The reason the parts are not viable to be printed flat is that this would eliminate the manufacturability of the parts. They must be printed vertically in order to allow for auto-ejection which then allows for mass production to the 100,000's. We then must engineer from that constraint
Or you could mold thousands of them as a time using traditional casting methods. While the curve would be annoying, they are not beyond the ability of using cores that can be removed chemically. Stating that you cannot make them any other way in just incorrect - and decades of making far more complicated structures already exist in many industries.
Why would you print inferior product by the hundred thousands, instead of a better product in custom, small-batch quantities? You think you're going to disrupt injection-molded parts industry with 3d printed product with a clearly inferior product, printed using sub-optimal bed orientation? How easy is it to snap one of those filters in half, since the layer lines run vertically, instead of stacking like plywood? It doesn't make sense.
I mean, Slant 3d seems to value novelty and differentiation over straightforward utility with these videos, like the mug handles with very poor mass separation. They illustrate the strength of the tech, and the ignorance of the design optimization related to the original design goals.
One fun thing to do is to set top and bottom layers to 0 and just let the infill pattern be your vent mesh. Grid for simplicity, Gyroid for style. Prints super fast and works great
That's an awesome idea! But does the gyroid pattern actually work, as I thought those patterns (at least in cura) would close the airflow.. but perhaps not, if you print it laying flat, which you would do while printing at home
@@sarthaksharma9656 Mass-manuracturing. If you print these fan shrouds flat, you fit maybe four per printer. If you stand them up, you can print 20+ at a time. It may not matter if you do one-offs on your personal printer but for a print farm that does parts by the thousands, it saves a lot of labor with re-setting printers between batches.
THANK YOU! I was sitting here trying to figure out why in the F they would print those parts vertically at all. Sometimes these videos are insightful and educational - this one missed the mark for me
My guess for printing the square vent covers vertically is due to the mass production aspect of this company, since they can fit more of these vent covers on the same plate of a single printer than laying them flat. But he really should have explained that. Still I don't see why he was printing the ones with long slats with the slats oriented horizontally instead of vertically.
@@Eidolon1andOnly Even if you can fit more on the plate at once... surely printing them vertically would require it to be thicker, and as a result make it need like, idk, but my uneducated guess is at least twice as much plastic (compared to just 1-3 layers printed flat), and so I would imagine that each piece should take like at least twice as long to print? Well, if printed individually, but I don’t see why printing multiple with the same head at once, would gain that much speed..? I guess for two of the axiis, you can combine the time it takes to move along that axis, doing it once for each part in a line, but, is “moving while not extruding filament” a substantial fraction of the printing time..? I guess maybe it could be..? I don’t have any experience in this area, but it still seems counterintuitive to me that it would be more efficient to have them vertical, and I’m still skeptical about it.. Edit: apparently it isn’t so that they can fit more on the plate, but because a lower footprint results in making it easier to have it automatically removed from the plate after being printed. Huh.
@@drdca8263 It would take roughly the same amount of plastic though vertically would require some support material. Though if you were to print just one individually, there'd be very little time difference between vertical and horizontal. But let's say, for example you have a plate capable of fitting four of these squares horizontally and it takes 8 hours to print all four, but you have an order for 120 of these square vent covers made. That means you'd have to scrape the plate after every 8 hours of printing 30 times to reach the goal of 120 total square vent covers. That's 240 hours of printing plus the time it takes to exchange out the plate or clear it off for the next print. 10+ full days. But if you were able to place 20 of these vent covers on the same plate by printing vertically, and it took 17 hours to print all 20, you're only looking at 10 full prints to reach your goal, and it would be done in a little over 170 hours with clearing the plate each time for the next print. That's only a little over 7 days. That's a pretty decent time saver.
Why did you choose to print vertically? Was it more cost efficient? You could also add chutes so that any airborne debris could drop vertically through the vent. I love the radial exit around the outside diameter.
He said in another video (the angle brackets one), that you always want a small footprint on the print plate, because that works better with auto ejection. More automation equals cheaper part.
@@lupeters213 there are bed surfaces that have really strong adhesion with certain plastics while hot, but basically zero adhesion once cooled down. I once printed a part with a massive footprint (over 50% of the print bed, I think the bed was over 200x200 mm) using PETG, and while it was printing and the bed was hot, it was stuck on like crazy (you could literally lift the printer by the part), but the moment the printing finished and the bed got to room temperature, you could literally just *pick up* the print off the plate. It was nuts, and I actually don't know what the bed material was, but it *is* possible (at least with the bed material + PETG combination)
@@TricksterRad yeah, while printing at home that is definitely the best solution, but these videos are to be seen as design guidelines for mass production with many parts printed and auto ejected. You don't want to let the bed cool and reheat it again in this scenario. The system slant3d uses involves no human input between prints. After it is finished, the piece will get pushed off the plate by the head and the next print starts
@@m00str why not? Letting the bed cool is something you may need to do regardless, depending on what you're printing, and it doesn't require any more human input than getting something to ram a part off the plate. It does slow down your production a little. But I'd argue that that might be worth it considering it is a much gentler way to eject prints, and allows you more flexibility in how you print your parts (sometimes you simply can't afford to print a part in the smallest bed surface area orientation)
@@TricksterRad it is indeed a lot gentler than kicking it off by force, but in some cases it can take a lot of time to cool down and a lot of energy to reheat, especially with larger print beds. I guess it's dependent on what you want. Pushing parts off without cooling could even damage some delicate parts. When I print at home i always pop the part off the bed while keeping it heated, just because I don't want to wait and reheat, but I'm removing the bed and bend it to not damage anything by ripping it off
Should be titled "Why We Don't 3-D Print Vent Covers" - All this is done better, cheaper, easier with stamped or molded pieces. Don't want light through? Just stack 3 plates with offset holes and spacers between the plates. 3D printing is great technology, but this is putting a square peg in a round hole. Every justification is a stretch. "Trap dirt?" That why we add screens which are easily cleaned - how are you going to get dirt (or see it) if it is trapped in a tunnel?
"Let's start with an intentionally bad design, then try to fix it to get still bad design, with additional drawbacks, then trying to fix that get even worse vent which is not a vent any more. And in the end let's show something not much related to the initial idea and probably useless but cool looking". Things like this make 3D printing look like marketing bubble or toys for rich people. Thank god there are others.
I have a feeling that the reason vent covers havent evolved very much over the decades for good reason. They already serve their purpose very well while being very quick and efficient to manufacture en masse. 3D printing a cover will always take considerably longer to do than stamping some sheet metal or brazing some wire stock in a certain way, including any post processing. Vent covers only need to prevent things from going into the fan blades while having minimal impact on airflow and noise generation. If you need dust filtration you use a thin sheet of mesh, not a 1" thick piece of plastic with convoluted tubes inside that can get clogged and/or drastically limit airflow. 3D printing can be very useful in a lot of places but this is not one of them.
@@slant3d homie we replaced horses because they were less efficient than cars. 3d printing is far less efficient for vent covers in pretty much every way
I've just been going through these videos today, but this is the first one that felt odd. the first two or three designs that had issues with needing support would have been much easier to vertically print if you just rotated them so the slots were vertical, wouldn't it?
Why not print slotted vent laying down, fixes the whole support removal issue when there aren't any to remove. Also, all these designs seem like adding more area which in return would not vent as efficiently.
Because then it will take much more bed area and since he is talking about designing for mass manufacturing you can fit way more of those parts when placing them vertically. The one with laying down configuration will add even more cost for just having someone to clear the bed so that another batch could be started than thr cost it saves by reducing post processing.
@@saadqadeer7807 so you're saying someone has to be available for mass production... I sure hope there's someone watching the printer, they are known to catch fire. Also, printing multiple at once or one at a time usually end up taking same amount of time. Also, if that is the main goal, printing multiple at one, as comment above mentioned... Turn it 90 degrees and shouldn't need the supports, and less surface area. This would be faster prints than vents 2 inches thick.
