@@jakeengland1430 - delta printers have the advantage of no Z banding from the bed moving up and down. Therefore, for the price you can get extremely high quality prints from a delta.
I don’t know if it will work but the process is amazing. If you are enjoying the challenge, go for it. It’s not about it being perfect right away it’s about innovating and making it better.
@@properprinting this looks very promising for light gantries (with carbon rods?) and yes you should definitely carry on with the project. Towards Open source/commercial frame-robot type things? Regarding the speed-boat-race me (and the mrs) were watching and talking about engineering principals. On a core-xy printer, surely the Gantry and head need to be light but the Frame as rigid as possible regardless of weight? Some people literally bolt their frame to the nearest brick wall ;) also... . I can see the benefit of triangles for maximum strength for the mass but could frame components be square (with internal triangular reinforcement?) for greater simplicity where weight is less critical? You'd only need 1 corner model :) xx
I’m skeptical of the new system though if you push forward you’ll learn stuff that will enable you to make a better thing so it may be worth it anyway if you enjoy it. It definitely looks good!
@@properprinting one of your major strengths (and the reason I think we enjoy your videos so much) is that you embrace all parts of the design process: you come up with a cool idea and get it from your brain into a real part or machine, and then you can critique your own work and feed that back into the design. Getting the whole design process right is not at all easy and yet every one of your videos shows us more of the art. Always inspiring.
Even if this doesn't work as parts for a high-speed 3D printer, I can see this being useful as a structural component for building things like boxes and enclosures. I could see this making for a very interesting looking PC case, or as a replacement for those expensive aluminum extrusions you see on some printers (less cutting and drilling).
Now, this might not be the fast printing revolution - but it could bring us closer to a stable and economic 3d printed 3d-printer frame, which I'm all for. I'd love to see you pursue this idea further!
Hi, this is a great idea. Here in my country (Paraguay), it's really difficult to buy v-slot to make 3d printers, then you only can laser cut mdf or something like that. This idea can really have a great potential for smaller countries to promote building stuff. I would really love to see a full 3d printer frame made with this idea.
Is it 100% the only solution? No, not even close, but it is pushing design and a new approach. I think it is incredibly interesting and think there is a ton of potential here. I would say keep working on it until you hit a dead end. It is a great way to work those engineering muscles and shares a new perspective approach with the community! Great work!
it's awesome to see someone finally utilize the triange. i just cannot for my life find out why everyone use square tubes and stuff, when the triangle has the best strength of every simple geometry.. but i'd make it simpler, less modular, and i'd make it with only 2 smoothrods at 2 edges along the triangle. also i'd print it in one piece so you don't have to tighten the modules together. can't wait to see how you'll advance with this project
Well, it's not hard to guess, is it? People use square-everythings because it makes everything so easy to design in a CAD with a big cartesian raster. But you're right, that is not ideal. (And it's not like more sophisticated geometries would put the software to its limit!)
the other major reason people use squares in 3d printers is that triangular t-slot extrusions aren't nearly as common or available in as many sizes, therefore they're more expensive - and a truss like this would be even more expensive to mass manufacture
I might suggest rather than an enclosed triangle ⛛ , an equilateral Y shape. Tubular roller shafts (rather than threaded) should give a better weight to strength ratio; the receiving cups could be much smaller, and use CA glue for bonding lengthways. Have the wide of triangle down on flat surface, apply CA to upwards facing leg 'cup', drop in tube, which could be held down with a milled jig to ensure straightness and uniformity of pressure while curing. Ends of tubes can be tapped as needed. Not sure about the rest since I am only a lowly Ender 3 V2 use, but fascinating to watch your design progression.
Try designing the segments to use 1 threaded rod through the center. Make the end segments a little more stout or a plate of some kind to allow more torque from the single rod and connecting points for the joint. For the corners, make them interlocking and completely round or what ever mating shape you want the rollers to be. OR 2 corners interlocking and 1 corner designed to hold a hollow tube for the linear rail.
Or, you could loctite in some threaded standoffs (like the ones used to mount PC motherboards). I'm taking my inspiration from archery arrows - hollow tube with a threaded insert epoxied into the end.
Yes, please, continue! When you ditch the thread, the printed wheels with pressed bearings should be enough for it to run smoothly. This could very well be the next iteration of reprap project. Just use 6x1mm ALU tube and tap for internal thread, and you're golden 🙂 The beauty of it lies in the minimal required tools - just tubing cutter, sandpaper and tap. This way you can eliminate the need for any saw or dremel with cutting disc...
Great idea! The tetrahedron is the most rigid structure from geometric principles. However exploring the unknown means that a few failures are to be expected and this is part of the learning process. They usually allow to generate new ideas. Two things though: - for smooth movements you need smooth surfaces. - things that assemble will be heavier than something in one part as you need to generate extra loads to keep the parts together. I'm eagerly awaiting next episode.🤩
I like this idea, but I don't think that printing them in small segments makes much sense. The printed pieces are probably heavier and less stiff as a result of being small segments. I would recommend that you create a parametric CAD version that generates some unit length printed truss (if you haven't already). I also think that using advanced vase-mode techniques can get you maximum strength to weight ratio. I recently released some advanced vase-mode cylinders on printables for model rocketry that have built-in strengthening ribs as part of the CAD model and are printed purely in vase-mode. I can't post the link here unfortunately but the model name is "Advanced Vase-Mode Hollow Cylinders (model rocketry related)" if you search it should come up.
Your tenacity with these designs is completely insane... and sincerely inspiring. Plus yes, I would love to see you make a printer out of this system, or at least a working gantry.
Looks cool, I like that it's modular Ideas, use the hollow steel tube: 1) at the end use a long screw and glue with J-B Weld original(or your glue of choise) 2) use one of those ball joints used in RC for the stearing, insert the side of the ball into the tube, form the tube ends to make them hug the ball, add glue if posible. The threaded side should stick out and allow you to apply tension, you can choose from steel, brass or alum.
I also had the thought, for people who want formula one levels of performance you are looking at X price range but for others who are happy with the weighted version, the price is X
I would do this as a drop in replacement for a voron gantry. That way people can scale it up or down depending on their printers needs. I think a hollow steel tube would be best as you can then tap them and screw in standoffs. Keep going and make this a full build
I like the design! If it doesn't work out as a linear rail, I think this has potential for robotics as a frame. Similar to the currently often used aluminium extrusions. As you already suggested, tubes instead of threaded rods might be a good alternative. If the wall thickness of the tube is thick enough, an inner thread instead of an outer thread could be used to tension the stacked modules. This has the benefit of using a smaller thread than one would need for an outer thread. (Hope this makes sense) However, the tubes then need to be shorter as the stacked modules.
As you say at the end, maybe scaling it up would do the job As it guest bigger it gets lighter in comparison with the linear rails since they are entirely made out of stainless, where getting bigger means more still while this in the other hand have all that empty spaces
This is a win win for everyone. It's entertaining, there is much to be learned through this iterative design process, and it's inspiring for everyone to see someone think differently and to approach problems in this manner. I highly recommend you continue.
I think even if it doesn't turn out to be any better, it's still a really cool modular building system. what I would personally do is not prioritize weight, and try to focus more on improving modularity and ease of use. modular systems are jack of all trades, unless extremely specific condition, a purpose-built system at equal cost will always perform better. if you want to make it lighter you might want to look into topology optimization for your 3D printed segments.
Definitely should try a carbon fiber tube on a micro-sized track roller. I love the direction this is heading in as a concept. Sure it's not the most rigid frame in the world, but it's definitely a cool exploration!
I love the idea. I was considering something similar once. However, it is not a reasonable way forward: Tests with 3D Printed parts show, that the stiffness of any printed material is not sufficient. When you upscale it or you try to go for higher speeds, it will introduce wobble. Now you add some stiffness with your metal rods, but this again adds weight. Way easier: Buy a 2cm x 2cm carbon fiber pipe and mount a linear rail to it. The endproduct is lighter. The stiffness is increadible - even if you upscale to a large format printer. These pipes are standard parts, meaning they are affordable. What you get is greater stiffness, less weight, less effort and a concept that works for every printer size.
I think this is a great idea. You could build a "3d printed 3d printer" with this concept. And yeah, it looks freaking cool. Just imagine a printer built out of those. You just have to figure out a way to put a belt in there for xy and for z.... Well, you just put 4 of those rails vertical. Nice job man!
