have a V shape channel at the third location. The V will prevent rotation and still allow rolling in 1 direction. The other two ball bearings will be able to take up the remainder of the wobble.
I do not think that will work, the V shape channels on the 3 Z axis will result in a kinematic mount? ruclips.net/video/ekUkI9iWUoM/видео.html
Only 1 is a V shape. Other 2 are the ball bearing on magnet sliding joint. Rotation of the unit will be constrained but it will still be able to slide/wobble in the plane perpendicular to Z
@@benharris4436 Yeah, my thought was that 1 V shape channel on each of the 3 Z axis will form a kinematic mount (across all 3 Z axis). But the points are not rigidly connected to the ball screw or the print bed due to the other 2 magnetic ball joints on each Z axis. So it probably won't be a kinematic mount.
But through the V shape channel, you create a "harder" pivot point compared to the magnet alignment. This could be worse if the ball nut hangs compared to the magnet coupling.
I don't know, I'm not convinced by the concept. I'd rather use belts.
I have though of that at first, but as @phibel mentions here above, that V grooved ball bearing will act as an off axis pivot point potentially inducing rotation deviation in case of wobbling.
@@MirageC I had a similar idea to your approach, also using ball bearings. I was going to use two cylindrical pins embedded into the print for the bearing to run in. Those units can be arranged 180° across from each other and will also constrain rotation. However you need a second stage offset 90° , so it may be a little less compact. I might try it if I have the time to build a printer at some point.
I just realized that you are French Canadian!impressed with the printer and your speech skills :)
I am a frog ! :)
Fantastic!!! A similar solution could be also implemented on my ender 3 that has a wobbly Z rod!!
If you're brave enough (and I definitely don't encourage it) you can do like I did and hand bend them very carefully back, they're sufficiently soft to be bent by hand at least on 400mm size, it just has to be veeeery slight bends at a time, always checking against a known very flat surface to see the results. It's easy to make it worse tho, specially applying the force on the wrong spot.
Haha buy a new rod
@@REDxFROG it's also probably bent if it's the same quality
This is very clever, excellent problem solving with limited tools. Bootstrapping precision with off the shelf components is no easy task
Awsome video! Informative and also entertaining, that‘s how it should be. Which 3d printing material have you used, there you have glued in the magnets?
The material used for the gray wings (glued) is Extrudr Greentec Pro Carbon Fiber.
Came for printer improvements, stayed for accent 👌🏼 seriously though, this fix was pure genius and you need way more subscribers
Thank you! :)
Why not have a belted Z axis? I have it on my Pro 3D V-King and I get no Z wobble + its cheap
A wonderful video and beautiful engineering design!
There is nothing wrong with belts. Use them if you like. However screws being a metall part, they can widthshand high temperatures much better. For example the best belts at the moment are made by Gates - they are rated for 135c, but they do tend to wear out faster at elevated temperatures nevertheless.
Also i wonder what happens if your z belt ever fails? Does one side of the bed just fall down until it jams on it's rail's thrust bearing?
@@dreamcat4 You can pretty easily calculate the amount of gearing required for the motors to hold the weight of the bed when turned off so your bed doesn't droop if you loose power or something. It is my opinion as well that if you've got this many motors you might as well go with belts since I don't really see many benefits for the ball screws.
@@dreamcat4 It is possible to isolate the belts from the heat (in some cases). And yes, belt failure would have your heatbed drop. However since belts are so dirt cheap and easy to replace, and dont have any of the problems associated with ball screws, I believe a well-designed Z-axis-system using belts (maybe more than one?) is superior in all aspects to a ball screw. It has to be overengineered so flexibility-related issues don't come into play, but it all pays off in my opinion. Z-wobble is just fucked man.
@@youngtschakaloff Thanks you your replies here. I think in response to your points... it seem like a difficult problem to thermally isolate z axis belts AND seal the chamber for higher temperatures. Compared to (for example) sealing away belts for the other x-y axis. Your other point about belts being cheap to replace: well that is only really true for regular belts. However the hi temp belts are twice the price. To you can immediately double that. And of course then the more belts you need to replace, that difference in cost is going to eventually become more significant at some point. Doubling the belt width also going to double the price of replacing them too. So it all stacks up. And depends entirely on the specific configuration. But more so the MTBF for belt failure become exponential past a certain point when increasing the temperature. And getting to around or above the rated spec. Therefore it's more difficult to know ahead of time designing the printer how durable the belts will actually be in practice. I.e. for a given target temperature that you are trying to hit (that is above 100c). Well... by comparison ball screws don't seem to have the same set of problems / trade offs. I am willing to accept the fact there are other possible issues with ball screws. Just that belts are not always so perfect either. It just remains very unclear to me what the actual other dealbreakers still are with ball screws, after seeing this wobble joint solution video. And the performance / results of it as demonstrated.