@@g0d77 I meant that the channel is about mass producing things with 3d printers that's why he didn't mention it. Yes it would take same time but in one of the scenarios you have to be able to swap the bed like 10 times to have same number of prints. About printer catching fire idk about that what he does about it but still that's doable with stuff like smock detectors or having fire extinguisher packets that explode and disperse the powder when they hit threshold temperature. And isn't rotating 90 degrees same thing as laying it down on bed?
If you craft your vortices proper you can literally separate the hot from the cold air - conserving the cool air and rejecting the hot air. Ranque-Hilsch vortex math exists.
@@slant3d Curious, what is an instance you've used actually used any of the designs other than the "slots are better than holes" tip? Filters are a much much better option for particulate capture, fans create turbulent enough flow for heat transfer, a regular impeller will have far better performance than a fan-to-radial-outlet design. As far as prototyping goes, 3D printing is king, but for any mass production, unless you can't create the geometries you need with injection molding or machining AND the best solution is a weird internal geometry... This video seems very "everything is a nail to a hammer" but with 3D printing.
I can think of very few, if any, situations where you'd want to restrict your airflow so much as to require a much higher static pressure in exchange for... wavy airflow holes.
I’ve never had any issues with vertical slots, even small ones, I do prefer diagonal slots though. Vertical slots probably aren’t that strong but diagonal slots would be stronger whilst being easier to print than horizontal slots. For most applications there won’t be much force on fan covers so the strength isn’t that important. If designing large fan covers I tend to use a hexagonal pattern tiled within a circle, you can orient it so that there are no bridges. When possible I do print fan covers flat against the bed though, that is best for strength and print ability.
S-Curve Air Vents!. Makes for a very cool, one of a kind heat exchanger made out of thermally conductive plastic! Could even make for air-tight sealed enclosure, avoiding dust inside sensitive electronics enclosure!❤👍
This video should be shown in all engineering school ... As ho NOT to approach design !!! It's a vent, it's purpose is to let airflow passes, your number one metric should always be air resistance measure. All what you are discussing here are just secondary detail. Function FIRST !!!!
But "look there's one step even further we can go" (8:55) ... "we can do whatever we want." Love this. The magic of engineering design, that enables transforming ideas into real solutions, that weren't possible before. In terms of design ideas, or design strategy, 3d printing gives engineers and designers a whole new manufacturing process that they can experiment with. Cost to experiment with, fail, and learn is much lower than other manufacturing processes too.
I have two questions. 1. Who is mass-manufacturing using a printer? 2. A slot has support along the axis of its length, but it has almost no support on the axis of its width. By using slots of only a single orientation you are only providing support in a single direction, am I not correct? If you used two slots perpendicular to each other, you would have support in every direction.
I agree. For mass production, ts not ideal to rely on the slicer for this. But for one-off or personal use, I think 0 top/bottom is good enough, though.
@@slant3d Is that going to work reliably, should you decide you want a geometry in the style of gyroid infill? EG, doesn't the treatment of narrow wedges (as between the mesh on either side of a geometrically defined gyroid) vary with slicer settings, between slicers, and between versions? (I'll also opine that a callout like "Gyroid infill, 20% with 0.8mm nozzle, pattern axis aligned with long axis of part" is really no different than the threading directions that people have been putting on drawings since... threads were standardized? Problem here isn't the mass production, but the handoff, with insufficient documentation, of a nonstandard feature, to an external vendor.)
You could easily make these patterns using whatever manufacturing technique you want. You also don't need supports for a 3d print like the ones you showed if you rotate the model before you print it.
Those S-Curves would be brilliant in window coverings like blinds: very good air flow, with complete privacy. Would probably be great sound-dampening too.
@@GarrettWAssuming the tubes through the plate have the same cross section throughout their length, it would be the same air flow as straight tubes minus the extra drag created by friction with the longer walls of the tube, which is negligible.
My first question would be: why are you printing these in the worst configuration possible? Just turn the part 90 deg to get rid of all the issues with supports.
While the support problems you noted with horizontal slots make sense, and vertical or diagonal slots would introduce the same islands that caused rejection of round holes, I wonder if diagonal slots with a nonplanar slicer could solve some of these issues. You would still need to build up the slot in layers, and that would require islands, but a couple of diagonal layers or a diagonal "stem" on each step would serve to lock the islands to a single common strand. As to the weird hole path concepts… I struggle to see any merit to them. Thinner pipes have more drag per cross section than wider pipes, and longer more than shorter. If the goal is to minimize restriction of airflow, the pipes should be as short as possible and as wide as possible. In other words, a sparsely perforated flat plate. Overwrought internal passages may be of interest to a heat sink, in which narrow spaces and turbulent flow may improve heat transfer, but for simply exchanging air these solutions are strictly worse. Your suggestions of internal dust traps in a part designed specifically to be impossible to thread a tool through are, frankly, deranged. Moreover, your claim that these forms are unmanufacturable with techniques other than 3D printing neglects processes such as lost material casting and multi-part assemblies. If it is possible to produce a turbine blade with internal coolant channels by non-3D printing processes, it seems strange to suggest that your vent covers will be any more challenging. Finally, it seems a dubious engineering philosophy to design parts to fit a process. Here, it seems you have gone even further, and are proposing designs merely to rationalize using a process you have already picked-perhaps even to justify purchase of expensive equipment. Instead, process selection and design should go hand-in-hand. If the requirement is a vent cover, then it seems sensible enough to design a stamped piece, or a bent-wire piece, or potentially a molded or even machined piece. 3D printing is not appropriate for the requirement of a thin, plate-like shape with a pattern of holes, and inventing reasons to use a slower and more error-prone process does a disservice to your design, the process, and the application.
Some great ideas, thanks! I realise that for mass production you'd probably want to stack as many of those vents on the build plate as possible in one job, but printing the vent flat would solve a lot of the "needs support to print" designs issues with the first models
@@evbunke2 3D printing two parts at once takes longer than printing the parts one at a time, as you have added additional motions to move from one part to the next. Slant3D also prefers to print parts with minimal contact to the build plate, as more contact area makes for more effort in removal of the part.
Another thing to note is that while these examples are flat plates, this could also be on some bigger part that necessitates a different printing orientation.
nice, vid but i feel there are some missing info. I assume there is a reason you don't want to print it flat on the plate, like u cant print a lot on a plate at once, making the big flat side of the print a no go? even tho it might be a when its to be installed so the screw wont act like a vedge in between the layer lines. The second example would that not be super easy to print if just rotated so the slots are vertical and not horizontal?
Printing something like this vertically would exponentially change print times. Instead of printing a large flat cross section, with many edges that could separate from the bed (although a larger surface area as well) you can print a much smaller cross section more quickly, as there is much less travel while printing the part. That would be the most notable reason.
You should see it more as a specific case where you have lots of islands. Of course in this very case if printed it flat, it would solve the problem. But learning wise it would be pointless.
I'm going through this process for a custom solar generator. Since the equipment will be used outdoors it needs to be at least IP40 (not waterproof, but that ambient rain and splashes of water don't get into) and the 3d design gives so many tools for that instead of just punching or drilling holes for a vent. Right now we're on tilted triangle holes, so water doesn't reach the fan and it drips out the little canal on the triangle's tip. For sure I'll be trying some of the ideas here.
Having a 3D printer is a licence for so many bad ways to design things. That hurt my head watching you show terribly inefficient cooling paths and missing the actual good possibilities in the technology.
I mean… these re cool ideas but I would be curious if these are actually an improvement. In my eyes using simpler covers in combination with ducts to direct or manipulate airflow sounds a t cheaper, simpler and easier to design.