Thanks! Making a whole printer will be an interesting next step and I hope it will look like some miniature futuristic stage at a festival. I should print band members on the plate, I mean stage xD
@@properprinting For more inspiration I would totally look at the Rook 3D printer. It's using a printed frame. With something like this, it moves us one step closer to self replicating 3D printers in all but the nuts and bolts 😆
Do the side rails need to be extra light also? It makes sense for the moving gantry to be light but I don't think the side rails need to be light as they shouldn't have the need to move. I can definitely see where it helps to have the mounting to the side rails still be light and thus possibly needing to have a custom mounting solution
I would definitely like to see you pursue this idea/concept as this has great potential. It may ultimately prove to not be suited for fast 3d printing, but it will drive innovation in the industry and others that rely on motion systems.
Do it. Speed of one isn't important. Just do one, print the next one with it, boom double speed. Have two nozzles on the same gantry, have two gantries with two nozzles, check their combined speed. Have 1+2. Use 0.6 for filling, 0.2 for outer shell plus a 0.2 water soluable. Lower the number of corner kinds by using a square. Make the interconnecting plastic parts lighter (no extra holes) and longer (less work to assemble). Have a version of interconnecting parts that don't expose the rods - only the gantry needs that, and would make the rest stiffer. Awesome project!
This is my first comment ever. I think there might be a misunderstanding with light as possible. The static parts should be as heavy as possible. Different materials should help with dampening the vibration. Only the moving parts should be light and stiff.
True that! This reminds me of a video Tom Sanladerer once made where he filled his extrusions with sand. Maybe I can make closed segments which you can fill up xD
@@properprinting You can refer to X1C's implementation for inspiration. It may save you some time. By the way, Winston Moy did the same with a Shapeoko CNC 3 years ago. You can find it on his channel.
Yes. The static structure must be able to absorb the forces generated during acceleration and deceleration. A heavier structure will reduce the magnitude of any vibration. I vaguely remember someone suggest using magnets as dampeners. The generated force can be minimised by reducing the weight of moving parts in the first place. Another thing to watch out for is the natural resonance frequency of the moving parts. The printer's frequencies of speed changes should not overlap with it.
So right off the bat I said to myself... Stainless Hypo-tubes instead of threaded rods. For reference, I've been a design engineer in the medical device space for the past 30+ years. As for lightening... Maybe a triangular beam that is not skeletonized but has gyroid infill, printed with one side on the build plate. That'll already make it as flat as the built plate, if it doesn't warp. I like this subject and I think you're on point thinking about the next step in 3D printing. As always, Awesome work! Thanks for sharing.
Id love to see this scaled up. Doesnt seem like the best option for going fast but still could make building large printers much cheaper. For example a 500mm sized printer, would be great to see.
Really cool idea, the customizable frame design seems like it'd be really useful for things like enclosures, hope you keep iterating to see where you can go with it. Another idea would be instead of trying to start from scratch and designing your own linear rail system, maybe you could look at an existing linear rail and see what the bare minimum that you could add to it to turn it into a gantry, combining the best of both worlds (a proven effective rail design and a cheap yet lightweight extrusion to attach it to).
This is an awesome idea with excellent implementation. I think instead of buying 100% threaded rod we should try to find some smooth steel rods(think Prusa style linear rods) that can easily be threaded with a DIE. Ideally sizes that can threaded directly or by just removing the surface finish/chrome with a file/grinder/belt sander. I'm picturing some sort of bearing harness that allows you to rotate it easily to get an even finish for threading with the DIE. If it catches on I'm sure China will eventually sell cheap pre-threaded rods but until then I personally would prefer stronger smooth rods over 100% threaded. They would likely be more concentric as well.
When it comes to choosing materials for mechanical assemblies, it's important to consider the specific requirements of the application. While threaded rods can be useful for transmitting tension loads or providing clamping force, they are not typically considered to be a primary structural material. This is because threaded rods do exhibit some degree of elasticity, which can be useful for producing compression when tightened. However, this elasticity can also result in lower stiffness and strength compared to other materials like steel beams or columns. If you're looking to achieve a high level of stiffness or strength in your assembly, it may be necessary to consider using stiffer materials and threading the ends as a means of connecting components. By selecting materials that are appropriate for the specific application and design requirements, you can create a strong and reliable assembly that will perform as intended. I also wanted to kindly mention that thermoplastics may not be the best choice of material for this particular application. This is because their young modulus, or stiffness, is generally lower than that of other materials like metals. However, if you prefer to stick with thermoplastics, I would suggest considering a carbon fiber reinforced filament. This type of filament will have greater stiffness while still maintaining a low weight. Please keep in mind that fiber blend filaments may not be as strong as natural filaments, but they can still be a good option for achieving the desired level of stiffness. May I suggest you studdy the history of fully 3D printed Reprap printers before going any further ? This will save you a lot of time. Regards.
Smooth rods where just the very ends are tapped and the track rollers you mentioned would eliminate the primary flaw with rough motion. Lightweight foaming filament would reduce weight. You *might* be able to get away with 3 rollers gripping 2 rods as if it was an inverted v wheel printer which means your extruder mount would become a pretty small triangle. Since the 3rd rod is now only needed for tension some of the truss supports connected to it could be eliminated. That same 3rd rod might also become some magnet wire etc if it isn't guiding a roller. If you wanted to be completely berserk about chasing grams you could sand the rods so the cross section would be a D shape rather than an O to further shave down ~40% of the mass of the two remaining rods. You'd need to get everyone on earth to agree not to sneeze to keep it from breaking, but you'd have bragging rights for a DIY ultralight X rail without having to metal print a titanium one. Breaking away from speedboat I think the structure being preloaded under compression and multiple materials might be interesting for resonance. Tune the threaded rods like guitar strings and fiddle with plastic type / truss design / print settings for the best possible starting point prior to input shaper. Come to think of it I'm now wondering what impact shoving a guitar string down the middle of standard aluminum extrusion and tightening it would have on resonance. Horrible waste of effort or would it push resonance to a higher frequency and thus out of the way of normal printing.
Do you need a rigid rod? If just compression what about tensioning with wire or a zip tie mechanism? Truss with slots for stock zipties might provide an adequate rail and make corner fittings much much simpler. I think you can use heat to shink them & add extra tension if needed, & later acetone or epoxy to glue tensioned ziptie to truss. It be relatively easy to span zipties through the hollow of the trust to different corners for extra strength & dial in (or out) any twist. Zips come in a huge variety of sizes & materials (stainless even) up to 60inches long. Hose clamp might also work. A metal zip could take a fillet or be bent like angle iron which might be superior to steel rod by weight.
I'd love to see how far this project can go. The original open source concept for 3D printers was to make a machine that make more machines. I think you're on the right track. It's super cool what you're doing, even if it's not the lightest setup, weight is not the only factor for printing well and with speed. Consider using carbon fiber rods. Cheaply available online.
Try to replace the threaded rod with round stock and get a die to put the threads on the ends just where you need them. I think putting sleeves over threaded rod will just increase the size and weight.
Have you ever seen the Positron? It's also got some clever ideas to print even faster--the most important, I think, is printing upside down to keep the print head as low as possible, reducing the wobble of the frame. The bed is then upside-down, which seems to work just fine. Maybe some of the concepts used there could serve as inspiration for your project? Combine an ultra-light frame with a low center of mass, and a low printhead that can dash back and forth without shaking the frame apart.
hollow pipe to roll on for example filement ptfe could work great with those metal rollers but you would need much smaller diameter rods or tension wires. the problem with that idea would be if overtensioned plastic structures would be prone to deforming you could use metal pipes but the question is do you even need to tension it, how straight would it be without tension or with minimal tension from single central rod while side tracks used as rails would be fited with pipes that are not under tension you could also make universal cornet pieces in shape of a hollow wireframe cube and tension the truss with central screw trough the cube itself (a bit limited to 3 sided use with straight alan key with ball key you couls probably use more sides)
I just had an idea for an ultra light gantry and had to come back to the comments of this video. Use 2 small aircraft cables pulled really tight as the gantry. You could have bearings run along the cables for the caraige. It would require the frame to be really strong but the actual moving parts could be extremely light. There would then also be a potential to electrify the cables and use the wheels to transfer power to the hotend to reduce electrical cable mass. You would then only need fan and thermistor wires. Isolating the aircraft cables may be more trouble than its worth though.
I would first start by comparing apples to apples, your rail is at least 1 and a half times longer than the voron rail. Next, you probably dont need the threaded rod to be as thick as it is, and third, if you go with hollow rod and those bearings (which IMO you should) you can put threaded rod through it to tension it. Also (IMO) you only need one smooth rod on each side of the Y gantry, facing directly in. You could also probably get by with an ovoid or diamond shaped prism for the x gantry that only has two rods through it. That would make a big difference I think.
I'd be interested in seeing this more. I also had the idea of smooth rods with threaded ends and concave bearings to roll against them, in hopes it would be smoother. as for lighter... .i think that's an as-you-go situation. unless your printing in aluminum or other lightweight yet sturdy metal or alloy, I don't see an easy solution all around, but step by step improvements as you go. Finding that a solution doesn't work, back to design for it , or perhaps finding a lightweight metal solution to replace it with, etc.
