Great stuff! It's very true that not enough time is spent designing the inside, I've made this mistake myself several times. This is a great example of how much better a part can be with a bit more consideration during design.
In this example with a "tower" there's a slicer solution that doesn't require a change in a model. Use cylindrical modifier that goes to the bottom and change infill to 100% under that protrusion (or play with other parameters). Very useful especially if you use someone else's stl.
@ Ah, thank you. I have tried PS before but, honestly, I could not get the hang of it. Probably because I learned with Cura. I'll look more into Orca though, I have heard good things.
@@RoseKindred Hi Rose. In cura you can use "support blocker" to create a cube that intersects with the model and then use "modify settings for overlaps" in "per model settings" to change in infill density
@@tharukakuruppu Ah, thank you. I never tried that way. I will look into it when I can. Typically I print with higher infill anyway, but I have noticed on my ant bait station there is a weakness in the screwhole posts I made. They looked much like this "tower" example.
The real "(re)design for 3d print" tutorial. That simple, that genius. Sometimes I do similar tricks, sometimes not. Should be shown to design team when you are "the printer guy" in company.
The real "(re)design for 3d print" tutorial. That simple, that genius. Sometimes I do similar tricks, sometimes not. Should be shown to design team when you are "the printer guy" in company.
You've strengthened that lateral force failure point. You've changed the cylinder from being solid to a shell, though. Your cylinder is less strong for compression and your prior failure point under lateral force is more likely to fail now under torsion. Still a good video on getting people to think about how they can design better.
Thanks! Yeah, I could have used a better example for sure. : ) We printed a similar part for a customer and didn't want to use his design for the video, hence the dumbed down version. This was designed for lateral force, just for the opening demo. If someone thinks differently about designing for 3d Printing, it was a win. Thanks for watching and commenting.
Some time ago I helped a friend with her mechatronics thesis. She designed a robot to be 3D printed, and I was the one who offered help with that, otherwise it would've been very expensive for her. Unfortunately, neither she or I were the ones to design the parts, rather her teammate who was absolutely zero skilled in 3D modeling, specially for additive manufacturing. Had I found this video back then, it would've been so useful, since his designs were extremely flawed precisely because of a bad implementation of pegs and whatnot. This was such a simple yet extremely effective solution, I'm definitely adding to my repertoire.
No one teaches engineers this very well, but: Use a peg for location, and a screw to hold something down. One screw and two pegs will locate and hold something against a flat surface.
LOL NOPE. Lets compare polycarbonate (one of the strongest material you can print) VS 6061-T6 aluminium alloy (one of the weakest popular metal alloys) Ultimate tensile strength: ~60 vs 310 MPa (PLA ~37MPa) Yield t. s. : ~80 vs 276 MPa (PLA ~55MPa) Modulus of Elasticity : 3 vs 69 GPa (PLA can be ~13 GPa) So you can see the difference is like 3 to 20 times in "strength". If you take into accound weak bonding betwen layers (70% of strength max, usualy around 40-50%) and fact that you can use good steel instead of average aluminium, the difference will be like 10 to ~200 times...
@@DrewLSsix Why? Maybe I wasn't specific enough... I fully agree that well designed (with 3d printing principles in mind) part will be waaaaay stronger than just designed part. I just don't agree that polimer can be nearly as strong as metal, because it is not true due to properties of materials. Those parts could be equally functional when metal strength is not needed or when you make fdm part way bigger. I just saw it worth mentioning because difference in strength can be like 2 orders of magnitude big, so imo comparing this type of materials isn't really adequate.
@@DrewLSsix Nah, he didn’t. The point was that geometry is the only thing that determines an object’s strength. He articulated in a very detailed manner why that is incorrect. It is possible (probable even) that the original comment missed the point, as silly as that is. I suspect the idea that was ineffectively communicated was that there are many functional parts which don’t need to be made of metal as long as the design process is adequate. Granted, this is only an educated guess, but it seems unreasonable to say the guy regurgitating relevant data sheets (in the way which most advocates what was originally said, btw) is the one who lacks understanding of what we’re talking about.
That’s a pretty smart way to solve the layer adhesion issue AND to use the infill material in a better way, thanks for sharing your knowledge on the topic!
I have been 3d printing for 3 years now. As a retired engineer, I can see you know very well what you are talking about. You have got a new sub. Bravo!
I have tried adding internal walls before, without much success. When adding an internal wall, in my experience, the slicer views the internal walls as external walls and therefore doesn't generate infill in the hollow areas of the print. I suspect this is why you have support material inside the part as opposed to infill? Either way, a really nice, concise, informative video!
Are you sure it's not layer time cooling slowing down the print? This is a very standard feature in slicers and if the layer is small, it slows down the print to allow it to cool better. This also causes the plastic to melt a lot better because it is going much slower, which in turn causes better layer adhesion. I suspect this is what is happening more than the hot plastic on cold theory. Most printers if not slowed by layer time, they push the limits of their hotend, which means less layer adhesion when adding more parts up to the point it stops slowing down because of layer time.
To further increase strenght.. print orientation so you have all the loads winthin a layer plane. For that part sideways print will need some support for the bin but it will not snap between layers. Much stronger. I'd also recommend adding a hole on the pin and insert a bolt/screw/rod afterward. The screw will also compress the layers resisting layer separation more if any tension appears. For really study pieces, the last resort before CNC is to have some metal hardware prepared (like some flat stock with a bolt welded, or tapped hole). then design the model to fit the metal hardware inside. Some clever orientation let you pause the print -> insert hardware -> unpause, having a complete 3dprinted outside. Pin down orientation would be beneficial for such hardware but will net more support on the flat face in air. Also in the improved sliced part, the top part has a layer or two of infill. I would increase top and bottom layers so there is no infill within shell wall. Als for using inside supports for shelled part - use 0 contact distance (soluble), also increase support thickness, alternate lines.. so the inside suppor can do something and not get loose and rattly inside. Another strength increase trick is to print a shell with a hole somewhere. Fill the inside with epoxy resin. Bonus points to add metal rods where possible as well.