@@dreamcat4 yes, great points you made there. You see, I talked to many people about the best kinetic system (cartesian, delta or coreXY) and it seems as if coreXY is the best, except for if you have Z-wobble issues. So I decided to construct and build my own printer. It is still an unfinished project but it will be a CoreXY with linear rails on all axis as well as worm gears and belts for the Z-movement. After putting the Z-system together and making everything work, I will isolate the whole print area and maybe make a custom watercooled hotend. That way no components, including belts are facing any heat. I am sure that this will give the best printing results for the money, all things considered (heated chamber too). I am planning to make a video when its all done but until then, there is some months of work left. Now, you got me thinking. Would it maybe make sense to use chains instead of belts? It's just that I don't think that noise ever was an issue for the Z-axis. Just an idea.
This is a video I didnt know I needed to see. I had an issue with my cnc router where the Z axis was off by up to .020 TIR along the entire work area. I have bought linear rails to replace the rods. But now I will check the ball screws as well.
Такую же вещь можно сделать и на винтовой подачи
Fantastic accomplished. Well done.
Thanks for sharing your experience with all of us👍😀
This needs to be patented. Even if you make it public, it's necessary to prevent some Jackass from doing it and making money from your hard work.
by the very nature of publishing it, it can't be patented - its now "prior art" that will also prevent other patenting the same thing (but ofc someone will change the font and get a questionable patent)
patents are an hinderance to progress. IF reprap was patented, what would have happened to the 3D printing community you think? Many stupid patents exists, for instance, the heated enclosure is patented... I could be partially fine with a patent offering a manufacturing method, or a non-trivial innovation, but many patents are simply preventing concepts from being used by others. IE : heated enclosure. Did they invent the oven? no. But using it for 3D printing is enough to be granted a patent it seems...
Open source is a constant source of innovation. See how Arduino changed stuff. Reprap too. Openbuilds. Or more recently how risc V is changing the micro controller ecosystem....
So thank you to the creator for his video, and documenting it. Indeed, open source is nothing without documentation!
This isn't actually new: Roll-in-Cage (RNC) isolation has been used in civil engineering to decouple the lateral movement of foundations from the structures they support in quake-prone areas for decades. The first ever suggestion (on record) of supporting structures with spherical bodies dates all the way back to 1906.
@@AudreyRobinel I believe the FDM 3d printing process was already patented long ago. I'm sure a google search or two would turn up some patent documents. I think reprap was only able move forward with things because the patent ran out but that is just speculation on my part.
@@K9Megahertz that was my point!
When the patent was active, not much happened in the consumer 3D printing space. When the patent expired, a new industry emerged. 3D printers went from enormous expensive machines to small, cheap and easy to use devices... Slicers got amazing, with plenty of amazing free software to back things up, such as marlin, octopring, prusaslicer, cura, etc...
Outstanding work sir
It’s good to see real engineering & 3D printing working hand in hand😱
What a great idea ! I can feel all the fun you had creating these magnetic sliders. By the way, ton accent Québécois est à peine perceptible. Excellente présentation. J'aime bien. Je m'abonne.
Merci :)
Good night.Have you tried to use 2 springs? There are some printers with springs attached to brass nut
How does this guy only have 12k subs? Excellent content, excellent presentation!
Great video and so informative on how the HevORT works!! I now understand how ball screws work and learned about the wobble wings!!
That was an elegant solution to a well explained problem. In a dusty environment (mine) I would mount the lower disk's surface proud of its plate and then extrude a shroud from around the upper disk to the lower disk to keep the system clean.
A plastic oldham coupler would probably fail through plastic deformation of the surfaces because of the static load, apart from work wear. Cross drilling 3 steel disks clamped together should work. Multiple balls per channel would ease the static load.