You could also print custom waveguides out of conductive filament. Previously impossible to manufacture structures are shifting several areas of thinking in multiple industries lol good stuff man
The second example with straight slots is a perfect case for rotating it 90 degrees so the slots print vertically as their would be no bridging and no support would be needed if the slot is cut with rounded edges making an arch.
Reminds me of something even crazier. The other day I watched a video of a company that had an AI design organic piped burn chambers for a rocket engine to achieve desired characteristics snd then print in metal.
Brilliant! Totally just redeemed my plans to use pc fans for my grow tent exhaust. I was literally just about to scrap the idea bc they would let too much light in
I get there may be a situation where you would need to print these in the orientation that you're talking about, but If I'm printing a fairly flat piece with slots in it, I'm going to lay it down on the bed. You're going to get a stronger print that way as well, because of the layer lines
What I often do in situations like this is to use a grid. But normally, FDM slicers really don't like grids, they will essentially play snake on the build plate and mess everything up. The trick here is to interleave columns and rows of the grid, with each of them being as tall as a layer height. They still mostly fuse, even though technically, it's just 50% of the material.
I think the throughline of the narrative has me confused: "It's better to print slots than holes, so don't design with holes. Not here's all the things you can do with 3D printing using holes instead of slots." Also, so many of those designs severely restrict the total airflow through the vent's area. It might be useful for some novel consumer products, but for industrial applications, the stamped metal louver plate still seems to be the best choice, plenty of airflow, and the benefits of angled slots, namely preventing falling particles, falling liquids from entering through the vent.
This is a fantastic idea I could see useful in various projects. I’ve never thought about this way of vent design before, thanks for sharing! Will definitely keep this in my back pocket for future enclosures
print vent as separate body without top and bottom and use slicer options for infill to choose appropriate inner structure. It will be the fastest print of the vent cover ever.
That last example is really good. I met someone at a conference in April this year who was doing just that - printing impossibly ducted copper heat sinks for high power components. When they handed me the sample on their exhibition stand, I had one of those moments where I just couldn't believe what I was seeing!
Unlike most of the other videos in this series, I'm not so convinced by most of these shrouds. I'm aware that your particular company has constraints (must print with minimal bed contact for auto-ejection system, etc.) that discourage flat objects, which explains the unusual and potentially weakened print orientations, but I'm really not convinced on the square grids of serpentine round holes and similar high-flow-restriction designs. Seems like you would need a VERY high static pressure fan to take advantage of them. Nice to keep the techniques in mind though, as it's entirely possible that you'll happen upon a use case where they ARE actually practical! Seems like a hex pattern of hexagonal holes (or making vertically serpentine horizontal slits rather than holes) could pack the serpentine designs a lot more densely, providing far more airflow per unit area. Seems like just a diagonal would be far easier to clean than serpentine patterns, though.
I feel like printing this part in a vertical orientation is an edge case that's applicable for printer farms. I don't think this is the optimal way to print it. Not only will it take longer, but it's going to introduce more visual defects. All of this in the name of part ejection. If you're a hobiest, none of this is as applicable.
Even easier to just remove top and bottom layer (for a section or whole part) in the slicer and let the infill produce very fast printing grills that look awesome!
For the example at 1:40 Why not print it laying flat? No supports needed, the thing becomes one of the easiest prints you can do if you just orient it the right way
I am at the beginning stages of production of my own inventions right now and the shear cost of molding just one of my 12 parts in their original design was $25k. That's just the mold, not material. I'm not even sure they could IM petg-cf. With my final production design the cost would come down since there are less "pulls" to do but it wouldn't matter. My parts cannot be molded. I'm sure a hundred people commenting here would love to say otherwise but a mold simply cannot eject my parts. And no you cannot simply design around it or modify it. Design for the process as stated. The only trouble is you cannot in my case, and yet I still need to make 10s of thousands of them if not more. If you have to make a mold then use cnc tooling, machining to cut into it as a completely separate process step then you are NOT injection molding parts, you are machining them. There is a huge difference between Injection Molding a part the way we all imagine it--stamp, drop, bin, repeat and what many of you commenting claim to be Injection Molding. If you have to use a mold, AND THEN use separate tooling on different machines to make cuts, holes, deburr or whatever the case may be then you are NOT injection molding in the traditional sense-you are machining. Sure the IM plant can come back to me and say "yes, we can make that." But making it happen in totality and simply stamping a part into a bin are 2 totally different animals. And the machining part I would imagine takes drastically more time than the molding part since one blade or bit can only cut one object at a time.
Hate to break it to you but a lot of, if not all of the designs you showed could be made using either basic tooling or positive to navigate mold construction, yes 3D printing does it far cheaper, far faster, and at a far larger scale. And for those who are wondering what positive to navigate mold construction is the process of making a master of the product/feature(s) you want then making navigate molds from the master to than make the final product. This is the whole "clone a thing" process but it can be used to make parts for castings.
"Why don't print it laying down?" Because then it will take much more bed area and since he is talking about designing for mass manufacturing you can fit way more of those parts when placing them vertically. The one with laying down configuration will add even more cost for just having someone to clear the bed so that another batch could be started than thr cost it saves by reducing post processing.
I'm on 6 minute wathing this, and you still pretending that you know what Arflow is and air has no dencity by pointing on s-shaped tiny tubes. Let me help with this - Airflow is free inertional moving of air molecules. What you pointing is while taking that brick with few low-diameter holes will not give you airflow, it will be airstuck. Higthly inefficient. You want airflow - make lot of open space without any elements that can prevent air to move straight line. For the best airflow possible you just need one big hole or nothing at all. If you want to make a filtration - make an air filter and do not tell anyone that it "provides airflow" - it doesn't. It actually prevents airflow from happening by making obstacles. Any directional changing of air flow will take energy from the air.
Water soluble filament could be useful for this too. And not just for supports. Say you want 1 thin line going through the mayerial. Instead of trying to print a tiny hole, print a thin line of water-soluble filament. Then ramp up extrusion with normal filament and squish against the thin line. In a sense, it's like adding a layer of subtractive manufacturing to it
One question... why don't you print it lying down? Like why add supports? ... I mean, with mass production in mind, I kinda know the answer, sorry. What about snake turn drills? It worked with tunnel drilling, but yes, we don't see them in parts manufacture.
I'm sorry, are you telling me you print these standing up? Why? There is no valid reason to do that. You waste material with supports, you risk warping, and you now have to worry about snaping it along the layer lines. Printing these flat makes so much more sense. You dont have to jump over any gaps, the layers are set out along both the width and length so layer adhesion is much less of an issue, and as long as you have good bed adhesion there should be little to no warping. Additionally if you must print vertical with holes, hexagons with the point facing up are better, especially if they just need to be a hole and not hold any other parts. Also that first vent you showed is not typical, not if you put a fan behind it. It would be really restrictive. An actual vent that you would normally see in the wild usually has the holes done up in concentric circles with a set spacing to maximize airflow. That strange star pattern you made would be aweful in comparison. You would be better off with just a big hole. One last thing, the reason we really use straight through vents and stamped ones with veins is not just because of the ease of manufacturing, but because straight through is the most efficient. And bends and turns only add resistance and thus lowers your cooling/airflow potential.
@@slant3d Hang on, why? Especially designing explicitly for mass production on FDM, if it's possible to create the geometry you need using infill, for which this is an excellent example, I would always want to take advantage of that, and design the part around the slicing profile I expect to use. Making the design parametric from the get-go can really help here, as the CAD can be tweaked and adjusted in step with the GCODE tuning. Some slicers will allow you to print two parts on the same bed with different settings, nothing says these can not be touching so as to fuse into one part, so I might sometimes combine an infill-only grating, with 1mm thick infill and just a couple of outer walls, with a more detailed surround at 0.6mm width and denser, standard infill, all from the same 0.6mm nozzle in the same part.