Oh man, that's a gamechanger. I was under the impression that slicers ignored any internal geometry like that. I think it is something I saw in a video or read somewhere ages ago that got imprinted that misconception on my brain. Thank you for this!
How do you achieve both goals - shell and add the walls and then turn everything else into a solid for infill? You might have two things going on - need for the pin to be strong against bending and also the entire structure strong against crushing.
So, we modeled the boss to make it stronger. If we also had a problem with it crushing, add a web in the middle. But, the internal radii strengthens the deck too. The model is on our site under 3D Printing for evaluation. Nice comment, and good thoughts!
5:34 Most people never mess with the cooling fan speeds, but in this scenario turning the fan real low or off, would work great in keeping the "part material" at the right temp to keep it soft .
Thanks for the tip about the layer adhesion on multiple parts. I’ve actually had this happen to me where a print of 2 worked fine, but a print of 10 failed. I bet this is why
I find this very interesting, but wonder if you left something out? First of all, I didn't realize that anyone was shelling parts out for printing parts. You may have said something I didn't pickup on, but how do you get the printer to take a shelled out part and print it as though there's material on the inside? There is no material on the inside void of the shell, at least in the CAD model unless you did something more in CAD that you didn't seem to show? So I'm guessing it might have something to do with something you said without explanation? Something about printing it as a shell? I'm pretty new to, and only beginning to learn the slicer software, but I'm not new to CAD.
Good questions. If you use Fusion 360, you can download the 3D Model from our site. It has the history of the modeling changes. Shell is just the perimeter passes, or Thickness of the walls. After we shelled the part, there can be no infill. But, the slicer will add Support material to the inside. This is a technique that allows you to reduce material, target stress points, and decrease cycle times for production runs. Thanks for watching!
Awesome stuff thank you. I caught your video on my feed last night and immediately went back to the drawing board on a part I designed. Using this same technique I revised the geometry in Fusion and did a new test print and it's essentially unbreakable now by hand. I'd have to take a hammer to it so thank you very much for this quick tip. Really good stuff 👍
Great tips. Really wish CADs can incorporate (and extend upon) the support generation features of slicers since it's IMO a step in designing and not manufacturing. Doing so also means we can have an easier time to do analysis (center of mass/FEM etc.) on the parts.
very useful information, I've printed thousands of parts and have never thought to add internal structures/walls in increase part strength... bahh. thank you!
The printer slowing down for the short layer times of that post has the biggest effect, much more so than the previous layer being cool. Pressing in a metal pin or screw from the back will give you ultimate strength and let you print many very quickly.
Love this 🤘🤘 I just had this issue with some yard signs I made for my daughter's birthday party. I thought something similar would be beneficial, so nice to see the confirmation on that!
All you did was essentially double the wall thickness by shelling it, because theres now an inner and an outer wall, which is why there is no more infill on the "inside". Your infill is now between the two walls as you can see when reviewing the slices
75% of that boss was printed on infill and 4 surface layers. You would need 100% infill to do what we did here. Now, we need now infill, and still get 100% of the boss build onto profile passes. This isn't a great example but just a concept.
I tend to agree that much of this was superfluous. Extending the handle down to the bottom of the part? Thats not gonna solve the FDM layer separation issue with this type of feature. The handle isn't breaking because the part is flexing so much on the top surface. It's breaking at the layers, because its a small cylinder with only a few perimeters. Making the handle more solid or adding the radius is about the only effective method other than print orientation.
I hear you. But keep in mind, those profile passes for the boss 0:03 extend all the way to the bottom of the part. And there’s radii on the outside and inside of the part. Give it a test, and let us know what you find.
@@LeCafeRacer we experienced no layer adhesion issues for either part. The issue was printing on top of infill versus profile passes. The results speak for themselves. If you give it a shot, let us know what your results show.
interesting ideas! I'm fortunate to only have to print the things I design myself, and I realized looking at how odd the example part seemed to me that it's a kind of feature I have practically subconsciously avoided, probably realizing it would not come out strong without even thinking about it, which is wild. extra note- turning off the cooling fan and extruding 20-50 degrees above what the filament recommends, plus lowering layer height to 50% nozzle width and upping extrusion width to 125-150% is a great way to make large or multiple parts still have good adhesion, though it will degrade bridging, angled overhands, and surface finish.
hey man, just an idea, that at least would help me a lot: would you mind showing off how you handle the sequential printing, in specific I mean modifiying the gcode. I did test that a little, with copy pasting 5 gcodes into 1 master gcode, but I still struggle a lot with it, and so I had the Idea if you maybe wanna share your skills and knowledge about this topic with us? would be very much appreciated, thanks a lot, cheers :)
Maybe I'm missing something but from everything I know about 3d printing, aren't model files saved as solid 3d volumes? How would you actually define an internal wall inside of the model like that? Maybe I'm missing something but it seems like that's something you'd need to do in the slicer, because the model itself should just be exported as a solid object.
Yeah, instead of leaving it solid, we "shelled it" and left walls. This way, you're not building a surface on infill. When you design anything, it should be designed with the manufacturing method in mind.
@@ChipMonkeys Yes but my point is, on a technical level, that shouldn't be possible with how meshes are exported and sliced as far as I'm aware. Any closed shape is by default sliced as a filled volume and as far as I'm aware file formats like .stl literally just export meshes, which wouldn't be capable of modeling internal solid geometry like that. I've seen people do stuff similar to this by making micro-cuts internally to make the slicer add additional "walls" inside of the part, but you seemed to somehow actually directly model internal geometry in a way which is preserved by the slicer. Are you saying that you literally turned the part entirely hollow and removed *_all_* of the infill throughout the entire thing? I interpreted your wording as in you were replacing the infill in that area with solid walls, not removing literally all of the internal infill but that's just about the only way I could see this working within the constraints of how modern CAD and slicer software works. The thing is, that'd be making a pretty massive compromise strength wise. One particular area of the part would be incredibly strong sure, but the rest of the part has 0% infill and is an empty void which would make it weak and, depending on the part, make it not even print correctly in the first place due to massive internal overhangs. If that's the case, then adding 'micro-cuts' would likely be a better method to strengthen the part in this manner since you could add internal wall-structures without losing infill through the rest of the part. (at least, a better method that isn't slicer-dependent. I'm sure some slicer has some fancy way of doing this builtin, but micro-cuts are the only slicer-agnostic way of doing it that I'm aware of) I'm not sure if fusion360 allows you to do this or not, but I'd be really interested in seeing a wireframe exploration of this part to understand exactly how it's actually being modeled as a mesh that the slicer can understand.