Look into Arris Composites. They specialize in conformal, compression molded carbon fiber composites. The performance of parts manufactured with their technology is honestly ridiculous; light, stiff, and resilient as hell. And since they're mostly carbon fiber, they're also electrically conductive, so parts can behave as their own strain sensors. I've never seen anything like it. The only drawback is tooling. But it could be a game changer for high performance FDM printers.
I agree with others, that this may not work for a 3D printer (probably not, I’d say), but I think has a lot of potential as a general structural system. As noted, it could be scaled up for larger objects. I wonder how it would compare with aluminum extrusions at larger dimensions? The extrusions might me more stiff, but they get *dang* heavy in larger cross-sections. What if the panels were, say, 10cm on a side, with 5 or 6mm rods in them? How stiff would the assembly be, for how much weight per unit length? Kudos for your creativity and pushing the envelope with different “crazy” ideas! I learn something or get useful ideas from every one of your videos, this time it was the idea of pre-tensioning the assembly to avoid stripping the threads on the Al rods. I knew this is done with prestressed concrete panels to overcome the low tensile strength of concrete, but I’d never thought of it as a way to save threads in a soft material the way you used it here.
That little chuckle at 12:30 is exactly the reason I still keep making stuff. Things break, designs don't work out, unexpected problems out the wazoo, smacking my head at the wall for hours, days, weeks. But at the end i have something that give me that little chuckle and suddenly the whole journey doesn't seem that bad after all.
12:30 You look completely thrilled there! And I understand you completely: There's a satisfaction looking at your idea as something real, something you can touch, something you can use... No words for it 😁👌🏼
Even if it isn't as advanced or as light as other methods, I feel that that fact that most of the structural "filler" is 3d printed is still amazing, and it looks like the concept has potential. maybe smooth rods with threaded ends?
Go for it but.... Try to use pccf (stiffness) Things like wheels you could use nylon (abs isnt Wear resistant at all) Also instead of that you could Just get 2 aluminium smooth rods and igus fi8 bearings and youll end up with really light thing but then you cant use direct drive system with is ok with klipper and pa. I have tried aluminium rods and whole x axis was lighter than Just 2 steel smoothrods :)
I love this truss idea, while it may not be the best option for a 3d printer, I still think you should try and make it one, I'm interested to see other uses for this sort of system, where the truss structure can be taken better advantage of
One thing with many of the gantries is their tubes are hollow. I know many hobby shops sell tiny metal tubes (~3mm), you could replace the threaded rods with those and screw a bolt inside the ends to keep the stack tensioned.
I suspect the main issue there is the bearing surfaces, whereever you've got plastic rubbing against plastic it's going to wear. linear slides have ball bearings going around in a circle against a hardened piece of metal which is supposed to be ground to be really flat so that it's super accurate. I could see this more as a replacement for alu extrusion for the frame although you'd have to compare the rigidity to alu
It requires a bit more machining steps, but what about precision ground rods for the reinforcing/bearing surface? Yes you'd have to tap each end for bolting together, but it would give you a fairly smooth surface for rollers to run on And as a former live sound engineer and rigger, what about using egg and pin connections? I have no idea how you'd machine the eggs to this scale, but clockmaker's tapered pins and a tapered broach to get the fitment just right may be a starting point! Also, what about a different geometry to the truss segments, using topology optimisation for an ideal strength to weight ratio?
It's cool looking, but I can't shake the feeling that a very different design would be needed for this use case. The third rail on the gantry is dead weight, as are the ends of each modular section (fewer, longer modules would reduce the weight). I'm also wondering if just a pipe could work as a gantry. It would need something else to keep the hotend from swaying forward and back (a slot in one side?), but it might be lighter than 3 threaded rods and their 3D printed supports.
Ideas.... 1) Don't use threaded rod, instead, use hollow tube, and tap the ends. now you've eliminated a lot of weight, because you no longer have solid heavy rods, and you no longer need the heavy standoffs, because you can now screw right into the tube. 2) By using smooth hollow tube, you no longer introduce additional friction trying to mesh geared teeth that serve no purpose. small bearings should ride quite nicely on them, could even use tiny V-bearings so they lock onto the metal tubes. 3) I would suggest making different types of the trusses. you don't need screw / mounting holes on all sides, so maybe only do one, or two sides, further reducing weight 4) to keep the parts from "slipping" on the contact faces, you can add tabs on one side of the truss module, and pockets on the other, so they made, creating a more sturdy mating surface. 5) If you need additional end points for mounting, you could add a central threaded rod, and use this as either a mounting point on either end, or as a method of pretensioning the entire truss assymbly
So if you do tubing or pipe, you can see about making seize fit ends that are tapped or threaded rod to attach or mount ends too. You could also see about a small spot welded and adding a nut inside the tubing, like an m2 or
I really love this modular truss design, it may not be the absolute best for a specific use case but it has a TON of potential for a general purpose building system.
Keep going for sure mate. could you print these beams completely on a large bed, forgo inserting rods and just print with dense/strong material for the rod sections and a lightweight one for the rest of it? Could then use wheels of the same material as the rod section, which would hopefully stop them from being destroyed so quickly. Not sure if the beam would hold up to stress without proper rods inserted but would be interesting to find out if you can completely print a printer.
You might be able to use bicycle wheel spokes instead of threaded rod since they are made to be a lightweight adjustable tensioner and can be bought in a variety of lengths and materials, including carbon fibre.
Some ideas: -use fiber reinforced polypropylene for stiffness and light weight. -use stainless steel tubes and a thread former (not cutter) for the tention threads. you can also still make your design smaller and maybe give the sections some interlocking features.
If you are going to use smooth rods and need to cut threads in them anyway, I think I'd use tubing and tap internal threads in the ends. You could user larger internal diameter tubes which might add to the stiffness without increasing the weight. This was a very nice instructional video. I don't think you are a native English speaker, but your English is very good and very understandable.
Please give us more. I have a few material suggestions: Polypropylene is extremely light, but a little flexible, might be useable because of the truss structure, fairly cheap too. Carbon fiber might be an option since you want stiffness and lightness.
Ohhh, I also tried to build a FDM printed truss system about a year ago and didn't like the result. My concept was like origami, printing all 3 sides in one flat piece to optimise print speed and layer orientation. Maybe use an unextruded length of filament along the length as a pin to secure the triangles.
I once made a somewhat similar structure with carbon fiber tubes. A bit like the pre-tensioning as you did with the aluminum ones. But fixing with glue.
I was gonna suggest smooth rods and grooved rollers but they you figured that out yourself. You should also only need 3 bearings per carriage, you're over constraining them which makes movement less smooth and more sensitive to alignment. On the small scale of the voron 0 you're unlikely to find success making this lighter and cheaper, but for a large format printer where weight really starts becoming an issue and there is more material that can be removed, you make have better luck. The straightness of the rails should be somewhat adjustable by tightening and loosening each of the three rods to tension one corner more or less to pull things straight. I don't think this will fix wonky rods though so maybe try buying precision ground linear rods?
Definitely has potential! I would love to see your version of a mostly 3D printed 3D printer. I know the Mulbot and some others already exist but your design skill is just on another level
Did you ever do a comparison of the assembled portals? It seems that the truss would have a more uniform stiffness. The base gantry is going to be stronger in the wide plane than the narrow plane by a large margin. I would like to see some strength tests done on your trusses. I considered something similar for building a 4' x 8' CNC router gantry. Only using MDF instead of printed plastic. I think you should(maybe you already have) keep going on this idea. Even if you don't make something revolutionary, you might come up with something that could make building homebrew 3D printers more accessible.
I love the work and the premise of the idea, I really would like to see how it develops. Honestly though I don’t think it would be a better solution as linear rails is used in many industries. What I would like to see is how it compares in terms of price. If you can have the frame this way and only need to use small rods and few fittings that would be a cool diy 3d printer project.
It's really cool but I think one of the main flaws is the build volume you'll get out of it, if I'm correct it should be smaller than the one you printed them on, plus I'm not sure about how rigid this would be if you stuck 2 pieces together to make a long one. I really like the idea but I don't see the point other than being cool (at least for now), it's not the lightest and probably not the cheapest to build, maybe it could be a great middle ground option with enough developpement
Would love to see additional effort into this. As for the weight problem. Maybe hollowed and smooth Titanium rods with the inner diameter threaded for your endcap screws to tighten it down? Or carbon fiber rods maybe? So many options
YES! More Videos!! You are on the right track, for the scale look at how to make the trusses with hollow tube without becoming breakable. I believe you are on the right direction by making them like Lighting trusses.
