When it comes to 3D printing using Smooth Overlay Modeling (FDM) technology, there are two main categories of printers: Cartesian and CoreXY, with the latter aimed at those looking for the fastest print speeds thanks to more flexible tool head configuration technology. The lower mass of the X/Y bottom bracket assembly means it can also move faster, prompting CoreXY FDM enthusiasts to experiment with carbon fiber and a recent [PrimeSenator] video where the X-beam is cut from aluminum tube and weighs even more than comparable. Carbon fiber tubes are lighter.
Because CoreXY FDM printers only move in the Z direction relative to the print surface, the X/Y axes are directly controlled by belts and drives. This means that the faster and more precisely you can move the extruder head along the linear guides, the faster you can (in theory) print. Dropping the heavier carbon fiber for these milled aluminum structures on the Voron Design CoreXY printer should mean less inertia, and initial demos are showing positive results.
What’s interesting about this “quick printing” community is that not only is the raw print speed, but the CoreXY FDM printers theoretically outperform them in terms of accuracy (resolution) and efficiency (like print volume). All of this makes these printers worth considering the next time you buy an FDM style printer.
Linear guides are designed to bend to the flatness in which they are installed. This means that the rail will bend the part it is attached to if the part they are attached to is not stiff enough. If that’s enough to worry me, I don’t know, I haven’t used linear guides before.
There are some very dedicated Voron users who only use linear rails with no other support, so it’s not the most rigid system to run on one of the machines with good results.
The CoreXY system moves its head in the X and Y directions. The Z axis is achieved by moving the print deck or gantry. The advantage is that the required movement of the bed is reduced, since movements in the Z-axis are always small and relatively infrequent.
As another commenter pointed out (sort of), the linear rails are now starting to look heavy. I was wondering if they could be made from something lighter like boron? (what could go wrong?)
In fact, I suspect that the best solution is not to separate the manuals from the support. My cheap and terrible printer uses a pair of steel rods as guides and supports, and I doubt that this design can compete with it in quality. (but definitely not accuracy and rigidity)
Installing hardened steel rods at diagonally opposite corners may work, but not with ready-made recirculating ball guides.
In the middle of the track there are holes cut by abrasive water jet to reduce weight. Make the rear side the inlet side so that the natural spread of the jet creates a slight cone and no sharp edges on the front side so that the wipers on the gate (if installed) do not snag or cut.
They are just hardened steel. Just mill them out of carbide. Turned parts from gauge pins in hardened 52100 bearing steel.
Impossible as the induction hardening applied during manufacture creates internal stresses in the rail (some Chinese magnesium alloy rails may not be hardened at all to be machined). management……
In fact, it’s not even a proper support for linear rails. For steel bars embedded in aluminum look at Nadella rails, this is basically a concept but since aluminum needs a large cross section to have some stiffness they are very heavy.
The German company FRANKE produces 4-sided aluminum rails with integrated steel raceways – light and strong, for example:
The stiffness of a beam increases with the square of the area. Aluminum is a third lighter and a third stronger. A small increase in section is more than enough to compensate for the loss in strength of the material. Usually half the weight gives you a slightly stiffer beam.
Using a surface grinder, the rails can be reduced to an H-shape with a sidewall web between the contact planes of the balls (they probably have 4 point contact, but you get the idea). TIL: Titanium (alloy) profiles also exist: https://www.plymouth.com/products/net-and-near-net-shapes/ but you have to ask the price.
Then there was a problem with the Plymouth Tube Company of America lol. After checking with virustotal, all tests showed no problems, except for “Yandex Safe Browsing”, which, in his opinion, contained malware.
I also think the linear rails look heavy and I love the idea of ​​integrated steel rails. I mean, this is for a 3DP, not a grinder – you can lose a lot of weight. Or use urethane/plastic wheels and ride straight on aluminum?
Let’s hope no one tries to build it out of Be There is an interesting comment in the video review about the use of carbon fiber. Now imagine a 5-6 axis machine that can wrap around a 3D printed mandrel in an optimized orientation. Couldn’t find much information about the CF winding project… maybe it is? https://www.youtube.com/watch?v=VEGMEFynPKs
Haven’t studied it carefully, but isn’t the track itself strong enough? Do you really need something more than just a corner bracket for attaching handrails to side rails?
My first thought was to cut the weight in half again by turning the triangles out of the corners instead of the tubes, but you’re right…
Is that much torsional rigidity required in this application? If so, mount the bracket “inside” the corner, perhaps with the screws used for the rails.
FYI: I found this video helpful for rules of thumb for different shapes of structures: https://youtu.be/cgLnADEfm6E
I think if you don’t have a milling machine you can go crazy with a drilling machine and just drill different sizes of holes and get pretty close to it.
This is, of course, a strange obsession (“but why?” is never a valid question in HaD), but it can be further optimized (facilitated) with a genetic algorithm to develop the most efficient part. You may have better results if you use a solid stock and let it cut once in the X-axis and once in the Y-axis.