@@robonator2945 Imagine you download an STL of a Glass bottle, then a ship. You then place the ship in the glass bottle and print them at once. It would print a ship in the bottle. It wouldn't just print the bottle. If you model internal structures, they will print too.
@6:15 you notice the parts dont complete in order, but at random. there is a way so you can have it go row by row so the extruder doesn't knock into the others. unfortunately the slicer doesn't really have a way to organize this how you want, but if you spend time figuring it out, you can make it work.
Great tips. Let me just be a bit nerdy and add to this that parts printed with a larger nozzle, such as 0.8mm, are generally stronger than those printed with a smaller nozzle, like 0.4mm. Let's compare your part (the pin you broke) with a hook, designed for wall mounting and hanging items ( just to make a point here, and draw from some data my company did quite recently). The advantages of using a larger nozzle become evident through specific tests: Tensile and Flexural tests that are assessing the structural integrity and resistance to bending forces critical for a hook/pin bearing loads, larger nozzle prints show improved layer adhesion, enhancing strength under tensile and bending stresses. 2. Evaluating the hook's toughness and its capacity to absorb energy (impact test) without fracturing, larger nozzle prints are likely to exhibit better resistance due to stronger layer bonds. 3. Determining the hook's endurance under repeated loading, the enhanced layer cohesion from a larger nozzle contributes to a longer fatigue life. 4. Measuring how the hook deforms over time ( a form of creep test) under a constant load, prints from a larger nozzle might offer superior performance because of improved material continuity. These tests did highlight the strengths of using a larger nozzle for printing objects like hooks and pins, emphasizing durability, strength, and resistance to deformation. But its a quite big However….. however using larger nozzles on smaller parts can be challenging, often requiring multiple fans, even with ABS. But smart design methods like you demonstrated here can really compensate for a lot of this, again really great content 👌
That is some good information. Very inciteful and very well presented. Everything you say makes sense but certainly isn't obvious. Thank you for the video
Plus, for some uses, I make a hole with nut socket, thru all the "chimney", to reinforce it with long M2 or M3 screw. We even make serious mechanical parts for clients with this method.
I get the premise of the video, but generally hollowing a part out will make it a lot slower to print because those inside model faces will print with outside surface speeds. (Not sure if s3d allows you to tweak those individually). I think you would have gotten the same result by just putting a hole through the shaft on your part and leaving the rest solid. Make use of your slicer differentiating between outside and inside walls as well as infill speed and the infill being properly connected to the inside walls.
I saw some video with that technique. Some other slicers seem to have other options that might help too. But, like you said, this video was more about a concept, than an example. Thanks for your comment! I never thought people would watch it. haha
I frequently print by object instead of by layer as insurance. That way if something gets off on a long print I still have 2-3 valid pieces and just need to reprint the last couple.
You sorta glossed over the part about filleting the internal sharp corners, but that's also a very good design rule for 3D printed parts! Glad you showed it!
I've never done internal structures like that, because I thought it would make the solid non-manifold for some reason and create errors in the slicer! Good to know!
I know what you’re saying, I think it has to do with the difference between how a slicer sees a step files vs a STL file. Sounds a good research project. : )
I’d leave the “chimny” feature open on the top and insert a steel dull pin. Or you can pause the print before it closes the top and insert the pin at that point. That will give you a strong part.
That's a good tip! But it would still break off at the infill. It needs to be modeled on the inside. Then the steel pin would go throughout the part. That would be crazy strong. The next step would be machining it! haha
Yeah I did came across such similar issues.. we just have to bend our mind to way of design for print instead of design for machining😊. Thanks for making this video ❤
@@ChipMonkeysyeah, as for sequential printing that’s my only choice if I print multiple parts be it same or different. If dimension does not fit the bed I don’t print. Yeah do think printability as previously I design machinability too. Love your channels are great subscribed. ❤thanks for making us makers better by sharing ❤
I’m about to start running production prints and was thinking about this very same problem earlier today. I appreciate the valuable insight you shared. Can’t wait to see how it translates to my first production run!
3D printing is straight forward and pretty easy compared to machining. But when something goes wrong, there are so many factors to getting a good print. We struggle sometimes. Glad to hear you are getting involved!
Great VIdeo and exactly what I was looking for to strengthen parts. Do you know if this method of capturing the internal geometry is possible using Orca Slicer or Cura? I don't use s3D.
Yeah, you'll need to play around with wall & flat thickness to avoid think layers of infill. This print was stronger because we didn't print the boss on a flat surface that was built on 15% infill : ) Thanks for your good comment!
I often change the axis of the part and split them, so I can glue them later together, it is way stronger than if i would print it horizontal. The superglue by them self is like an extra layer you can't break. And the fact if you print your model on the side, you can't break also the little handle. But i do it only for mechanical things which have to be strong. Also, I print them slower for better layer connection and as minimum fan speed as possible, to let not cool down the lower layer so fast. Sorry for my bad English, it's not my first language . I just work as innovation developer and consultant, so i also have to work very often with 3D Printers but also way other filaments in the high temperature zones like PI or PAI or PFA/PVDF. But it works for every filament in the same way if you want to get the max out of it.
Love it! Very rarely do I see this talked about. Only one question. Are you using a "3D mouse" to orbit and zoom around the part? Sometimes I see these weird fluid movements in videos and I am wondering what input device they are using, because it never looks like that when I just use my regular mouse.