None of this material is stiff. Next time print in CF Nylon for stiffness of the plastic parts and use CF hollow rods. This way you can keep the look and make the part ridgit. This kind of application is perfect for CF. If you want great performance use a CF square tube and cut holes into it. Or use thin gauge steel sheets with lots of depressions, cut outs, flares and so on.
The threaded rods are screw rods so have slanted cuts while rollers have groved that are perpendicular to mountings. That means rollers do not match squarely with threads. Result breakage of roller groves.
I think the Lego 3d printer concept is really interesting and definitely worth it. Maybe you won't get the fastest or lightest or even the most precise 3D printer out there, but I think that the best of Lego is their capacity of give a lot of freedom of creation coupled with great practicality. I think you should aim at that.
You should think about the hollow steel rods with some threaded inserts soldered or brazed to them! Or some smaller threaded rod cut to a smaller length and use that as an insert, again either soldered/brazed. For the whole process only cutting external threads on some titanium rod might be preferable though.
I'd love to see more of this project. It may not currently be cheaper, stronger, more durable or lighter but it is cooler looking and it's more customizable. I wonder if you could use pullstruded carbon rod instead of the metal rods and do a sort of tensegrity to keep the beam components together. Just a piece of kevlar or similar line through the middle of the beam with tensioners on the ends of the beam (tension in the kevlar means compression in the truss segments). Then rather than rollers, get some of the IGUS Iglidur filament and print custom bushings. Also not sure, but I suspect the truss structure could be optimized. I bet you'd need a lot less filament than you think to "truss-out" 3 carbon fiber rods running axially, at least to be good enough for a 3D printer.
You may want to try to use less hardware overall, you should try to make the pieces of the truss index with echother and have only one threaded rod running down the middle for tension.
Running the system in tension seems like a positive improvement to me. In the 24/7 printing video about lightest gantries we learn that weight isn't the only factor involved in higher acceleration using input shaping. I would not be surprised if a tensioned gantry would have a far higher input shaping acceleration than a non tensioned one.
You need the shape, not the entire rail. Print as sections and space accordingly as required for the desired regidity. Use hollow tube and epoxy threads in the ends. This also opens the door for CF tube. The result should be consideraly lighter and equally or more rigid.
I think some people have mentioned this already, but this would be great for robotics or engineering projects. It would be pretty cool to see a whole 3d printer assembled using some kind of modular components like these.
Realy nice disign! I cold imagine incresing the segment size and useing a hollow support structure like a printed tube with checker plate-ish features could increase the stiffness while maintaining a light wight, possibly enough to ditch the M3-Rods and use a pice of 1,75 filament as a rail instead. This could reduce the price and increase the accessibility while maintaining a light wight.
This is a great video. Your CAD skills are top notch, especially the ability to make that complex corner piece printable without supports. The idea itself is great and I would love to see more of it. Smooth titanium rods are definitely the way to go. As far as filament goes, here are some things to things to think of. For Stiffness, from most stiff to least, there is (Carbon Fiber Filament, HIPS, PLA, Poly Carbonate and then everything else). For Density from lowest to highest ( Polypropylene 0.9 g/cm3, HIPS 1.03 g/cm3, ABS 1.04 g/cm3, Nylon 1.06- 1.14 g/cm3, ASA 1.07 g/cm3) and then everything else). For Strength from strongest to weakest (PVA, Polycarbonate. Nylon, PLA, ASA, PETG and then everything else). For Durability from most durable to least, (Nylon, Polycarb, ASA, Polypropylene, ABS, PETG, HIPS, PVA For a part that needs to be stiff and light but not necessarily durable (Carbon Fiber Filament) For a part that needs to be light and durable but not necessarily stiff (ASA or Nylon) For a part that needs to be stiff and durable and light (HIPS) You can find more material info on Simplify3d's resources page.
I saw an interesting design concept. A single smooth rod for X and a single smooth rod for Y, with the hotend assembly mounted at the intersection of the two rods on LMXUU linear bearings contained in a single block. Each rod constrains the potential for rotation on the other rod. The Bowden-based hotend assembly just has to be mounted such that it avoids the rods. With carbon fiber rods and Delrin bushings instead of bearings it can be much lighter. There is also another design concept that uses DIN Rails. They're not the lightest things, but drill enough holes and they can be reduced in weight a lot. Of course, using a Bowden system reduces the weight to a minimum, but you hve the trade-off of needing to do longer retraction. High quality PTFE tubing can make a difference here, along with well-tuned Linear Advance. We certainly can do better than 20x20 extrusions and rollers!
I have a couple ideas that may help, first I like where you were going at the end of the video, go with steel or titanium rods and get sections as long as you can to drive cost down on the individual sections so you can cut them to length yourself and add threads only as far as you need. Next, I would remove the flats on the end of each gantry section and make a flat end piece that can go onto either end of each full length of gantry, I dont know if that would reduce stiffness but I think it would reduce or eliminate the need for supports on the individual sections further reducing print time and cost, plus if total removal does affect stiffness you can print one flat to go between each section. Final thoughts, steel is more readily available and cheaper for the masses, so if the point is cheap and mostly printed go with steel, if the point is as light as possible go with titanium.
Thanks for your suggestions! I'm currently working on a redesign and the sections can be printed without using support. I want to make the system in such way that multiple ways on tensioning can be used. This makes it easier to source parts and it's possible to decide to use different methods for different purposes. Like light for the gantry, and cheap/heavy for the frame or incredibly cheap if you just want to put something together.
If you can get your hands on titanium thin walled tube ,go that route with those steel bearings. The trick with making stiffness and a light frame is to go thin walled,and large diameters. You will find the compromise between weight and rigidity. My friends brag about their carbon bikes costing thousands, my bike is lighter and is a steel alloy frame. Thin walled ,double butted, stiff ,strong. Think bigger,thinner. Try titanium bike frame tubes.
Take a look at concrete beams that are prestressed with tension cables. What if you ran 1 very light cable through the center of of your truss then tension it? It would be very strong and light. Then you could replace the rods with something lighter just for your gantry rollers. No need for them to hold the segments together. Maybe carbon fiber tubes. Or maybe lose the rods and just 3d print a trough for bearings to run in? That assembly would be extremely light.
I think that maybe the heavier but cheaper version of the system could be experimented with in order to possibly provide people with a rail and frame system that could be used to make printer builds (or more likely rebuilds) cheaper, even if it was not used for the full frame. It does seem in its current state to be more of an alternative solution rather than an improvement right now but for some people that in itself may be compelling enough to work with it, at least in the earlier "cheap" version using steel instead of titanium.
Please continue this! I'd absolutely love to learn from what u end up doing so that I could maybe make a whole printer that's light and easy to bring around someday! Also, it might be interesting to integrate the threaded rods for tool head movement. If you did that though I'm almost certain that you'd lose the core xy system that the voron has
I think this has great potential but I think you should focus less on the light weight and focus more on the fact that you created a printable truss. I would love to be able to go to a local hardware store, pickup threaded rods etc, print some parts and be able to make a solid frame vs ordering extrusions. It looks so cool and I think you should build a printer no matter the weight! So cool
You should switch to a non threaded rod and then use a die to add threads only to the ends of the rods. It's more work, but it would roll better and wear less over time. You might be able to use something like a straight pull aluminum bicycle spoke too.
I’ve threaded rod ends before when I was making RC cars, I has to anneal the rods in an oven first and then harden them again after threading but it works.
Has this idea potential, or shall I ditch it altogether and move on with the next project?
its wayy too complicated. needs to look and be simpler!
Just go with the rule of cool just because its not BETTER doesn't mean its not stil FUN. Thats why delta printers are still a thing after all haha
@@jakeengland1430 - delta printers have the advantage of no Z banding from the bed moving up and down. Therefore, for the price you can get extremely high quality prints from a delta.
I love it! Please continue with this idea!
@@hayden9944 2 words, belted Z
I don’t know if it will work but the process is amazing. If you are enjoying the challenge, go for it. It’s not about it being perfect right away it’s about innovating and making it better.
This is not about making it better, it's about the view counts.
I think this system has enormous potential, would love to see how far you can push this
Thanks! I have this feeling too :D
@@properprinting this looks very promising for light gantries (with carbon rods?) and yes you should definitely carry on with the project. Towards Open source/commercial frame-robot type things? Regarding the speed-boat-race me (and the mrs) were watching and talking about engineering principals. On a core-xy printer, surely the Gantry and head need to be light but the Frame as rigid as possible regardless of weight? Some people literally bolt their frame to the nearest brick wall ;) also...
.