@@slant3d whats the reason for that? im totally not a 3d printing person, but somewhere in here was said that 1 part has to be printed at a time, so i assume both orientations would print in the same time, and its not like horizontal piece takes space from other pieces because its only one printing at a time
highlights: - the guy goes off on some ill considered rant about vortex generators and laminar vs turbulent flow, and somehow managed to completely ignore (and catastrophically fuck up) porosity, static pressure, air flow impedance, ease of cleaning and maintenance... - at one point he mocks up the weird disc with the holes redirecting to the sides thing with 5 holes, and a 5 bladed fan. does he not know why most fans have prime numbers of blades? our protagonist has just re-invented the old timey air raid siren! - near the end our hero suggests 3D printing copper heat sinks. in a video about mass production you can actually use 3D printing for all sorts of cool well optimized shrouds and ducts for a fan that would be difficult to make through other means, but the trick to doing that is to actually think about the application, not vaguely imagine what you think a fan does in an entirely abstract machine.
It's really cool to see the kind of innovation that comes from a relatively new manufacturing technique. The further into the video you go, the more insane the geometry gets. And while I imagine you technically *could* manufacture these parts in other ways, it'd probably be stupid expensive and complicated to, and likely even then still likely with extreme compromise. For example, several smaller parts later joined together would be one way. It just wouldn't make any sense to do. I was actually quite critical of the last Slant 3D video I watched (the mug handle video) because I felt it was very dismissive of other manufacturing techniques as antiquated. But this doesn't seem to be the case here. Instead it's just showing off the really cool things that are being made practical with the use of 3D printing. Doing so also provides solid examples for why someone might want to choose 3D printing over other techniques for high volume production. It does also show its limitations and potential ways to mitigate them. For example the basic vent designs with 3D printing may not be an ideal combination, but the slants work much better. (Though I assume this introduces more drag, something that may very well be an acceptable trade-off) Some things that came to mind for me while watching these specific examples were the complicated hydraulic circuits that are often found in things like automatic transmissions. As well, I imagine you could do other crazy geometries such as Tesla valves. The whole idea of complex internal geometries offers a ton of potential. Overall, I'd love to see more content like this, and it makes me really excited to see what other really cool things people can do with this technology or how principles already outlined here may be applied for other use cases!
Just print the piece flat and no supports needed for the first 2, and any word about difficulty with the basic vents is stuff much easier to do with traditional manufacturing methods, and so this 3d print ideas are solving an imaginary problem
The intent of the video is understandable, but is nothing more than 3Dprinting basics and promises. A tested, real world example with actual, indended performance, is absolutely necessary.
Printing that orientation would serve well for higher volume because of how much you could fit on a plate, but that is no reason to complicated the design.
It really sound to me how to make your life more difficult. I understand why you wouldn’t place it face down but on the other hand it make for a easier and cleaner print and much stronger also.
I don't get your 'assumed' 3D printing orientation. Surely you would print those first examples flat on the build plate then you wouldn't get any of the problems you mentioned!
@@slant3d It seems that there's many things here that hobbyist 3D printers find unintuitive, because they think in terms of 3D printing at home, not mass production. Like if you want slots and can't print it flat, surely you'd orient the slots upward? Then there would be nothing to bridge.
So you really never thought to lay the part flat on the bed instead of building it on it's edge? Slicers do have a feature that allows you to reorient your part - you know that right?
I couldn´t follow for a second, cause I couldn´t understand, why there´s any advantage in printing those vertically, while printing them flat would have eliminated the issues in the first place...
I'm no engineer, so take my ideas with a pinch of salt. If we've designed a fan plate to act like a dust filter, then we have designed a dust filter, haven't we? A dust filter that would either clog up and give us zero airflow, or give us poor airflow and let all the dust in anyway. So this hybrid is both a bad fan plate, and a really bad dust filter. I dont say it's impossible to design an actually useful fan plate that uses 3D-printed micro-structures (easilly plugged tiny canals, that kill the airflow), but then we'd need to have an entire structure designed to take the air in, swirl it around to separate the dust, eject the dust through side ports, and let the clean air inside. I have never seen a dust filter that doesnt require maintenance. It seems like we'd need wuite a fan behind this "fan plate" to get any air inside through that. Or maybe itd be possible to nake afan plate from mumtiple layers of sparse grids of filament that act like a dust mesh... Wait... Uhh.. What about an ionizer-based 3D-printed air filter? It ionizes the dust particles and makes them stick to the walls of... Our chasis. Maybe we xould 3D print a large block that'd have relatively wide canals - say a mm. Thatd gently wobble the air left and right and provide extra paths for the dust to fall into due to inertia... Then wed just need a tray underneeth to catch it all sonwe can clean it every now and then - or just drop it on the floor and let sonething else deal with it. If this fan plate.. I mean filter block works, we could revolutionize the industry. Or evolutionize ourselves out of the business...
The reason the parts are not viable to be printed flat is that this would eliminate the manufacturability of the parts. They must be printed vertically in order to allow for auto-ejection which then allows for mass production to the 100,000's. We then must engineer from that constraint
Or you could mold thousands of them as a time using traditional casting methods. While the curve would be annoying, they are not beyond the ability of using cores that can be removed chemically. Stating that you cannot make them any other way in just incorrect - and decades of making far more complicated structures already exist in many industries.
belt printers could auto-eject these.
That explains the first one but the two after could be fixed without support by rotating them 90°
Why would you print inferior product by the hundred thousands, instead of a better product in custom, small-batch quantities?
You think you're going to disrupt injection-molded parts industry with 3d printed product with a clearly inferior product, printed using sub-optimal bed orientation?
How easy is it to snap one of those filters in half, since the layer lines run vertically, instead of stacking like plywood?
It doesn't make sense.
@@davidmcintosh19 exactly what i was thinking and i came to the comments to see if anyone else agreed
This sounds like a way to decrease air flow and increase noise. I encourage you to run it through a simulator to check efficiency.
You can design it in any way that meets the specs of the project. You could even design them to increase air flow and decrease noise.
@@slant3d you say that the additional flow resistance can increase the air flow ??
Presumably the "specs of the project" required noisy, lossy flow.
I mean, Slant 3d seems to value novelty and differentiation over straightforward utility with these videos, like the mug handles with very poor mass separation. They illustrate the strength of the tech, and the ignorance of the design optimization related to the original design goals.
Oh yeah, this is the idiot who 'redesigned' the mug handle to be shittier
One fun thing to do is to set top and bottom layers to 0 and just let the infill pattern be your vent mesh. Grid for simplicity, Gyroid for style. Prints super fast and works great
That's an awesome idea! But does the gyroid pattern actually work, as I thought those patterns (at least in cura) would close the airflow.. but perhaps not, if you print it laying flat, which you would do while printing at home
In Prusa slicer, you can add modifiers to remove those top and bottom layers in specific locations of your model.
why dont just print it flat
@@sarthaksharma9656 Mass-manuracturing. If you print these fan shrouds flat, you fit maybe four per printer. If you stand them up, you can print 20+ at a time. It may not matter if you do one-offs on your personal printer but for a print farm that does parts by the thousands, it saves a lot of labor with re-setting printers between batches.
get an oscilloscope, find some software that lets you make a map of all of the readings that it takes in aggregate, and speak into it.
For the first few examples, you can also just lay the part flat on the front face and you wouldn't have to worry about support material.
@bensmith3890 you could also rotate it 90° so there's no need to support the vents
THANK YOU! I was sitting here trying to figure out why in the F they would print those parts vertically at all. Sometimes these videos are insightful and educational - this one missed the mark for me
My guess for printing the square vent covers vertically is due to the mass production aspect of this company, since they can fit more of these vent covers on the same plate of a single printer than laying them flat. But he really should have explained that. Still I don't see why he was printing the ones with long slats with the slats oriented horizontally instead of vertically.