That’s a 3D Connexion controller. I bought a few of them, but it looks like they went up! www.amazon.com/3Dconnexion-SpaceMouse-Wireless-universal-receiver/dp/B079V367MM/ref=mp_s_a_1_2?adgrpid=57563685513&dib=eyJ2IjoiMSJ9.TiHiPktVgJmYlWNZBjltpWbGV9VexWKwf58nqtxDv2UwtHtnR8d8GOKd2Cto_e1BM_2JDp2-8XQ_Jl9asjUam5XQK88nluWXB0ozUkvPXp3XKQz4qIXfWTrq_kOwQig9euazpsKBzDPbUeIEtduHcFxnAprPluo5l-TTd8vob56aktu7pE1iAFnKG1nwdtvdm5dmTCBpUS0YLG0KyKYXfQ.XXpE8Z1bzkF-habs67GQ_Yh4m8R1temKlheerBvRxzo&dib_tag=se&hvadid=651162589606&hvdev=m&hvlocphy=9013299&hvnetw=g&hvqmt=e&hvrand=14994002162514758618&hvtargid=kwd-296149135190&hydadcr=20137_13445650&keywords=3dconnexion&qid=1710159678&sr=8-2
Wow, what an eye opening video. Thanks for the heads up about how they need to be printed one at a time. Sound's like maybe you learned that the hard way.
Lots of good tips here. I do got to do this sometimes but then a voice in the back of my head says; you cant do that because the top surface wont bridge well over that gap, (without infill) but ofcourse theres not reason you cant use internal supports! I stopped using supports that build off of the model some time ago because I had so many poor results trying to remove all reamains of the support, but ofcourse that doesnt apply to internal stuff. Good to have a reminder. ------------ Another simple way you can achieve the same thing is by making a small hole in your peg that goes all the way down to just above the base of your model. Ideally this hole has a radius equal to that of your peg minus your intended perimiter/wall loop thickness. Though you can just play with the loop count till they match.
@ChipMonkeys haha no it's all good I understood what your trying to say I think. I have been thinking lately about how it is probably best to avoid Infill from the conception of how you will model most parts. In terms of being able to control where the internal bracing is relative to what you are actually doing with the part, cross braceing, things like the fillets in the internal corners, But also in terms of potentially making the switch to other production methods easier like injection molding, machining etc. As a way of potentially limiting how far you part might stray from something that wouldn't require total re design to your assembly to dfm it properly. Anyway, I'm an industrial designer so that's how I'm thinking about it.
I think you’re fine without interior supports when you have a flat (horizontal) “roof”. That bridges fine for such a small area. Double the size and I’d probably suggest differently. Slopes would be different.
a few years ago, in 2019, i ran into this issue, i just used the "cut" option in my slicer to separate the part, and thats it. The slicer sees it as a floating piece and will ask if you want supports, just hit no. and BOOM, you have a more solid piece
Prusaslicer under output options. Have to set avoidance height and radius so your nozzle won't hit the last completed part or so your gantry can't crash into one while traveling.
Great stuff! It's very true that not enough time is spent designing the inside, I've made this mistake myself several times. This is a great example of how much better a part can be with a bit more consideration during design.
Thanks for watching and the well thought out comment.
In this example with a "tower" there's a slicer solution that doesn't require a change in a model. Use cylindrical modifier that goes to the bottom and change infill to 100% under that protrusion (or play with other parameters). Very useful especially if you use someone else's stl.
Can I ask what slicer you are using? That seems like a great tip and I am not familiar with it, but I only use cura for FDM.
@@RoseKindred You could try Prusaslicer or Orcaslicer for this. :)
@ Ah, thank you. I have tried PS before but, honestly, I could not get the hang of it. Probably because I learned with Cura. I'll look more into Orca though, I have heard good things.
@@RoseKindred Hi Rose. In cura you can use "support blocker" to create a cube that intersects with the model and then use "modify settings for overlaps" in "per model settings" to change in infill density
@@tharukakuruppu Ah, thank you. I never tried that way. I will look into it when I can. Typically I print with higher infill anyway, but I have noticed on my ant bait station there is a weakness in the screwhole posts I made. They looked much like this "tower" example.
Excellent ideas and honsetly the best practical ideas I'v seen in a while.
Thanks Terry!
Literally the best design video I have seen in months. I never would have thought of this and it helps so much.
Thanks!
The real "(re)design for 3d print" tutorial. That simple, that genius. Sometimes I do similar tricks, sometimes not. Should be shown to design team when you are "the printer guy" in company.
The real "(re)design for 3d print" tutorial. That simple, that genius. Sometimes I do similar tricks, sometimes not. Should be shown to design team when you are "the printer guy" in company.
You've strengthened that lateral force failure point. You've changed the cylinder from being solid to a shell, though. Your cylinder is less strong for compression and your prior failure point under lateral force is more likely to fail now under torsion. Still a good video on getting people to think about how they can design better.
Thanks! Yeah, I could have used a better example for sure. : ) We printed a similar part for a customer and didn't want to use his design for the video, hence the dumbed down version. This was designed for lateral force, just for the opening demo. If someone thinks differently about designing for 3d Printing, it was a win. Thanks for watching and commenting.
Great tip, never thought of doing that before but makes perfect sense, thanks! 👍
Great insight! Thanks for sharing!
Glad to hear you took something away from this video you can use!
thanks for the great tip i can use it right away
That's awesome! I'm glad it was useful.
Some time ago I helped a friend with her mechatronics thesis. She designed a robot to be 3D printed, and I was the one who offered help with that, otherwise it would've been very expensive for her. Unfortunately, neither she or I were the ones to design the parts, rather her teammate who was absolutely zero skilled in 3D modeling, specially for additive manufacturing. Had I found this video back then, it would've been so useful, since his designs were extremely flawed precisely because of a bad implementation of pegs and whatnot. This was such a simple yet extremely effective solution, I'm definitely adding to my repertoire.
No one teaches engineers this very well, but: Use a peg for location, and a screw to hold something down. One screw and two pegs will locate and hold something against a flat surface.
Nice! Intentionally designing for 3D printing makes a massive difference. I've found some engineered prints can be nearly as strong as metal.