I can see the benefit of triangles for maximum strength for the mass but could frame components be square (with internal triangular reinforcement?) for greater simplicity where weight is less critical? You'd only need 1 corner model :) xx
Everything that set us closer to a fully 3d printable 3d printer is to be considered big innovation for 3d printing. Great work!!
there's printers with 3d printed linears
I’m skeptical of the new system though if you push forward you’ll learn stuff that will enable you to make a better thing so it may be worth it anyway if you enjoy it. It definitely looks good!
Thanks and a bit of skepticism is good!
@@properprinting one of your major strengths (and the reason I think we enjoy your videos so much) is that you embrace all parts of the design process: you come up with a cool idea and get it from your brain into a real part or machine, and then you can critique your own work and feed that back into the design. Getting the whole design process right is not at all easy and yet every one of your videos shows us more of the art. Always inspiring.
@@peterspencer6442 fact-check: true!
Even if this doesn't work as parts for a high-speed 3D printer, I can see this being useful as a structural component for building things like boxes and enclosures. I could see this making for a very interesting looking PC case, or as a replacement for those expensive aluminum extrusions you see on some printers (less cutting and drilling).
Now, this might not be the fast printing revolution - but it could bring us closer to a stable and economic 3d printed 3d-printer frame, which I'm all for. I'd love to see you pursue this idea further!
Yes! It might become a cheaper alternative to exstrusions
Aren't extrusions already cheaper than that? It would be a different story if those blocks are injection moulded tho.
Hi, this is a great idea. Here in my country (Paraguay), it's really difficult to buy v-slot to make 3d printers, then you only can laser cut mdf or something like that. This idea can really have a great potential for smaller countries to promote building stuff. I would really love to see a full 3d printer frame made with this idea.
+1 Morocco ! :D
Is it 100% the only solution? No, not even close, but it is pushing design and a new approach. I think it is incredibly interesting and think there is a ton of potential here. I would say keep working on it until you hit a dead end. It is a great way to work those engineering muscles and shares a new perspective approach with the community! Great work!
How can I "...throw more Input Shaper on it..." - I want that, I need that, I love round corners :-D
it's awesome to see someone finally utilize the triange. i just cannot for my life find out why everyone use square tubes and stuff, when the triangle has the best strength of every simple geometry.. but i'd make it simpler, less modular, and i'd make it with only 2 smoothrods at 2 edges along the triangle. also i'd print it in one piece so you don't have to tighten the modules together. can't wait to see how you'll advance with this project
Well, it's not hard to guess, is it? People use square-everythings because it makes everything so easy to design in a CAD with a big cartesian raster. But you're right, that is not ideal. (And it's not like more sophisticated geometries would put the software to its limit!)
you have to use what is available... lightweight triangle extruded shapes are hard too find...
the other major reason people use squares in 3d printers is that triangular t-slot extrusions aren't nearly as common or available in as many sizes, therefore they're more expensive - and a truss like this would be even more expensive to mass manufacture
I might suggest rather than an enclosed triangle ⛛ , an equilateral Y shape. Tubular roller shafts (rather than threaded) should give a better weight to strength ratio; the receiving cups could be much smaller, and use CA glue for bonding lengthways. Have the wide of triangle down on flat surface, apply CA to upwards facing leg 'cup', drop in tube, which could be held down with a milled jig to ensure straightness and uniformity of pressure while curing. Ends of tubes can be tapped as needed. Not sure about the rest since I am only a lowly Ender 3 V2 use, but fascinating to watch your design progression.
Just found your channel, man this frame looks SO sick. It’s not a failure it’s a start of something great!
@@holeechitfukinfast.4900 Agreed. Whenever a new video from him pops up, I instantly turn into a 5 year old.
Try designing the segments to use 1 threaded rod through the center. Make the end segments a little more stout or a plate of some kind to allow more torque from the single rod and connecting points for the joint. For the corners, make them interlocking and completely round or what ever mating shape you want the rollers to be. OR 2 corners interlocking and 1 corner designed to hold a hollow tube for the linear rail.
If you went with the hollow tubing, you could thread it on the inside to provide extra support at the ends.
I definitely agree, smooth rods in the linear rail portion would definitely be better
Or, you could loctite in some threaded standoffs (like the ones used to mount PC motherboards). I'm taking my inspiration from archery arrows - hollow tube with a threaded insert epoxied into the end.
Yes, please, continue! When you ditch the thread, the printed wheels with pressed bearings should be enough for it to run smoothly. This could very well be the next iteration of reprap project. Just use 6x1mm ALU tube and tap for internal thread, and you're golden 🙂 The beauty of it lies in the minimal required tools - just tubing cutter, sandpaper and tap. This way you can eliminate the need for any saw or dremel with cutting disc...
Great idea! The tetrahedron is the most rigid structure from geometric principles. However exploring the unknown means that a few failures are to be expected and this is part of the learning process. They usually allow to generate new ideas.
Two things though:
- for smooth movements you need smooth surfaces.
- things that assemble will be heavier than something in one part as you need to generate extra loads to keep the parts together.
I'm eagerly awaiting next episode.🤩
I like this idea, but I don't think that printing them in small segments makes much sense. The printed pieces are probably heavier and less stiff as a result of being small segments. I would recommend that you create a parametric CAD version that generates some unit length printed truss (if you haven't already).
I also think that using advanced vase-mode techniques can get you maximum strength to weight ratio. I recently released some advanced vase-mode cylinders on printables for model rocketry that have built-in strengthening ribs as part of the CAD model and are printed purely in vase-mode. I can't post the link here unfortunately but the model name is "Advanced Vase-Mode Hollow Cylinders (model rocketry related)" if you search it should come up.
Your tenacity with these designs is completely insane... and sincerely inspiring. Plus yes, I would love to see you make a printer out of this system, or at least a working gantry.
Looks cool, I like that it's modular
Ideas, use the hollow steel tube:
1) at the end use a long screw and glue with J-B Weld original(or your glue of choise)
2) use one of those ball joints used in RC for the stearing, insert the side of the ball into the tube, form the tube ends to make them hug the ball, add glue if posible. The threaded side should stick out and allow you to apply tension, you can choose from steel, brass or alum.
Keep going with this idea, smooth hollow rods (carbon fiber, titanium or similar) with smooth bearing would make all the difference
I also had the thought, for people who want formula one levels of performance you are looking at X price range but for others who are happy with the weighted version, the price is X
I would do this as a drop in replacement for a voron gantry. That way people can scale it up or down depending on their printers needs. I think a hollow steel tube would be best as you can then tap them and screw in standoffs. Keep going and make this a full build
I like the design! If it doesn't work out as a linear rail, I think this has potential for robotics as a frame. Similar to the currently often used aluminium extrusions.
As you already suggested, tubes instead of threaded rods might be a good alternative. If the wall thickness of the tube is thick enough, an inner thread instead of an outer thread could be used to tension the stacked modules. This has the benefit of using a smaller thread than one would need for an outer thread. (Hope this makes sense) However, the tubes then need to be shorter as the stacked modules.
As you say at the end, maybe scaling it up would do the job
As it guest bigger it gets lighter in comparison with the linear rails since they are entirely made out of stainless, where getting bigger means more still while this in the other hand have all that empty spaces
This is a win win for everyone. It's entertaining, there is much to be learned through this iterative design process, and it's inspiring for everyone to see someone think differently and to approach problems in this manner. I highly recommend you continue.
I think even if it doesn't turn out to be any better, it's still a really cool modular building system. what I would personally do is not prioritize weight, and try to focus more on improving modularity and ease of use.
modular systems are jack of all trades, unless extremely specific condition, a purpose-built system at equal cost will always perform better.
if you want to make it lighter you might want to look into topology optimization for your 3D printed segments.
Definitely should try a carbon fiber tube on a micro-sized track roller.
I love the direction this is heading in as a concept. Sure it's not the most rigid frame in the world, but it's definitely a cool exploration!
I love the idea. I was considering something similar once.
However, it is not a reasonable way forward:
Tests with 3D Printed parts show, that the stiffness of any printed material is not sufficient.
When you upscale it or you try to go for higher speeds, it will introduce wobble.
Now you add some stiffness with your metal rods, but this again adds weight.
Way easier:
Buy a 2cm x 2cm carbon fiber pipe and mount a linear rail to it.
The endproduct is lighter.
The stiffness is increadible - even if you upscale to a large format printer.
These pipes are standard parts, meaning they are affordable.
What you get is greater stiffness, less weight, less effort and a concept that works for every printer size.
I think this is a great idea. You could build a "3d printed 3d printer" with this concept. And yeah, it looks freaking cool. Just imagine a printer built out of those. You just have to figure out a way to put a belt in there for xy and for z.... Well, you just put 4 of those rails vertical. Nice job man!