@@Eidolon1andOnly Even if you can fit more on the plate at once... surely printing them vertically would require it to be thicker, and as a result make it need like, idk, but my uneducated guess is at least twice as much plastic (compared to just 1-3 layers printed flat),
and so I would imagine that each piece should take like at least twice as long to print?
Well, if printed individually, but I don’t see why printing multiple with the same head at once, would gain that much speed..?
I guess for two of the axiis, you can combine the time it takes to move along that axis, doing it once for each part in a line,
but, is “moving while not extruding filament” a substantial fraction of the printing time..?
I guess maybe it could be..?
I don’t have any experience in this area, but it still seems counterintuitive to me that it would be more efficient to have them vertical, and I’m still skeptical about it..
Edit: apparently it isn’t so that they can fit more on the plate, but because a lower footprint results in making it easier to have it automatically removed from the plate after being printed. Huh.
@@drdca8263 It would take roughly the same amount of plastic though vertically would require some support material. Though if you were to print just one individually, there'd be very little time difference between vertical and horizontal. But let's say, for example you have a plate capable of fitting four of these squares horizontally and it takes 8 hours to print all four, but you have an order for 120 of these square vent covers made. That means you'd have to scrape the plate after every 8 hours of printing 30 times to reach the goal of 120 total square vent covers. That's 240 hours of printing plus the time it takes to exchange out the plate or clear it off for the next print. 10+ full days. But if you were able to place 20 of these vent covers on the same plate by printing vertically, and it took 17 hours to print all 20, you're only looking at 10 full prints to reach your goal, and it would be done in a little over 170 hours with clearing the plate each time for the next print. That's only a little over 7 days. That's a pretty decent time saver.
Why did you choose to print vertically? Was it more cost efficient? You could also add chutes so that any airborne debris could drop vertically through the vent. I love the radial exit around the outside diameter.
He said in another video (the angle brackets one), that you always want a small footprint on the print plate, because that works better with auto ejection. More automation equals cheaper part.
@@lupeters213 there are bed surfaces that have really strong adhesion with certain plastics while hot, but basically zero adhesion once cooled down. I once printed a part with a massive footprint (over 50% of the print bed, I think the bed was over 200x200 mm) using PETG, and while it was printing and the bed was hot, it was stuck on like crazy (you could literally lift the printer by the part), but the moment the printing finished and the bed got to room temperature, you could literally just *pick up* the print off the plate. It was nuts, and I actually don't know what the bed material was, but it *is* possible (at least with the bed material + PETG combination)
@@TricksterRad yeah, while printing at home that is definitely the best solution, but these videos are to be seen as design guidelines for mass production with many parts printed and auto ejected. You don't want to let the bed cool and reheat it again in this scenario. The system slant3d uses involves no human input between prints. After it is finished, the piece will get pushed off the plate by the head and the next print starts
@@m00str why not? Letting the bed cool is something you may need to do regardless, depending on what you're printing, and it doesn't require any more human input than getting something to ram a part off the plate. It does slow down your production a little. But I'd argue that that might be worth it considering it is a much gentler way to eject prints, and allows you more flexibility in how you print your parts (sometimes you simply can't afford to print a part in the smallest bed surface area orientation)
@@TricksterRad it is indeed a lot gentler than kicking it off by force, but in some cases it can take a lot of time to cool down and a lot of energy to reheat, especially with larger print beds.
I guess it's dependent on what you want. Pushing parts off without cooling could even damage some delicate parts.
When I print at home i always pop the part off the bed while keeping it heated, just because I don't want to wait and reheat, but I'm removing the bed and bend it to not damage anything by ripping it off
Should be titled "Why We Don't 3-D Print Vent Covers" - All this is done better, cheaper, easier with stamped or molded pieces. Don't want light through? Just stack 3 plates with offset holes and spacers between the plates. 3D printing is great technology, but this is putting a square peg in a round hole. Every justification is a stretch. "Trap dirt?" That why we add screens which are easily cleaned - how are you going to get dirt (or see it) if it is trapped in a tunnel?
"Let's start with an intentionally bad design, then try to fix it to get still bad design, with additional drawbacks, then trying to fix that get even worse vent which is not a vent any more. And in the end let's show something not much related to the initial idea and probably useless but cool looking". Things like this make 3D printing look like marketing bubble or toys for rich people. Thank god there are others.
For some of the bridging, island, and support issues, have you ever considered *CHANGING THE PRINT ORIENTATION.*
I have a feeling that the reason vent covers havent evolved very much over the decades for good reason. They already serve their purpose very well while being very quick and efficient to manufacture en masse. 3D printing a cover will always take considerably longer to do than stamping some sheet metal or brazing some wire stock in a certain way, including any post processing. Vent covers only need to prevent things from going into the fan blades while having minimal impact on airflow and noise generation. If you need dust filtration you use a thin sheet of mesh, not a 1" thick piece of plastic with convoluted tubes inside that can get clogged and/or drastically limit airflow. 3D printing can be very useful in a lot of places but this is not one of them.
Just like Horses
@@slant3d homie we replaced horses because they were less efficient than cars.
3d printing is far less efficient for vent covers in pretty much every way
@@baboom-wof Especially when you're losing air flow by making it do loops.
@@slant3d You haven't built a car, you built a donkey
I've just been going through these videos today, but this is the first one that felt odd. the first two or three designs that had issues with needing support would have been much easier to vertically print if you just rotated them so the slots were vertical, wouldn't it?
well said
I think he wanted to avoid having many "islands" horizontally. Still there are countless more sensible ways of making a cool yet working fan shroud
I feel like printing things in the absolute stupidest orientation is a great way to get comments on your videos🥴😉
Why not print slotted vent laying down, fixes the whole support removal issue when there aren't any to remove. Also, all these designs seem like adding more area which in return would not vent as efficiently.
Or just rotate it 90 degrees
Because then it will take much more bed area and since he is talking about designing for mass manufacturing you can fit way more of those parts when placing them vertically.
The one with laying down configuration will add even more cost for just having someone to clear the bed so that another batch could be started than thr cost it saves by reducing post processing.
@@saadqadeer7807 so you're saying someone has to be available for mass production... I sure hope there's someone watching the printer, they are known to catch fire. Also, printing multiple at once or one at a time usually end up taking same amount of time. Also, if that is the main goal, printing multiple at one, as comment above mentioned... Turn it 90 degrees and shouldn't need the supports, and less surface area. This would be faster prints than vents 2 inches thick.
@@g0d77 I meant that the channel is about mass producing things with 3d printers that's why he didn't mention it. Yes it would take same time but in one of the scenarios you have to be able to swap the bed like 10 times to have same number of prints. About printer catching fire idk about that what he does about it but still that's doable with stuff like smock detectors or having fire extinguisher packets that explode and disperse the powder when they hit threshold temperature.
And isn't rotating 90 degrees same thing as laying it down on bed?
If you craft your vortices proper you can literally separate the hot from the cold air - conserving the cool air and rejecting the hot air. Ranque-Hilsch vortex math exists.
The internal tubes are going to cause a huge amount ot air resistance. I can see how it could work but you cant just slap that on any old fan.
Very true. It is a design concept that requires expertise to implement in a particular situation.
@@slant3d Curious, what is an instance you've used actually used any of the designs other than the "slots are better than holes" tip? Filters are a much much better option for particulate capture, fans create turbulent enough flow for heat transfer, a regular impeller will have far better performance than a fan-to-radial-outlet design. As far as prototyping goes, 3D printing is king, but for any mass production, unless you can't create the geometries you need with injection molding or machining AND the best solution is a weird internal geometry... This video seems very "everything is a nail to a hammer" but with 3D printing.
I can think of very few, if any, situations where you'd want to restrict your airflow so much as to require a much higher static pressure in exchange for... wavy airflow holes.
I’ve never had any issues with vertical slots, even small ones, I do prefer diagonal slots though. Vertical slots probably aren’t that strong but diagonal slots would be stronger whilst being easier to print than horizontal slots. For most applications there won’t be much force on fan covers so the strength isn’t that important.