LOL NOPE. Lets compare polycarbonate (one of the strongest material you can print) VS 6061-T6 aluminium alloy (one of the weakest popular metal alloys)
Ultimate tensile strength: ~60 vs 310 MPa (PLA ~37MPa)
Yield t. s. : ~80 vs 276 MPa (PLA ~55MPa)
Modulus of Elasticity : 3 vs 69 GPa (PLA can be ~13 GPa)
So you can see the difference is like 3 to 20 times in "strength". If you take into accound weak bonding betwen layers (70% of strength max, usualy around 40-50%) and fact that you can use good steel instead of average aluminium, the difference will be like 10 to ~200 times...
@@tomekpekiyou missed the point entirely. Congratulations.
@@DrewLSsix Why? Maybe I wasn't specific enough... I fully agree that well designed (with 3d printing principles in mind) part will be waaaaay stronger than just designed part. I just don't agree that polimer can be nearly as strong as metal, because it is not true due to properties of materials. Those parts could be equally functional when metal strength is not needed or when you make fdm part way bigger. I just saw it worth mentioning because difference in strength can be like 2 orders of magnitude big, so imo comparing this type of materials isn't really adequate.
@@DrewLSsix Nah, he didn’t. The point was that geometry is the only thing that determines an object’s strength. He articulated in a very detailed manner why that is incorrect.
It is possible (probable even) that the original comment missed the point, as silly as that is. I suspect the idea that was ineffectively communicated was that there are many functional parts which don’t need to be made of metal as long as the design process is adequate. Granted, this is only an educated guess, but it seems unreasonable to say the guy regurgitating relevant data sheets (in the way which most advocates what was originally said, btw) is the one who lacks understanding of what we’re talking about.
@@tomekpeki wait a second... you're saying plastic isn't as strong as metal? I'm gonna need to seem some sources to back up those numbers
this stuff should be optimized by the slicers so they maximize strength
That’s a pretty smart way to solve the layer adhesion issue AND to use the infill material in a better way, thanks for sharing your knowledge on the topic!
Very cool idea. This was awesome! I almost never do this, but you earned a sub from the first video of yours I've watched.
Thanks Man!
I have been 3d printing for 3 years now. As a retired engineer, I can see you know very well what you are talking about. You have got a new sub. Bravo!
Thanks! We learned the hard way here. : )
I have tried adding internal walls before, without much success. When adding an internal wall, in my experience, the slicer views the internal walls as external walls and therefore doesn't generate infill in the hollow areas of the print. I suspect this is why you have support material inside the part as opposed to infill? Either way, a really nice, concise, informative video!
Yeah, infill turns to Support material. You should be getting your strength for the Wall count. Inside and outside. Glad you took something away : )
Are you sure it's not layer time cooling slowing down the print? This is a very standard feature in slicers and if the layer is small, it slows down the print to allow it to cool better. This also causes the plastic to melt a lot better because it is going much slower, which in turn causes better layer adhesion. I suspect this is what is happening more than the hot plastic on cold theory. Most printers if not slowed by layer time, they push the limits of their hotend, which means less layer adhesion when adding more parts up to the point it stops slowing down because of layer time.
Good thought! We printed these 1 at a time with all the same settings. The only change we made was modeling the inside.
great video and tips. Being new to 3d modeling I would be interested in a tutorial of how you duplicated the part and then made the changes.
Are you using Fusion 360? If so, I uploaded the model to our site with the History so you can see set by step.
@@ChipMonkeys I am using Fusion 360
@@ChipMonkeys I am using Fusion360, I looked on your site but was unable to find it.
@@ChrisOndrovic www.cnctutorials.com/3dprinting/
@@ChipMonkeys thanks got it
To further increase strenght.. print orientation so you have all the loads winthin a layer plane.
For that part sideways print will need some support for the bin but it will not snap between layers. Much stronger.
I'd also recommend adding a hole on the pin and insert a bolt/screw/rod afterward. The screw will also compress the layers resisting layer separation more if any tension appears.
For really study pieces, the last resort before CNC is to have some metal hardware prepared (like some flat stock with a bolt welded, or tapped hole). then design the model to fit the metal hardware inside. Some clever orientation let you pause the print -> insert hardware -> unpause, having a complete 3dprinted outside. Pin down orientation would be beneficial for such hardware but will net more support on the flat face in air.
Also in the improved sliced part, the top part has a layer or two of infill. I would increase top and bottom layers so there is no infill within shell wall. Als for using inside supports for shelled part - use 0 contact distance (soluble), also increase support thickness, alternate lines.. so the inside suppor can do something and not get loose and rattly inside.
Another strength increase trick is to print a shell with a hole somewhere. Fill the inside with epoxy resin. Bonus points to add metal rods where possible as well.
I think you have a lot of interestimg knowledge.
Thanks!
@@noxenia Thank you for apreciation.
I am passionate about mechanical engineering and i do have studies for it (phd).
I like the way you approached this. And mentioned the sequential parts for the layer adhesion.
Oh man, that's a gamechanger. I was under the impression that slicers ignored any internal geometry like that. I think it is something I saw in a video or read somewhere ages ago that got imprinted that misconception on my brain. Thank you for this!
How do you achieve both goals - shell and add the walls and then turn everything else into a solid for infill? You might have two things going on - need for the pin to be strong against bending and also the entire structure strong against crushing.
So, we modeled the boss to make it stronger. If we also had a problem with it crushing, add a web in the middle. But, the internal radii strengthens the deck too. The model is on our site under 3D Printing for evaluation. Nice comment, and good thoughts!
5:34 Most people never mess with the cooling fan speeds, but in this scenario turning the fan real low or off, would work great in keeping the "part material" at the right temp to keep it soft .
Quick, to the point informative video; and I don’t feel like I missed anything either. That was refreshing.
Glad to hear it!
Brother, I hadn't thought of trying this. I'm on it, thanks!
Thanks, I just used this in a little hand crank for my bandsaw tensioner. Printing now!