Thanks! Making a whole printer will be an interesting next step and I hope it will look like some miniature futuristic stage at a festival. I should print band members on the plate, I mean stage xD
“3D printed 3D printer” soooo… Reprap?
@@properprinting haha lol that would be so funny! I can already see the video in front of me, edited with music and lights shining from all sides 😂
@@properprinting For more inspiration I would totally look at the Rook 3D printer. It's using a printed frame. With something like this, it moves us one step closer to self replicating 3D printers in all but the nuts and bolts 😆
Do the side rails need to be extra light also? It makes sense for the moving gantry to be light but I don't think the side rails need to be light as they shouldn't have the need to move. I can definitely see where it helps to have the mounting to the side rails still be light and thus possibly needing to have a custom mounting solution
I would definitely like to see you pursue this idea/concept as this has great potential. It may ultimately prove to not be suited for fast 3d printing, but it will drive innovation in the industry and others that rely on motion systems.
Do it. Speed of one isn't important. Just do one, print the next one with it, boom double speed.
Have two nozzles on the same gantry, have two gantries with two nozzles, check their combined speed. Have 1+2. Use 0.6 for filling, 0.2 for outer shell plus a 0.2 water soluable.
Lower the number of corner kinds by using a square.
Make the interconnecting plastic parts lighter (no extra holes) and longer (less work to assemble).
Have a version of interconnecting parts that don't expose the rods - only the gantry needs that, and would make the rest stiffer.
Awesome project!
This is my first comment ever.
I think there might be a misunderstanding with light as possible. The static parts should be as heavy as possible. Different materials should help with dampening the vibration.
Only the moving parts should be light and stiff.
True that! This reminds me of a video Tom Sanladerer once made where he filled his extrusions with sand. Maybe I can make closed segments which you can fill up xD
Yes, I always thought Tom is called SANDladerer ;) he should have left out the moving parts.
@@properprinting You can refer to X1C's implementation for inspiration. It may save you some time.
By the way, Winston Moy did the same with a Shapeoko CNC 3 years ago. You can find it on his channel.
Yes. The static structure must be able to absorb the forces generated during acceleration and deceleration. A heavier structure will reduce the magnitude of any vibration. I vaguely remember someone suggest using magnets as dampeners.
The generated force can be minimised by reducing the weight of moving parts in the first place.
Another thing to watch out for is the natural resonance frequency of the moving parts. The printer's frequencies of speed changes should not overlap with it.
So right off the bat I said to myself... Stainless Hypo-tubes instead of threaded rods. For reference, I've been a design engineer in the medical device space for the past 30+ years. As for lightening... Maybe a triangular beam that is not skeletonized but has gyroid infill, printed with one side on the build plate. That'll already make it as flat as the built plate, if it doesn't warp.
I like this subject and I think you're on point thinking about the next step in 3D printing.
As always, Awesome work!
Thanks for sharing.
Id love to see this scaled up. Doesnt seem like the best option for going fast but still could make building large printers much cheaper. For example a 500mm sized printer, would be great to see.
Really cool idea, the customizable frame design seems like it'd be really useful for things like enclosures, hope you keep iterating to see where you can go with it. Another idea would be instead of trying to start from scratch and designing your own linear rail system, maybe you could look at an existing linear rail and see what the bare minimum that you could add to it to turn it into a gantry, combining the best of both worlds (a proven effective rail design and a cheap yet lightweight extrusion to attach it to).
Well done. buddy.
The result is impressive. Really modular and very interesting, just like you told me earlier =)
This is an awesome idea with excellent implementation. I think instead of buying 100% threaded rod we should try to find some smooth steel rods(think Prusa style linear rods) that can easily be threaded with a DIE. Ideally sizes that can threaded directly or by just removing the surface finish/chrome with a file/grinder/belt sander. I'm picturing some sort of bearing harness that allows you to rotate it easily to get an even finish for threading with the DIE. If it catches on I'm sure China will eventually sell cheap pre-threaded rods but until then I personally would prefer stronger smooth rods over 100% threaded. They would likely be more concentric as well.
When it comes to choosing materials for mechanical assemblies, it's important to consider the specific requirements of the application. While threaded rods can be useful for transmitting tension loads or providing clamping force, they are not typically considered to be a primary structural material. This is because threaded rods do exhibit some degree of elasticity, which can be useful for producing compression when tightened. However, this elasticity can also result in lower stiffness and strength compared to other materials like steel beams or columns.
If you're looking to achieve a high level of stiffness or strength in your assembly, it may be necessary to consider using stiffer materials and threading the ends as a means of connecting components. By selecting materials that are appropriate for the specific application and design requirements, you can create a strong and reliable assembly that will perform as intended.
I also wanted to kindly mention that thermoplastics may not be the best choice of material for this particular application. This is because their young modulus, or stiffness, is generally lower than that of other materials like metals. However, if you prefer to stick with thermoplastics, I would suggest considering a carbon fiber reinforced filament. This type of filament will have greater stiffness while still maintaining a low weight. Please keep in mind that fiber blend filaments may not be as strong as natural filaments, but they can still be a good option for achieving the desired level of stiffness.
May I suggest you studdy the history of fully 3D printed Reprap printers before going any further ? This will save you a lot of time. Regards.
Smooth rods where just the very ends are tapped and the track rollers you mentioned would eliminate the primary flaw with rough motion. Lightweight foaming filament would reduce weight. You *might* be able to get away with 3 rollers gripping 2 rods as if it was an inverted v wheel printer which means your extruder mount would become a pretty small triangle. Since the 3rd rod is now only needed for tension some of the truss supports connected to it could be eliminated. That same 3rd rod might also become some magnet wire etc if it isn't guiding a roller. If you wanted to be completely berserk about chasing grams you could sand the rods so the cross section would be a D shape rather than an O to further shave down ~40% of the mass of the two remaining rods. You'd need to get everyone on earth to agree not to sneeze to keep it from breaking, but you'd have bragging rights for a DIY ultralight X rail without having to metal print a titanium one.
Breaking away from speedboat I think the structure being preloaded under compression and multiple materials might be interesting for resonance. Tune the threaded rods like guitar strings and fiddle with plastic type / truss design / print settings for the best possible starting point prior to input shaper.
Come to think of it I'm now wondering what impact shoving a guitar string down the middle of standard aluminum extrusion and tightening it would have on resonance. Horrible waste of effort or would it push resonance to a higher frequency and thus out of the way of normal printing.
Do you need a rigid rod? If just compression what about tensioning with wire or a zip tie mechanism?
Truss with slots for stock zipties might provide an adequate rail and make corner fittings much much simpler. I think you can use heat to shink them & add extra tension if needed, & later acetone or epoxy to glue tensioned ziptie to truss.
It be relatively easy to span zipties through the hollow of the trust to different corners for extra strength & dial in (or out) any twist.
Zips come in a huge variety of sizes & materials (stainless even) up to 60inches long. Hose clamp might also work. A metal zip could take a fillet or be bent like angle iron which might be superior to steel rod by weight.
I'd love to see how far this project can go. The original open source concept for 3D printers was to make a machine that make more machines. I think you're on the right track. It's super cool what you're doing, even if it's not the lightest setup, weight is not the only factor for printing well and with speed.
Consider using carbon fiber rods. Cheaply available online.
Try to replace the threaded rod with round stock and get a die to put the threads on the ends just where you need them. I think putting sleeves over threaded rod will just increase the size and weight.
Have you ever seen the Positron? It's also got some clever ideas to print even faster--the most important, I think, is printing upside down to keep the print head as low as possible, reducing the wobble of the frame. The bed is then upside-down, which seems to work just fine. Maybe some of the concepts used there could serve as inspiration for your project? Combine an ultra-light frame with a low center of mass, and a low printhead that can dash back and forth without shaking the frame apart.
hollow pipe to roll on for example filement ptfe could work great with those metal rollers but you would need much smaller diameter rods or tension wires. the problem with that idea would be if overtensioned plastic structures would be prone to deforming
you could use metal pipes but the question is do you even need to tension it, how straight would it be without tension or with minimal tension from single central rod while side tracks used as rails would be fited with pipes that are not under tension
you could also make universal cornet pieces in shape of a hollow wireframe cube and tension the truss with central screw trough the cube itself (a bit limited to 3 sided use with straight alan key with ball key you couls probably use more sides)
I just had an idea for an ultra light gantry and had to come back to the comments of this video. Use 2 small aircraft cables pulled really tight as the gantry. You could have bearings run along the cables for the caraige. It would require the frame to be really strong but the actual moving parts could be extremely light. There would then also be a potential to electrify the cables and use the wheels to transfer power to the hotend to reduce electrical cable mass. You would then only need fan and thermistor wires. Isolating the aircraft cables may be more trouble than its worth though.