If designing large fan covers I tend to use a hexagonal pattern tiled within a circle, you can orient it so that there are no bridges. When possible I do print fan covers flat against the bed though, that is best for strength and print ability.
S-Curve Air Vents!. Makes for a very cool, one of a kind heat exchanger made out of thermally conductive plastic! Could even make for air-tight sealed enclosure, avoiding dust inside sensitive electronics enclosure!❤👍
This video should be shown in all engineering school ...
As ho NOT to approach design !!!
It's a vent, it's purpose is to let airflow passes, your number one metric should always be air resistance measure. All what you are discussing here are just secondary detail. Function FIRST !!!!
But "look there's one step even further we can go" (8:55) ... "we can do whatever we want." Love this. The magic of engineering design, that enables transforming ideas into real solutions, that weren't possible before.
In terms of design ideas, or design strategy, 3d printing gives engineers and designers a whole new manufacturing process that they can experiment with. Cost to experiment with, fail, and learn is much lower than other manufacturing processes too.
I have two questions.
1. Who is mass-manufacturing using a printer?
2. A slot has support along the axis of its length, but it has almost no support on the axis of its width. By using slots of only a single orientation you are only providing support in a single direction, am I not correct? If you used two slots perpendicular to each other, you would have support in every direction.
Use gyroid infill with no perimeters in the vented area, preferably with a larger nozzle.
I was thinking the same. Gyroid infill with no top and bottom. It's simpler to design and print
That is possible. But in mass production you generally do not want to rely on slicer settings to get part properties. They are best created in CAD
I did that for a sink drain strainer
I agree. For mass production, ts not ideal to rely on the slicer for this. But for one-off or personal use, I think 0 top/bottom is good enough, though.
@@slant3d Is that going to work reliably, should you decide you want a geometry in the style of gyroid infill?
EG, doesn't the treatment of narrow wedges (as between the mesh on either side of a geometrically defined gyroid) vary with slicer settings, between slicers, and between versions?
(I'll also opine that a callout like "Gyroid infill, 20% with 0.8mm nozzle, pattern axis aligned with long axis of part" is really no different than the threading directions that people have been putting on drawings since... threads were standardized? Problem here isn't the mass production, but the handoff, with insufficient documentation, of a nonstandard feature, to an external vendor.)
You could easily make these patterns using whatever manufacturing technique you want. You also don't need supports for a 3d print like the ones you showed if you rotate the model before you print it.
Those S-Curves would be brilliant in window coverings like blinds: very good air flow, with complete privacy. Would probably be great sound-dampening too.
just use fabric
@@GarrettWAssuming the tubes through the plate have the same cross section throughout their length, it would be the same air flow as straight tubes minus the extra drag created by friction with the longer walls of the tube, which is negligible.
My first question would be: why are you printing these in the worst configuration possible? Just turn the part 90 deg to get rid of all the issues with supports.
*takes it a step further and prints a solid rectangle*
Not sure why you'd want to mass print the the angled slot ones when you can probably just stamp it out of metal like any hvac vent. Maybe aesthetics?
Stamped sheets stacks is one way for Mass Production
While the support problems you noted with horizontal slots make sense, and vertical or diagonal slots would introduce the same islands that caused rejection of round holes, I wonder if diagonal slots with a nonplanar slicer could solve some of these issues. You would still need to build up the slot in layers, and that would require islands, but a couple of diagonal layers or a diagonal "stem" on each step would serve to lock the islands to a single common strand.
As to the weird hole path concepts… I struggle to see any merit to them. Thinner pipes have more drag per cross section than wider pipes, and longer more than shorter. If the goal is to minimize restriction of airflow, the pipes should be as short as possible and as wide as possible. In other words, a sparsely perforated flat plate. Overwrought internal passages may be of interest to a heat sink, in which narrow spaces and turbulent flow may improve heat transfer, but for simply exchanging air these solutions are strictly worse. Your suggestions of internal dust traps in a part designed specifically to be impossible to thread a tool through are, frankly, deranged.
Moreover, your claim that these forms are unmanufacturable with techniques other than 3D printing neglects processes such as lost material casting and multi-part assemblies. If it is possible to produce a turbine blade with internal coolant channels by non-3D printing processes, it seems strange to suggest that your vent covers will be any more challenging.
Finally, it seems a dubious engineering philosophy to design parts to fit a process. Here, it seems you have gone even further, and are proposing designs merely to rationalize using a process you have already picked-perhaps even to justify purchase of expensive equipment. Instead, process selection and design should go hand-in-hand. If the requirement is a vent cover, then it seems sensible enough to design a stamped piece, or a bent-wire piece, or potentially a molded or even machined piece. 3D printing is not appropriate for the requirement of a thin, plate-like shape with a pattern of holes, and inventing reasons to use a slower and more error-prone process does a disservice to your design, the process, and the application.
"Creating these islands makes the print slower, so we made a part an order of magnitude bigger. Definitely quicker to print"
Why do you print the vents vertically? And why not use infill patterns?
I was working on a 3d printed heat recovery vent last year. Kinda glad I never finished it because now I can build it with these ideas!
Thanks
Some great ideas, thanks!
I realise that for mass production you'd probably want to stack as many of those vents on the build plate as possible in one job, but printing the vent flat would solve a lot of the "needs support to print" designs issues with the first models
You never batch parts in mass production. Thanks for watching
@@slant3dwhay? In production you print parts one at a time?
@@evbunke2 3D printing two parts at once takes longer than printing the parts one at a time, as you have added additional motions to move from one part to the next. Slant3D also prefers to print parts with minimal contact to the build plate, as more contact area makes for more effort in removal of the part.
Another thing to note is that while these examples are flat plates, this could also be on some bigger part that necessitates a different printing orientation.
@@evbunke2 print failure in one part can cause other parts to fail as well, so you'll have wasted a lot more time and filament
nice, vid but i feel there are some missing info.
I assume there is a reason you don't want to print it flat on the plate, like u cant print a lot on a plate at once, making the big flat side of the print a no go?
even tho it might be a when its to be installed so the screw wont act like a vedge in between the layer lines.
The second example would that not be super easy to print if just rotated so the slots are vertical and not horizontal?
He's using it as an example if you were in a scenario where the vent absolutely had to be printed vertically
Printing something like this vertically would exponentially change print times. Instead of printing a large flat cross section, with many edges that could separate from the bed (although a larger surface area as well) you can print a much smaller cross section more quickly, as there is much less travel while printing the part. That would be the most notable reason.
Printing flat on the plate is not always possible for the part. And the large number of holes increases surface area which increases print time.
You should see it more as a specific case where you have lots of islands. Of course in this very case if printed it flat, it would solve the problem. But learning wise it would be pointless.
so to conclude if you do this at home just print it flat with no supports.
that was a pretty long add just to say you can make something with less airflow
For things like the holes I just print it flat, write some G code to kick the part off and print it again.
Trying to find a problem for a solution
I'm going through this process for a custom solar generator. Since the equipment will be used outdoors it needs to be at least IP40 (not waterproof, but that ambient rain and splashes of water don't get into) and the 3d design gives so many tools for that instead of just punching or drilling holes for a vent.
Right now we're on tilted triangle holes, so water doesn't reach the fan and it drips out the little canal on the triangle's tip. For sure I'll be trying some of the ideas here.
Having a 3D printer is a licence for so many bad ways to design things. That hurt my head watching you show terribly inefficient cooling paths and missing the actual good possibilities in the technology.
I mean… these re cool ideas but I would be curious if these are actually an improvement. In my eyes using simpler covers in combination with ducts to direct or manipulate airflow sounds a t cheaper, simpler and easier to design.
A really nice design concept to think about! Gave me an idea to incorporate some Tesla valves to create a one-way airflow.
That is a good idea. We didn't have time to include it in this video.