Well, it finished printing and holy cow is this thing strong! Definitely adding this to my bag-o-tricks, thank you so much!
That's awesome! Thanks for the update! Glad to hear you got some good results.
Thanks for the tip about the layer adhesion on multiple parts. I’ve actually had this happen to me where a print of 2 worked fine, but a print of 10 failed. I bet this is why
Been printing for years and actually learned something. You should have way more followers. Liked and subscribed 👌
I find this very interesting, but wonder if you left something out? First of all, I didn't realize that anyone was shelling parts out for printing parts. You may have said something I didn't pickup on, but how do you get the printer to take a shelled out part and print it as though there's material on the inside? There is no material on the inside void of the shell, at least in the CAD model unless you did something more in CAD that you didn't seem to show? So I'm guessing it might have something to do with something you said without explanation? Something about printing it as a shell? I'm pretty new to, and only beginning to learn the slicer software, but I'm not new to CAD.
Good questions. If you use Fusion 360, you can download the 3D Model from our site. It has the history of the modeling changes. Shell is just the perimeter passes, or Thickness of the walls. After we shelled the part, there can be no infill. But, the slicer will add Support material to the inside. This is a technique that allows you to reduce material, target stress points, and decrease cycle times for production runs. Thanks for watching!
Really great Advice! I'll be putting it into action!
Thanks for your comment! I'm glad you have an application in mind.
Awesome stuff thank you. I caught your video on my feed last night and immediately went back to the drawing board on a part I designed. Using this same technique I revised the geometry in Fusion and did a new test print and it's essentially unbreakable now by hand. I'd have to take a hammer to it so thank you very much for this quick tip. Really good stuff 👍
Great detail, very useful. And your production tip is amazing!! Thank you!!! Watch to the end!
Great tips. Really wish CADs can incorporate (and extend upon) the support generation features of slicers since it's IMO a step in designing and not manufacturing. Doing so also means we can have an easier time to do analysis (center of mass/FEM etc.) on the parts.
Good info. Clear, to the point, no jackass dramatics. Thanks.
Haha
that was some really valuable knowledge. Massive respect, please keep on making more!
Great info! Thank you 🙏
I hope we can make more! Thanks for your comment!
very useful information, I've printed thousands of parts and have never thought to add internal structures/walls in increase part strength... bahh. thank you!
The printer slowing down for the short layer times of that post has the biggest effect, much more so than the previous layer being cool. Pressing in a metal pin or screw from the back will give you ultimate strength and let you print many very quickly.
Love this 🤘🤘 I just had this issue with some yard signs I made for my daughter's birthday party. I thought something similar would be beneficial, so nice to see the confirmation on that!
I'm glad to hear you made some use of this tip! Thanks for letting us know!
This is unbelievably valuable information, thank you for sharing! A new standard in design for 3DP!
Thanks for letting us know!
All you did was essentially double the wall thickness by shelling it, because theres now an inner and an outer wall, which is why there is no more infill on the "inside". Your infill is now between the two walls as you can see when reviewing the slices
75% of that boss was printed on infill and 4 surface layers. You would need 100% infill to do what we did here. Now, we need now infill, and still get 100% of the boss build onto profile passes. This isn't a great example but just a concept.
I tend to agree that much of this was superfluous. Extending the handle down to the bottom of the part? Thats not gonna solve the FDM layer separation issue with this type of feature.
The handle isn't breaking because the part is flexing so much on the top surface. It's breaking at the layers, because its a small cylinder with only a few perimeters. Making the handle more solid or adding the radius is about the only effective method other than print orientation.
I hear you. But keep in mind, those profile passes for the boss 0:03 extend all the way to the bottom of the part. And there’s radii on the outside and inside of the part. Give it a test, and let us know what you find.
@@LeCafeRacer we experienced no layer adhesion issues for either part. The issue was printing on top of infill versus profile passes. The results speak for themselves. If you give it a shot, let us know what your results show.
interesting ideas! I'm fortunate to only have to print the things I design myself, and I realized looking at how odd the example part seemed to me that it's a kind of feature I have practically subconsciously avoided, probably realizing it would not come out strong without even thinking about it, which is wild.
extra note- turning off the cooling fan and extruding 20-50 degrees above what the filament recommends, plus lowering layer height to 50% nozzle width and upping extrusion width to 125-150% is a great way to make large or multiple parts still have good adhesion, though it will degrade bridging, angled overhands, and surface finish.
Thanks for the suggestion! I'll give that a try.
internal fillets on the wall thickness sounds amazing to reduce stress concentrations!!!!
I still haven't broke that part. I really thought it was going to break. haha
First time I see this. It's brilliant!
Thanks Nick!
Very happy I came across your video. Extremely simplified explanations that're easy to understand. Thanks for the tips mate!
hey man, just an idea, that at least would help me a lot: would you mind showing off how you handle the sequential printing, in specific I mean modifiying the gcode. I did test that a little, with copy pasting 5 gcodes into 1 master gcode, but I still struggle a lot with it, and so I had the Idea if you maybe wanna share your skills and knowledge about this topic with us? would be very much appreciated, thanks a lot, cheers :)
We are working on it! There are ways to do it that do not require stitching code together. That's a good idea for a short video. Thanks for the idea!
@@ChipMonkeysthat sounds great! looking fwd for it
Maybe I'm missing something but from everything I know about 3d printing, aren't model files saved as solid 3d volumes? How would you actually define an internal wall inside of the model like that? Maybe I'm missing something but it seems like that's something you'd need to do in the slicer, because the model itself should just be exported as a solid object.
Yeah, instead of leaving it solid, we "shelled it" and left walls. This way, you're not building a surface on infill. When you design anything, it should be designed with the manufacturing method in mind.
@@ChipMonkeys Yes but my point is, on a technical level, that shouldn't be possible with how meshes are exported and sliced as far as I'm aware. Any closed shape is by default sliced as a filled volume and as far as I'm aware file formats like .stl literally just export meshes, which wouldn't be capable of modeling internal solid geometry like that.