I would first start by comparing apples to apples, your rail is at least 1 and a half times longer than the voron rail. Next, you probably dont need the threaded rod to be as thick as it is, and third, if you go with hollow rod and those bearings (which IMO you should) you can put threaded rod through it to tension it. Also (IMO) you only need one smooth rod on each side of the Y gantry, facing directly in. You could also probably get by with an ovoid or diamond shaped prism for the x gantry that only has two rods through it. That would make a big difference I think.
I'd be interested in seeing this more. I also had the idea of smooth rods with threaded ends and concave bearings to roll against them, in hopes it would be smoother. as for lighter... .i think that's an as-you-go situation. unless your printing in aluminum or other lightweight yet sturdy metal or alloy, I don't see an easy solution all around, but step by step improvements as you go. Finding that a solution doesn't work, back to design for it , or perhaps finding a lightweight metal solution to replace it with, etc.
If you were to tap hollow tubes I would recommend using a form tap so you aren't losing as much wall thickness at the ends
Look into Arris Composites. They specialize in conformal, compression molded carbon fiber composites. The performance of parts manufactured with their technology is honestly ridiculous; light, stiff, and resilient as hell. And since they're mostly carbon fiber, they're also electrically conductive, so parts can behave as their own strain sensors. I've never seen anything like it. The only drawback is tooling. But it could be a game changer for high performance FDM printers.
I agree with others, that this may not work for a 3D printer (probably not, I’d say), but I think has a lot of potential as a general structural system. As noted, it could be scaled up for larger objects.
I wonder how it would compare with aluminum extrusions at larger dimensions? The extrusions might me more stiff, but they get *dang* heavy in larger cross-sections. What if the panels were, say, 10cm on a side, with 5 or 6mm rods in them? How stiff would the assembly be, for how much weight per unit length?
Kudos for your creativity and pushing the envelope with different “crazy” ideas!
I learn something or get useful ideas from every one of your videos, this time it was the idea of pre-tensioning the assembly to avoid stripping the threads on the Al rods. I knew this is done with prestressed concrete panels to overcome the low tensile strength of concrete, but I’d never thought of it as a way to save threads in a soft material the way you used it here.
That little chuckle at 12:30 is exactly the reason I still keep making stuff. Things break, designs don't work out, unexpected problems out the wazoo, smacking my head at the wall for hours, days, weeks.
But at the end i have something that give me that little chuckle and suddenly the whole journey doesn't seem that bad after all.
12:30 You look completely thrilled there! And I understand you completely: There's a satisfaction looking at your idea as something real, something you can touch, something you can use... No words for it 😁👌🏼
Even if it isn't as advanced or as light as other methods, I feel that that fact that most of the structural "filler" is 3d printed is still amazing, and it looks like the concept has potential. maybe smooth rods with threaded ends?
Go for it but....
Try to use pccf (stiffness)
Things like wheels you could use nylon (abs isnt Wear resistant at all)
Also instead of that you could Just get 2 aluminium smooth rods and igus fi8 bearings and youll end up with really light thing but then you cant use direct drive system with is ok with klipper and pa. I have tried aluminium rods and whole x axis was lighter than Just 2 steel smoothrods :)
I love this truss idea, while it may not be the best option for a 3d printer, I still think you should try and make it one, I'm interested to see other uses for this sort of system, where the truss structure can be taken better advantage of
One thing with many of the gantries is their tubes are hollow. I know many hobby shops sell tiny metal tubes (~3mm), you could replace the threaded rods with those and screw a bolt inside the ends to keep the stack tensioned.
I suspect the main issue there is the bearing surfaces, whereever you've got plastic rubbing against plastic it's going to wear.
linear slides have ball bearings going around in a circle against a hardened piece of metal which is supposed to be ground to be really flat so that it's super accurate.
I could see this more as a replacement for alu extrusion for the frame although you'd have to compare the rigidity to alu
It requires a bit more machining steps, but what about precision ground rods for the reinforcing/bearing surface? Yes you'd have to tap each end for bolting together, but it would give you a fairly smooth surface for rollers to run on
And as a former live sound engineer and rigger, what about using egg and pin connections? I have no idea how you'd machine the eggs to this scale, but clockmaker's tapered pins and a tapered broach to get the fitment just right may be a starting point!
Also, what about a different geometry to the truss segments, using topology optimisation for an ideal strength to weight ratio?
It's cool looking, but I can't shake the feeling that a very different design would be needed for this use case.
The third rail on the gantry is dead weight, as are the ends of each modular section (fewer, longer modules would reduce the weight).
I'm also wondering if just a pipe could work as a gantry. It would need something else to keep the hotend from swaying forward and back (a slot in one side?), but it might be lighter than 3 threaded rods and their 3D printed supports.
Ideas....
1) Don't use threaded rod, instead, use hollow tube, and tap the ends. now you've eliminated a lot of weight, because you no longer have solid heavy rods, and you no longer need the heavy standoffs, because you can now screw right into the tube.
2) By using smooth hollow tube, you no longer introduce additional friction trying to mesh geared teeth that serve no purpose. small bearings should ride quite nicely on them, could even use tiny V-bearings so they lock onto the metal tubes.
3) I would suggest making different types of the trusses. you don't need screw / mounting holes on all sides, so maybe only do one, or two sides, further reducing weight
4) to keep the parts from "slipping" on the contact faces, you can add tabs on one side of the truss module, and pockets on the other, so they made, creating a more sturdy mating surface.
5) If you need additional end points for mounting, you could add a central threaded rod, and use this as either a mounting point on either end, or as a method of pretensioning the entire truss assymbly
So if you do tubing or pipe, you can see about making seize fit ends that are tapped or threaded rod to attach or mount ends too.
You could also see about a small spot welded and adding a nut inside the tubing, like an m2 or
I really love this modular truss design, it may not be the absolute best for a specific use case but it has a TON of potential for a general purpose building system.
Keep going for sure mate. could you print these beams completely on a large bed, forgo inserting rods and just print with dense/strong material for the rod sections and a lightweight one for the rest of it? Could then use wheels of the same material as the rod section, which would hopefully stop them from being destroyed so quickly. Not sure if the beam would hold up to stress without proper rods inserted but would be interesting to find out if you can completely print a printer.
You might be able to use bicycle wheel spokes instead of threaded rod since they are made to be a lightweight adjustable tensioner and can be bought in a variety of lengths and materials, including carbon fibre.
Some ideas:
-use fiber reinforced polypropylene for stiffness and light weight.
-use stainless steel tubes and a thread former (not cutter) for the tention threads.
you can also still make your design smaller and maybe give the sections some interlocking features.
If you are going to use smooth rods and need to cut threads in them anyway, I think I'd use tubing and tap internal threads in the ends. You could user larger internal diameter tubes which might add to the stiffness without increasing the weight. This was a very nice instructional video. I don't think you are a native English speaker, but your English is very good and very understandable.
Please give us more. I have a few material suggestions: Polypropylene is extremely light, but a little flexible, might be useable because of the truss structure, fairly cheap too. Carbon fiber might be an option since you want stiffness and lightness.
Ohhh, I also tried to build a FDM printed truss system about a year ago and didn't like the result.
My concept was like origami, printing all 3 sides in one flat piece to optimise print speed and layer orientation. Maybe use an unextruded length of filament along the length as a pin to secure the triangles.
I once made a somewhat similar structure with carbon fiber tubes. A bit like the pre-tensioning as you did with the aluminum ones. But fixing with glue.
I was gonna suggest smooth rods and grooved rollers but they you figured that out yourself. You should also only need 3 bearings per carriage, you're over constraining them which makes movement less smooth and more sensitive to alignment. On the small scale of the voron 0 you're unlikely to find success making this lighter and cheaper, but for a large format printer where weight really starts becoming an issue and there is more material that can be removed, you make have better luck.
The straightness of the rails should be somewhat adjustable by tightening and loosening each of the three rods to tension one corner more or less to pull things straight. I don't think this will fix wonky rods though so maybe try buying precision ground linear rods?
Definitely has potential! I would love to see your version of a mostly 3D printed 3D printer. I know the Mulbot and some others already exist but your design skill is just on another level
Did you ever do a comparison of the assembled portals?
It seems that the truss would have a more uniform stiffness. The base gantry is going to be stronger in the wide plane than the narrow plane by a large margin.
I would like to see some strength tests done on your trusses. I considered something similar for building a 4' x 8' CNC router gantry. Only using MDF instead of printed plastic.
I think you should(maybe you already have) keep going on this idea. Even if you don't make something revolutionary, you might come up with something that could make building homebrew 3D printers more accessible.