You could also print custom waveguides out of conductive filament. Previously impossible to manufacture structures are shifting several areas of thinking in multiple industries lol good stuff man
Why not just print the original flat on the bed?
The second example with straight slots is a perfect case for rotating it 90 degrees so the slots print vertically as their would be no bridging and no support would be needed if the slot is cut with rounded edges making an arch.
This is clearly a representation of the phrase "3d printers let us solve problems that didn't exist before" 😆
Just like cars nevers solved the problem of transport. Horses were just fine.
@@slant3d this is silly because long distance transport as well as caring for a horse are pretty big downsides
Reminds me of something even crazier. The other day I watched a video of a company that had an AI design organic piped burn chambers for a rocket engine to achieve desired characteristics snd then print in metal.
Brilliant! Totally just redeemed my plans to use pc fans for my grow tent exhaust. I was literally just about to scrap the idea bc they would let too much light in
BRILLIANT VID, VRO. LOVE THE INNOVATIONS!
This wHole video is great practical showcase.
Thanks for watching
Wow. Thanks for interesting ideas!
I get there may be a situation where you would need to print these in the orientation that you're talking about, but If I'm printing a fairly flat piece with slots in it, I'm going to lay it down on the bed. You're going to get a stronger print that way as well, because of the layer lines
What I often do in situations like this is to use a grid. But normally, FDM slicers really don't like grids, they will essentially play snake on the build plate and mess everything up. The trick here is to interleave columns and rows of the grid, with each of them being as tall as a layer height. They still mostly fuse, even though technically, it's just 50% of the material.
Would have been cool to try the grill pattern from the Mac Pro. That overlapping sphere geometry could be optimized for 3D pretty well I think
It is a cool pattern. But in printing is has the same problem as normal holes. Many small "islands" that can lead to failure in production
How to design a fan cover that is guaranteed to irreparably clog up in no time flat.
I think the throughline of the narrative has me confused:
"It's better to print slots than holes, so don't design with holes. Not here's all the things you can do with 3D printing using holes instead of slots."
Also, so many of those designs severely restrict the total airflow through the vent's area.
It might be useful for some novel consumer products, but for industrial applications, the stamped metal louver plate still seems to be the best choice, plenty of airflow, and the benefits of angled slots, namely preventing falling particles, falling liquids from entering through the vent.
This is a fantastic idea I could see useful in various projects. I’ve never thought about this way of vent design before, thanks for sharing! Will definitely keep this in my back pocket for future enclosures
Man, I absolutely love your focus on design in printing over just talking hardware etc like everyone else.
Keep it up guys! Loving every second.
Thank you so much for watching.
print vent as separate body without top and bottom and use slicer options for infill to choose appropriate inner structure. It will be the fastest print of the vent cover ever.
Great video presentation of the world of possibilities with 3d printing 🙂 thanks!
Glad you enjoyed it!
That last example is really good. I met someone at a conference in April this year who was doing just that - printing impossibly ducted copper heat sinks for high power components. When they handed me the sample on their exhibition stand, I had one of those moments where I just couldn't believe what I was seeing!
It is highly under utilized
Great channel!!
I did a vent once for my 3d printer filter de se gyroid infill and not printing the vertical using 👌
my mind all the time during this video, "print it laydown on bed"
Drastically increases cost in mass production and is slower
Unlike most of the other videos in this series, I'm not so convinced by most of these shrouds. I'm aware that your particular company has constraints (must print with minimal bed contact for auto-ejection system, etc.) that discourage flat objects, which explains the unusual and potentially weakened print orientations, but I'm really not convinced on the square grids of serpentine round holes and similar high-flow-restriction designs. Seems like you would need a VERY high static pressure fan to take advantage of them. Nice to keep the techniques in mind though, as it's entirely possible that you'll happen upon a use case where they ARE actually practical!
Seems like a hex pattern of hexagonal holes (or making vertically serpentine horizontal slits rather than holes) could pack the serpentine designs a lot more densely, providing far more airflow per unit area. Seems like just a diagonal would be far easier to clean than serpentine patterns, though.
I feel like printing this part in a vertical orientation is an edge case that's applicable for printer farms. I don't think this is the optimal way to print it. Not only will it take longer, but it's going to introduce more visual defects. All of this in the name of part ejection. If you're a hobiest, none of this is as applicable.
You could solve most of your challenges by changing the print orientation.
Great video, thanks!
I thought the curve was going to go upwards instead of sideways, to ensure it's printing one contiguous slice instead of many small ones.
Or Make it out of mostly gyroid infill, then its fairly optimal perhaps.
Even easier to just remove top and bottom layer (for a section or whole part) in the slicer and let the infill produce very fast printing grills that look awesome!
Erm for the slot one why not print flat, or just rotate 90 degrees so they are upright?
Could use testla valve filters to controll air flow direction
Ummmm, print them flat???
Wouldn't it make more sense to print vents flat on the bed?
to the point at 2:15 about slots needing support material.. why not just print the plate lying down? or print vertical slots?
For the example at 1:40
Why not print it laying flat? No supports needed, the thing becomes one of the easiest prints you can do if you just orient it the right way
I think they need to make a lot so they have to be put up down so the print takes less space on the bed but I don’t know if this is right
@@Burntsteak-ut6hm if you lay it flat, you can print it faster, compensating for less parts per plate
@@jackgamer6307 how did you respond so fast
@@Burntsteak-ut6hm I was on my phone and got a notification
@@jackgamer6307 k
I am at the beginning stages of production of my own inventions right now and the shear cost of molding just one of my 12 parts in their original design was $25k. That's just the mold, not material. I'm not even sure they could IM petg-cf. With my final production design the cost would come down since there are less "pulls" to do but it wouldn't matter. My parts cannot be molded. I'm sure a hundred people commenting here would love to say otherwise but a mold simply cannot eject my parts. And no you cannot simply design around it or modify it. Design for the process as stated. The only trouble is you cannot in my case, and yet I still need to make 10s of thousands of them if not more. If you have to make a mold then use cnc tooling, machining to cut into it as a completely separate process step then you are NOT injection molding parts, you are machining them. There is a huge difference between Injection Molding a part the way we all imagine it--stamp, drop, bin, repeat and what many of you commenting claim to be Injection Molding. If you have to use a mold, AND THEN use separate tooling on different machines to make cuts, holes, deburr or whatever the case may be then you are NOT injection molding in the traditional sense-you are machining. Sure the IM plant can come back to me and say "yes, we can make that." But making it happen in totality and simply stamping a part into a bin are 2 totally different animals. And the machining part I would imagine takes drastically more time than the molding part since one blade or bit can only cut one object at a time.
Hate to break it to you but a lot of, if not all of the designs you showed could be made using either basic tooling or positive to navigate mold construction, yes 3D printing does it far cheaper, far faster, and at a far larger scale. And for those who are wondering what positive to navigate mold construction is the process of making a master of the product/feature(s) you want then making navigate molds from the master to than make the final product. This is the whole "clone a thing" process but it can be used to make parts for castings.
What is the software used to make this stuff? Ive seen many other 3d print TH-cam channels use this software or softwares similar.
these things are so bad for airflow, you might as well duct tape the fan cover shut
"Why don't print it laying down?"
Because then it will take much more bed area and since he is talking about designing for mass manufacturing you can fit way more of those parts when placing them vertically.
The one with laying down configuration will add even more cost for just having someone to clear the bed so that another batch could be started than thr cost it saves by reducing post processing.
Love the channel! What printers do you use?