The lost nut motion due to it partial rotating that you illustrated was worth several thou of linear error. I am using a double hinged linkage from the nut to the track truck to allow for all possible deviations of the screw to the track distance and stopping nut backlash .
so now i know why the leadscrew nuts on my printer are meant to be somewhat loose. Always tempted to tighten them down. I know much better now. Thank you. The animations were really good and helped me understand.
I really need to know where that mini Arbor press is from??!! Awesome video very informative also!!
I like the thought process and the final solution is elegant
littel long to see where you want to go... but you and your work are amazing..... you have done incredible works: thank you a lot.
regards;
This video was awesome. This is the kind of stuff I like doing. You just caught yourself another subscriber. Keep it up!
This was actually right in my wheelhouse. Very enjoyable.
Really like the presentation of a problem, and then the demonstration of a solution. This is excellent. I might share it with a friend of mine, and am for sure sharing it with my kids.
I just faced this problem and thinking about solving it without extra money. Your project is brilliant! Well done
as a engineer ( in a different field ) i can truly appreciate the work you have done here, i'm not often impressed but today i was. good job sir.
This is absolutely amazing! Instant sub. Not only was your idea very intelligent, but the entire presentation was great. 18 minutes flew by.
Thank you! :)
Great video and exploration. I didn't know what the wobble wings were for until now
This has taught me so much! This has saved me from so much time and money trying ideas myself!
Dude - Amazing video! One of my favourite TV shows is "How it's made" - I have watched countless of episodes and it surprises me that I have never heard about a ball screw before. Watching that animation of how it works is so satisfying that I think I will have to buy one for myself just to fidget with 😂
Thanks for the comment! Glad you liked it :)
Also. Can I say excellent video. After a career in mechanism and mechanical design, I can say, based on graduates that I saw, universities just don't include this sort of detail consideration. Keep it up.
Clever and inexpensive workaround! Once the steel balls wear through the plating on the magnets, the exposed magnet will quickly oxidize and MAY give a rough rolling surface and could change the dimensional spacing. I look forward to a follow-up video!
Yes, magnets plating does wear out after time. This is what i am working on now: raw.githubusercontent.com/MirageC79/HevORT/master/images/DSC00123.jpg
I encountered the same problem on my tinym, which I decided to use a 1204 ball screw. For sure I'll implement your idea on that! Thanks man!
Thanks man! Just done magnetic coupler for my cheapo ender but without additional magnets. But now i need second mount point for lead screw
All your content blows me away. Amazing knowledge and creativity.
Thank you!
Amazing Work. That is so easy, cheap and perfectly. You must get an Award😎
Your contend are very professional
Thank you Dennis.
An Amazing solution which I am sure some company out there needs it but totally unaware a solution exists
You are a very resourceful person. I am encouraging myself to remodel my printer with the tips that you teach
That's what I call Quality Content leading to somewhere where people actually can learn and advance. Thank you very much!
Brilliant idea, I wish I could like it more than once. You've just got yourself another subscriber 👍
Cool! Love the "not fall of the dance floor" line!
Great job! I didn't think of magnets immediatly, I think I would have but the ball in the middle definetly not 😂
I will try this in future designs of 3D printers!
Just discovered this video and immediately subscribed to tour channel; I am a hobbyist aiming at building a precision heavy cnc, so as costs are closely looked at since it is a hobby activity, your shared explanation is very valuable for peoples like me.
Many thanks!!
Thanks for an awesome investigation! It's time to look for some cheap ball screws now!
Amazing work and equally amazing job on making the video. Its pretty rare these days that Ill sit through an entire video like this without skipping ahead but you nailed the content and kept my ADHD focused lol.
incredible sir. high quality to the point i am speechless.
this is brilliant and really encompasses what it means to be a hobbyist and a maker.
Second time watching this video. This is very cool. I don't have a 3D printer but I like learning about them. The problem solving method you used here is something I can take note of for possible use on different types of machines I may build in the future. Thanks for sharing this really cool solution!
Wow! Great thinking, problem solving, and creativity!!
great video. Another option to eliminate back driving is to use a counterbalance system with pulleys and weights. You can run higher accelerations/speeds and reduce friction. It will also make Z more responsive overall
this is a great video! Thanks for the assistance. I love that you went through all the steps as you worked your way to a solution.
Glad you enjoyed it!