I know bioevolution techniques are all the rage right now, but I’d go for fractals because they look more scientific and don’t rely on repetitive guesswork … Now this might be old school as we call it, Fractal Punk 90- X?
I think the cost of using a solid material will far outweigh any benefits. You’ve sanded down most of the material, which will make it much larger.
Why assume a transition to hard stocks? Interesting optimization techniques can still be applied to square tubes.
Also, as far as square pipe optimization goes, I think you’ll actually get very little change in quality. The triangles in the truss are already optimal, the attachment points are more technologically advanced. If you translate this into a question of “what design is best for this application” (like full structural analysis for a 3D printer or something), then yes, you can definitely find places to cut weight.
A more achievable optimization method is topology optimization. I’ve only played around with this in SolidWorks, but I think there are plugins to do this with FreeCAD.
After watching the video, there are some (relatively) easily achievable results that need further optimization (although, even as the owner of a Core-XY machine, I personally see no interest in this rabbit hole):
- Moved the rail closer to the side for better stiffness (currently it will experience macro-deflection of the beam as well as deflection of the strut mounted on it)
- Classical truss optimization: The design of truss trusses has not been optimized, and even without the efforts to implement advanced optimization tools, truss design is a very developed field. After reading bridge design textbooks, he could probably reduce the weight by another third without losing stiffness .
While in practice it’s already quite light (and seems stiff enough not to noticeably affect repeatability), I don’t see the point in improving it further, at least not without first addressing the rail weight problem (as other people say).
“Having read bridge design textbooks, he could probably reduce the weight by another third without sacrificing stiffness.”
Cut *weight*? I agree that he probably increased *strength*, but where did the extra weight come from? Most of the remaining metal is used for rails, not trusses.
Use the same aluminum screws that RC enthusiasts use and sand down the linear guides so you can shave off a few grams.
Oh, and by the way, at a car forum about ten years ago it was discovered that filling the thresholds with foam can greatly increase the rigidity of some cars (improve handling, etc.)
So it might be an idea to try using a very light thin wall tube, perhaps for a brazed, brazed, brazed or similar mounting plate filled with expanding foam.
This should be obvious, but of course you want to do any kind of burning, melting, heating, heating, hot types before the foam fills up.
The aerospace industry is similar to honeycomb composite panels. Extremely thin carbon fiber or aluminum body with a typical Kevlar honeycomb structure in the middle. Very rigid and very light.
I don’t think thin wall pipes are the way to go. I’ve never been a big fan of injection-molded CFRP (it loses many of the advantages of UD CFRP, which is the long average filament length that gives it such great strength), and aluminum isn’t usually sold thin enough to save weight significantly. I imagine it would be possible to grind it very finely, but the knocking might prevent grinding fine enough.
If I were going in that direction, I would take a thin sheet of bi-directional CFRP from one of my favorite budget product sites, cut it to size, and glue it to closed cell foam, perhaps wrapping it in layers of CFRP or fiberglass. This will give it more rigidity in the movement and printhead support shafts, and the wrapper will give it enough torsional rigidity to withstand any small protruding moments from the printhead.
I applaud the effort and ingenuity, but I can’t help but feel it’s a waste of energy trying to squeeze every last drop out of a design that isn’t designed for the future at all. The only possible way forward is mass parallel 3D printing to reduce print times. Once someone hacks all these designs, there will be no competition.
But I think from a structural standpoint it’s probably a bigger issue – the strength of carbon fiber is mostly in those long fully encapsulated fibers and you cut them all to make it lighter and you don’t really use the same way for useful reinforcement – now creating a “pipe” or CF truss that weaves where you need it, works in the right direction, would be pretty impressive as they have a CNC router where they can carve an extrusion head.
Trying to find a compromise between doing what you say (which is the best way) and taking a simple DIY approach is one of the arguments for using what is sometimes called forged carbon fiber. But I think I got the idea to try the same basic shape, only in Zr magnesium alloy (or some other really high strength magnesium alloy). Good magnesium alloys have a higher strength to weight ratio than aluminum. They are still not as “strong” as carbon fiber if I remember correctly, but they are much stiffer, which I think will make a difference for this application.
I doubt it’s really “lighter than comparable carbon fiber tubing” – I mean it’s a kind of carbon fiber, stronger and lighter than materials like aluminum.
We used a few CF tubes in a project that was (literally) paper thin and was much stronger than the thicker, heavier aluminum equivalent, no matter how many speed holes you wanted to add.
I think it’s either “because I can”, “because it looks cool”, maybe “because I can’t afford a CF tube” or maybe “because we’re doing it with a completely different/inappropriate tube CF Compare norms.
Define “Stronger” – as a word, it’s so contextual, are you really aiming for stiffness, yield strength, etc.?