I've seen people do stuff similar to this by making micro-cuts internally to make the slicer add additional "walls" inside of the part, but you seemed to somehow actually directly model internal geometry in a way which is preserved by the slicer.
Are you saying that you literally turned the part entirely hollow and removed *_all_* of the infill throughout the entire thing? I interpreted your wording as in you were replacing the infill in that area with solid walls, not removing literally all of the internal infill but that's just about the only way I could see this working within the constraints of how modern CAD and slicer software works. The thing is, that'd be making a pretty massive compromise strength wise. One particular area of the part would be incredibly strong sure, but the rest of the part has 0% infill and is an empty void which would make it weak and, depending on the part, make it not even print correctly in the first place due to massive internal overhangs. If that's the case, then adding 'micro-cuts' would likely be a better method to strengthen the part in this manner since you could add internal wall-structures without losing infill through the rest of the part. (at least, a better method that isn't slicer-dependent. I'm sure some slicer has some fancy way of doing this builtin, but micro-cuts are the only slicer-agnostic way of doing it that I'm aware of)
I'm not sure if fusion360 allows you to do this or not, but I'd be really interested in seeing a wireframe exploration of this part to understand exactly how it's actually being modeled as a mesh that the slicer can understand.
@@robonator2945 Imagine you download an STL of a Glass bottle, then a ship. You then place the ship in the glass bottle and print them at once. It would print a ship in the bottle. It wouldn't just print the bottle. If you model internal structures, they will print too.
Nice Vid 👍
Have a nice weekend 😊🐈🐾🐾
Thank you! You too!
Two great sugestions - thanks for a simple yet straight to the point video - kudos!
Thanks!
@6:15 you notice the parts dont complete in order, but at random. there is a way so you can have it go row by row so the extruder doesn't knock into the others. unfortunately the slicer doesn't really have a way to organize this how you want, but if you spend time figuring it out, you can make it work.
Yeah, this slicer can’t do it. What a bummer.
Great tips. Let me just be a bit nerdy and add to this that parts printed with a larger nozzle, such as 0.8mm, are generally stronger than those printed with a smaller nozzle, like 0.4mm. Let's compare your part (the pin you broke) with a hook, designed for wall mounting and hanging items ( just to make a point here, and draw from some data my company did quite recently). The advantages of using a larger nozzle become evident through specific tests:
Tensile and Flexural tests that are assessing the structural integrity and resistance to bending forces critical for a hook/pin bearing loads, larger nozzle prints show improved layer adhesion, enhancing strength under tensile and bending stresses.
2. Evaluating the hook's toughness and its capacity to absorb energy (impact test) without fracturing, larger nozzle prints are likely to exhibit better resistance due to stronger layer bonds.
3. Determining the hook's endurance under repeated loading, the enhanced layer cohesion from a larger nozzle contributes to a longer fatigue life.
4. Measuring how the hook deforms over time ( a form of creep test) under a constant load, prints from a larger nozzle might offer superior performance because of improved material continuity.
These tests did highlight the strengths of using a larger nozzle for printing objects like hooks and pins, emphasizing durability, strength, and resistance to deformation.
But its a quite big However….. however using larger nozzles on smaller parts can be challenging, often requiring multiple fans, even with ABS.
But smart design methods like you demonstrated here can really compensate for a lot of this, again really great content 👌
That is some good information. Very inciteful and very well presented. Everything you say makes sense but certainly isn't obvious. Thank you for the video
Plus, for some uses, I make a hole with nut socket, thru all the "chimney", to reinforce it with long M2 or M3 screw. We even make serious mechanical parts for clients with this method.
I get the premise of the video, but generally hollowing a part out will make it a lot slower to print because those inside model faces will print with outside surface speeds. (Not sure if s3d allows you to tweak those individually).
I think you would have gotten the same result by just putting a hole through the shaft on your part and leaving the rest solid. Make use of your slicer differentiating between outside and inside walls as well as infill speed and the infill being properly connected to the inside walls.
I saw some video with that technique. Some other slicers seem to have other options that might help too. But, like you said, this video was more about a concept, than an example. Thanks for your comment! I never thought people would watch it. haha
Great Tip. Really enjoyed your format and video length. Just right. Thanks for sharing.
Thanks for the video. You can still print all the parts but try annealing them afterwards and they should be stronger than the first model you showed.
Not all materials are good for Annealing. Like PLA : ) ABS, ASA, and PP benefit from annealing. This job was PLA. Thanks!
I frequently print by object instead of by layer as insurance. That way if something gets off on a long print I still have 2-3 valid pieces and just need to reprint the last couple.
Good video mate, simple and efficient. Thank you.
Very interesting! Quite a few fusion workflow tips in here i would've never considered - thank you!
Awesome! I learned the hard way : )
I thought of modeling the inside of a part before but never actually done it.. really need to give that shell modifier a try 🤔
Let us know about your results!
You sorta glossed over the part about filleting the internal sharp corners, but that's also a very good design rule for 3D printed parts! Glad you showed it!
I've never done internal structures like that, because I thought it would make the solid non-manifold for some reason and create errors in the slicer!
Good to know!
I know what you’re saying, I think it has to do with the difference between how a slicer sees a step files vs a STL file. Sounds a good research project. : )
Ok, all makes sense except how do you make it print each item sequentially?
Thanks! We are actively working on that video. Should be up this weekend.
Thank you! Great advices!
great video thanks
I’d leave the “chimny” feature open on the top and insert a steel dull pin. Or you can pause the print before it closes the top and insert the pin at that point. That will give you a strong part.
That's a good tip! But it would still break off at the infill. It needs to be modeled on the inside. Then the steel pin would go throughout the part. That would be crazy strong. The next step would be machining it! haha
What a great video lots of good info in here.
Great content man, thank you for the lessons there!
Glad you took something from it! Thanks for letting us know.
Very informative! Thank you!
Yeah I did came across such similar issues.. we just have to bend our mind to way of design for print instead of design for machining😊. Thanks for making this video ❤
Yeah, I'm always getting onto engineers about machinability! Now it's Printablitiy.