We design and manufacture our own printers for our giant print farms. The Megafarm is speced for over 3000 machines
I'm on 6 minute wathing this, and you still pretending that you know what Arflow is and air has no dencity by pointing on s-shaped tiny tubes. Let me help with this - Airflow is free inertional moving of air molecules. What you pointing is while taking that brick with few low-diameter holes will not give you airflow, it will be airstuck. Higthly inefficient. You want airflow - make lot of open space without any elements that can prevent air to move straight line. For the best airflow possible you just need one big hole or nothing at all. If you want to make a filtration - make an air filter and do not tell anyone that it "provides airflow" - it doesn't. It actually prevents airflow from happening by making obstacles. Any directional changing of air flow will take energy from the air.
mind blowing
Water soluble filament could be useful for this too. And not just for supports.
Say you want 1 thin line going through the mayerial.
Instead of trying to print a tiny hole, print a thin line of water-soluble filament. Then ramp up extrusion with normal filament and squish against the thin line.
In a sense, it's like adding a layer of subtractive manufacturing to it
One question... why don't you print it lying down?
Like why add supports? ... I mean, with mass production in mind, I kinda know the answer, sorry.
What about snake turn drills? It worked with tunnel drilling, but yes, we don't see them in parts manufacture.
Just print the vent horizontal
Not always possible
I'm sorry, are you telling me you print these standing up? Why? There is no valid reason to do that. You waste material with supports, you risk warping, and you now have to worry about snaping it along the layer lines. Printing these flat makes so much more sense. You dont have to jump over any gaps, the layers are set out along both the width and length so layer adhesion is much less of an issue, and as long as you have good bed adhesion there should be little to no warping. Additionally if you must print vertical with holes, hexagons with the point facing up are better, especially if they just need to be a hole and not hold any other parts. Also that first vent you showed is not typical, not if you put a fan behind it. It would be really restrictive. An actual vent that you would normally see in the wild usually has the holes done up in concentric circles with a set spacing to maximize airflow. That strange star pattern you made would be aweful in comparison. You would be better off with just a big hole.
One last thing, the reason we really use straight through vents and stamped ones with veins is not just because of the ease of manufacturing, but because straight through is the most efficient. And bends and turns only add resistance and thus lowers your cooling/airflow potential.
At some point you just straight up use infill geometry. Gyroid probably flows pretty well.
This is true. But you generally do not want to rely on the slicer to create geometry. It is best done in CAD
@@slant3d Hang on, why? Especially designing explicitly for mass production on FDM, if it's possible to create the geometry you need using infill, for which this is an excellent example, I would always want to take advantage of that, and design the part around the slicing profile I expect to use. Making the design parametric from the get-go can really help here, as the CAD can be tweaked and adjusted in step with the GCODE tuning. Some slicers will allow you to print two parts on the same bed with different settings, nothing says these can not be touching so as to fuse into one part, so I might sometimes combine an infill-only grating, with 1mm thick infill and just a couple of outer walls, with a more detailed surround at 0.6mm width and denser, standard infill, all from the same 0.6mm nozzle in the same part.
why not print them horizontally? (hole cylinders perpendicular to bed)
Not Viable for mass production
@@slant3d whats the reason for that? im totally not a 3d printing person, but somewhere in here was said that 1 part has to be printed at a time, so i assume both orientations would print in the same time, and its not like horizontal piece takes space from other pieces because its only one printing at a time
highlights:
- the guy goes off on some ill considered rant about vortex generators and laminar vs turbulent flow, and somehow managed to completely ignore (and catastrophically fuck up) porosity, static pressure, air flow impedance, ease of cleaning and maintenance...
- at one point he mocks up the weird disc with the holes redirecting to the sides thing with 5 holes, and a 5 bladed fan. does he not know why most fans have prime numbers of blades? our protagonist has just re-invented the old timey air raid siren!
- near the end our hero suggests 3D printing copper heat sinks. in a video about mass production
you can actually use 3D printing for all sorts of cool well optimized shrouds and ducts for a fan that would be difficult to make through other means, but the trick to doing that is to actually think about the application, not vaguely imagine what you think a fan does in an entirely abstract machine.
Allways these titles and thumbnails showing pics of iMPoSsIBlE oBJeCts
just say "conventionally impossible" please
It's really cool to see the kind of innovation that comes from a relatively new manufacturing technique. The further into the video you go, the more insane the geometry gets. And while I imagine you technically *could* manufacture these parts in other ways, it'd probably be stupid expensive and complicated to, and likely even then still likely with extreme compromise. For example, several smaller parts later joined together would be one way. It just wouldn't make any sense to do.
I was actually quite critical of the last Slant 3D video I watched (the mug handle video) because I felt it was very dismissive of other manufacturing techniques as antiquated. But this doesn't seem to be the case here. Instead it's just showing off the really cool things that are being made practical with the use of 3D printing. Doing so also provides solid examples for why someone might want to choose 3D printing over other techniques for high volume production. It does also show its limitations and potential ways to mitigate them. For example the basic vent designs with 3D printing may not be an ideal combination, but the slants work much better. (Though I assume this introduces more drag, something that may very well be an acceptable trade-off)
Some things that came to mind for me while watching these specific examples were the complicated hydraulic circuits that are often found in things like automatic transmissions. As well, I imagine you could do other crazy geometries such as Tesla valves. The whole idea of complex internal geometries offers a ton of potential.
Overall, I'd love to see more content like this, and it makes me really excited to see what other really cool things people can do with this technology or how principles already outlined here may be applied for other use cases!
Just print the piece flat and no supports needed for the first 2, and any word about difficulty with the basic vents is stuff much easier to do with traditional manufacturing methods, and so this 3d print ideas are solving an imaginary problem
The intent of the video is understandable, but is nothing more than 3Dprinting basics and promises. A tested, real world example with actual, indended performance, is absolutely necessary.
Relatively space
Facebook
Printing that orientation would serve well for higher volume because of how much you could fit on a plate, but that is no reason to complicated the design.
It really sound to me how to make your life more difficult. I understand why you wouldn’t place it face down but on the other hand it make for a easier and cleaner print and much stronger also.
That and can keep flying bugs from getting in too as well
I don't get your 'assumed' 3D printing orientation. Surely you would print those first examples flat on the build plate then you wouldn't get any of the problems you mentioned!
Higher cost to mass produce
@@slant3d It seems that there's many things here that hobbyist 3D printers find unintuitive, because they think in terms of 3D printing at home, not mass production. Like if you want slots and can't print it flat, surely you'd orient the slots upward? Then there would be nothing to bridge.
So you really never thought to lay the part flat on the bed instead of building it on it's edge? Slicers do have a feature that allows you to reorient your part - you know that right?
what does he carry?
I couldn´t follow for a second, cause I couldn´t understand, why there´s any advantage in printing those vertically, while printing them flat would have eliminated the issues in the first place...
I'm no engineer, so take my ideas with a pinch of salt.
If we've designed a fan plate to act like a dust filter, then we have designed a dust filter, haven't we?
A dust filter that would either clog up and give us zero airflow, or give us poor airflow and let all the dust in anyway.
So this hybrid is both a bad fan plate, and a really bad dust filter.
I dont say it's impossible to design an actually useful fan plate that uses 3D-printed micro-structures (easilly plugged tiny canals, that kill the airflow), but then we'd need to have an entire structure designed to take the air in, swirl it around to separate the dust, eject the dust through side ports, and let the clean air inside. I have never seen a dust filter that doesnt require maintenance.
It seems like we'd need wuite a fan behind this "fan plate" to get any air inside through that.
Or maybe itd be possible to nake afan plate from mumtiple layers of sparse grids of filament that act like a dust mesh... Wait... Uhh..
What about an ionizer-based 3D-printed air filter? It ionizes the dust particles and makes them stick to the walls of... Our chasis.
Maybe we xould 3D print a large block that'd have relatively wide canals - say a mm. Thatd gently wobble the air left and right and provide extra paths for the dust to fall into due to inertia... Then wed just need a tray underneeth to catch it all sonwe can clean it every now and then - or just drop it on the floor and let sonething else deal with it.
If this fan plate.. I mean filter block works, we could revolutionize the industry.
Or evolutionize ourselves out of the business...