I love the work and the premise of the idea, I really would like to see how it develops. Honestly though I don’t think it would be a better solution as linear rails is used in many industries. What I would like to see is how it compares in terms of price. If you can have the frame this way and only need to use small rods and few fittings that would be a cool diy 3d printer project.
Also I subbed just in case you upload more stuff about this as I don’t want to miss any 😂
I like the way you think completely out of the box, and try before the project dies. please don't give up
Nice! I’ve been thinking about something with a similar form factor but larger for long and rigid DIY CNC rails.
It's really cool but I think one of the main flaws is the build volume you'll get out of it, if I'm correct it should be smaller than the one you printed them on, plus I'm not sure about how rigid this would be if you stuck 2 pieces together to make a long one. I really like the idea but I don't see the point other than being cool (at least for now), it's not the lightest and probably not the cheapest to build, maybe it could be a great middle ground option with enough developpement
Would love to see additional effort into this.
As for the weight problem.
Maybe hollowed and smooth Titanium rods with the inner diameter threaded for your endcap screws to tighten it down? Or carbon fiber rods maybe? So many options
YES! More Videos!! You are on the right track, for the scale look at how to make the trusses with hollow tube without becoming breakable. I believe you are on the right direction by making them like Lighting trusses.
None of this material is stiff. Next time print in CF Nylon for stiffness of the plastic parts and use CF hollow rods. This way you can keep the look and make the part ridgit. This kind of application is perfect for CF.
If you want great performance use a CF square tube and cut holes into it. Or use thin gauge steel sheets with lots of depressions, cut outs, flares and so on.
The threaded rods are screw rods so have slanted cuts while rollers have groved that are perpendicular to mountings. That means rollers do not match squarely with threads. Result breakage of roller groves.
I think the Lego 3d printer concept is really interesting and definitely worth it. Maybe you won't get the fastest or lightest or even the most precise 3D printer out there, but I think that the best of Lego is their capacity of give a lot of freedom of creation coupled with great practicality.
I think you should aim at that.
You should think about the hollow steel rods with some threaded inserts soldered or brazed to them!
Or some smaller threaded rod cut to a smaller length and use that as an insert, again either soldered/brazed.
For the whole process only cutting external threads on some titanium rod might be preferable though.
I'd love to see more of this project. It may not currently be cheaper, stronger, more durable or lighter but it is cooler looking and it's more customizable.
I wonder if you could use pullstruded carbon rod instead of the metal rods and do a sort of tensegrity to keep the beam components together. Just a piece of kevlar or similar line through the middle of the beam with tensioners on the ends of the beam (tension in the kevlar means compression in the truss segments). Then rather than rollers, get some of the IGUS Iglidur filament and print custom bushings.
Also not sure, but I suspect the truss structure could be optimized. I bet you'd need a lot less filament than you think to "truss-out" 3 carbon fiber rods running axially, at least to be good enough for a 3D printer.
You may want to try to use less hardware overall, you should try to make the pieces of the truss index with echother and have only one threaded rod running down the middle for tension.
Print a triangular profile in vase mod with a 1mm nozzle. Curve the contour to have a concave area at each corner for the rods.
Running the system in tension seems like a positive improvement to me. In the 24/7 printing video about lightest gantries we learn that weight isn't the only factor involved in higher acceleration using input shaping. I would not be surprised if a tensioned gantry would have a far higher input shaping acceleration than a non tensioned one.
You need the shape, not the entire rail. Print as sections and space accordingly as required for the desired regidity. Use hollow tube and epoxy threads in the ends. This also opens the door for CF tube. The result should be consideraly lighter and equally or more rigid.
I think some people have mentioned this already, but this would be great for robotics or engineering projects. It would be pretty cool to see a whole 3d printer assembled using some kind of modular components like these.
Realy nice disign! I cold imagine incresing the segment size and useing a hollow support structure like a printed tube with checker plate-ish features could increase the stiffness while maintaining a light wight, possibly enough to ditch the M3-Rods and use a pice of 1,75 filament as a rail instead. This could reduce the price and increase the accessibility while maintaining a light wight.
This is a great video. Your CAD skills are top notch, especially the ability to make that complex corner piece printable without supports. The idea itself is great and I would love to see more of it.
Smooth titanium rods are definitely the way to go.
As far as filament goes, here are some things to things to think of.
For Stiffness, from most stiff to least, there is (Carbon Fiber Filament, HIPS, PLA, Poly Carbonate and then everything else).
For Density from lowest to highest ( Polypropylene 0.9 g/cm3, HIPS 1.03 g/cm3, ABS 1.04 g/cm3, Nylon 1.06- 1.14 g/cm3, ASA 1.07 g/cm3) and then everything else).
For Strength from strongest to weakest (PVA, Polycarbonate. Nylon, PLA, ASA, PETG and then everything else).
For Durability from most durable to least, (Nylon, Polycarb, ASA, Polypropylene, ABS, PETG, HIPS, PVA
For a part that needs to be stiff and light but not necessarily durable (Carbon Fiber Filament)
For a part that needs to be light and durable but not necessarily stiff (ASA or Nylon)
For a part that needs to be stiff and durable and light (HIPS)
You can find more material info on Simplify3d's resources page.
I saw an interesting design concept. A single smooth rod for X and a single smooth rod for Y, with the hotend assembly mounted at the intersection of the two rods on LMXUU linear bearings contained in a single block. Each rod constrains the potential for rotation on the other rod. The Bowden-based hotend assembly just has to be mounted such that it avoids the rods. With carbon fiber rods and Delrin bushings instead of bearings it can be much lighter. There is also another design concept that uses DIN Rails. They're not the lightest things, but drill enough holes and they can be reduced in weight a lot. Of course, using a Bowden system reduces the weight to a minimum, but you hve the trade-off of needing to do longer retraction. High quality PTFE tubing can make a difference here, along with well-tuned Linear Advance. We certainly can do better than 20x20 extrusions and rollers!
I have a couple ideas that may help, first I like where you were going at the end of the video, go with steel or titanium rods and get sections as long as you can to drive cost down on the individual sections so you can cut them to length yourself and add threads only as far as you need. Next, I would remove the flats on the end of each gantry section and make a flat end piece that can go onto either end of each full length of gantry, I dont know if that would reduce stiffness but I think it would reduce or eliminate the need for supports on the individual sections further reducing print time and cost, plus if total removal does affect stiffness you can print one flat to go between each section.
Final thoughts, steel is more readily available and cheaper for the masses, so if the point is cheap and mostly printed go with steel, if the point is as light as possible go with titanium.
Thanks for your suggestions! I'm currently working on a redesign and the sections can be printed without using support. I want to make the system in such way that multiple ways on tensioning can be used. This makes it easier to source parts and it's possible to decide to use different methods for different purposes. Like light for the gantry, and cheap/heavy for the frame or incredibly cheap if you just want to put something together.
If you can get your hands on titanium thin walled tube ,go that route with those steel bearings. The trick with making stiffness and a light frame is to go thin walled,and large diameters. You will find the compromise between weight and rigidity.
My friends brag about their carbon bikes costing thousands, my bike is lighter and is a steel alloy frame. Thin walled ,double butted, stiff ,strong. Think bigger,thinner. Try titanium bike frame tubes.
Take a look at concrete beams that are prestressed with tension cables. What if you ran 1 very light cable through the center of of your truss then tension it? It would be very strong and light. Then you could replace the rods with something lighter just for your gantry rollers. No need for them to hold the segments together. Maybe carbon fiber tubes. Or maybe lose the rods and just 3d print a trough for bearings to run in? That assembly would be extremely light.
I think that maybe the heavier but cheaper version of the system could be experimented with in order to possibly provide people with a rail and frame system that could be used to make printer builds (or more likely rebuilds) cheaper, even if it was not used for the full frame. It does seem in its current state to be more of an alternative solution rather than an improvement right now but for some people that in itself may be compelling enough to work with it, at least in the earlier "cheap" version using steel instead of titanium.
Please continue this! I'd absolutely love to learn from what u end up doing so that I could maybe make a whole printer that's light and easy to bring around someday!
Also, it might be interesting to integrate the threaded rods for tool head movement. If you did that though I'm almost certain that you'd lose the core xy system that the voron has
I think this has great potential but I think you should focus less on the light weight and focus more on the fact that you created a printable truss. I would love to be able to go to a local hardware store, pickup threaded rods etc, print some parts and be able to make a solid frame vs ordering extrusions. It looks so cool and I think you should build a printer no matter the weight! So cool
You should switch to a non threaded rod and then use a die to add threads only to the ends of the rods. It's more work, but it would roll better and wear less over time. You might be able to use something like a straight pull aluminum bicycle spoke too.
I’ve threaded rod ends before when I was making RC cars, I has to anneal the rods in an oven first and then harden them again after threading but it works.