@@ChipMonkeysyeah, as for sequential printing that’s my only choice if I print multiple parts be it same or different. If dimension does not fit the bed I don’t print. Yeah do think printability as previously I design machinability too. Love your channels are great subscribed. ❤thanks for making us makers better by sharing ❤
@@Jim_One-wl4ke Thanks for the nice reply and for watching! I hope there is some educational value here : )
that's amazing information! thank you for sharing it and explaining it so well
Please more of that mix, design, slice, tips. Abo out!
Concise and helpful. Thanks for showing this.
Glad it was helpful!
Really useful concepts there, thanks for sharing!
Thanks for letting us know!
Very usefuls these hints❤❤
Thanks! Glad you could take something away from this.
I’m about to start running production prints and was thinking about this very same problem earlier today. I appreciate the valuable insight you shared. Can’t wait to see how it translates to my first production run!
Let us know how it goes! Thanks for watching.
Nice video and explanation. I’ve learned something new today. New to 3D printing
3D printing is straight forward and pretty easy compared to machining. But when something goes wrong, there are so many factors to getting a good print. We struggle sometimes. Glad to hear you are getting involved!
Great VIdeo and exactly what I was looking for to strengthen parts. Do you know if this method of capturing the internal geometry is possible using Orca Slicer or Cura? I don't use s3D.
Yeah, you'll need to play around with wall & flat thickness to avoid think layers of infill. This print was stronger because we didn't print the boss on a flat surface that was built on 15% infill : ) Thanks for your good comment!
I often change the axis of the part and split them, so I can glue them later together, it is way stronger than if i would print it horizontal. The superglue by them self is like an extra layer you can't break. And the fact if you print your model on the side, you can't break also the little handle. But i do it only for mechanical things which have to be strong. Also, I print them slower for better layer connection and as minimum fan speed as possible, to let not cool down the lower layer so fast. Sorry for my bad English, it's not my first language . I just work as innovation developer and consultant, so i also have to work very often with 3D Printers but also way other filaments in the high temperature zones like PI or PAI or PFA/PVDF. But it works for every filament in the same way if you want to get the max out of it.
Thanks for the tips! Your English is great.
Love it! Very rarely do I see this talked about. Only one question. Are you using a "3D mouse" to orbit and zoom around the part? Sometimes I see these weird fluid movements in videos and I am wondering what input device they are using, because it never looks like that when I just use my regular mouse.
That’s a 3D Connexion controller. I bought a few of them, but it looks like they went up! www.amazon.com/3Dconnexion-SpaceMouse-Wireless-universal-receiver/dp/B079V367MM/ref=mp_s_a_1_2?adgrpid=57563685513&dib=eyJ2IjoiMSJ9.TiHiPktVgJmYlWNZBjltpWbGV9VexWKwf58nqtxDv2UwtHtnR8d8GOKd2Cto_e1BM_2JDp2-8XQ_Jl9asjUam5XQK88nluWXB0ozUkvPXp3XKQz4qIXfWTrq_kOwQig9euazpsKBzDPbUeIEtduHcFxnAprPluo5l-TTd8vob56aktu7pE1iAFnKG1nwdtvdm5dmTCBpUS0YLG0KyKYXfQ.XXpE8Z1bzkF-habs67GQ_Yh4m8R1temKlheerBvRxzo&dib_tag=se&hvadid=651162589606&hvdev=m&hvlocphy=9013299&hvnetw=g&hvqmt=e&hvrand=14994002162514758618&hvtargid=kwd-296149135190&hydadcr=20137_13445650&keywords=3dconnexion&qid=1710159678&sr=8-2
@@ChipMonkeys Awesome, that's what I thought it was! Thanks!
Great idea. Thank you.
Thanks for watching!
A heated chamber will greatly improve layer adhesion btw. You should be able to print nonsequentially with much better strength
Thanks for the tips great video
Thanks for watching and commenting!
Great video! :)
Thanks!
Instantly subscribe. Great explanation! I learned a lot :)
"I designed my part so strong I can't break it by hand". Man that's not embarrassing, that's awesome!
Hahaha! I really thought it was going to break
Wow, what an eye opening video. Thanks for the heads up about how they need to be printed one at a time. Sound's like maybe you learned that the hard way.
I learn everything the hard way. haha. Desperation can be far more valuable than education sometimes.
Good tips, thanks.
Lots of good tips here. I do got to do this sometimes but then a voice in the back of my head says; you cant do that because the top surface wont bridge well over that gap, (without infill) but ofcourse theres not reason you cant use internal supports! I stopped using supports that build off of the model some time ago because I had so many poor results trying to remove all reamains of the support, but ofcourse that doesnt apply to internal stuff. Good to have a reminder. ------------ Another simple way you can achieve the same thing is by making a small hole in your peg that goes all the way down to just above the base of your model. Ideally this hole has a radius equal to that of your peg minus your intended perimiter/wall loop thickness. Though you can just play with the loop count till they match.
You're correct! I saw some using that method. I'm trying to introduce a concept of modeling. This was not the best example. : )
@ChipMonkeys haha no it's all good I understood what your trying to say I think. I have been thinking lately about how it is probably best to avoid Infill from the conception of how you will model most parts. In terms of being able to control where the internal bracing is relative to what you are actually doing with the part, cross braceing, things like the fillets in the internal corners, But also in terms of potentially making the switch to other production methods easier like injection molding, machining etc. As a way of potentially limiting how far you part might stray from something that wouldn't require total re design to your assembly to dfm it properly. Anyway, I'm an industrial designer so that's how I'm thinking about it.
I tried this on a print but prusaslicer would not infill at all.
The Slicer doesn't look at it as the "inside" anymore. Your model is just Walls or Shells now. Any infill will be Support material.
Thanks. That was very informative and useful.
I think you’re fine without interior supports when you have a flat (horizontal) “roof”. That bridges fine for such a small area. Double the size and I’d probably suggest differently. Slopes would be different.
You're right! I printed the real customer part without internal support. It bridged and printed great. Thanks for your comment!
Love this. Have just the part that needs redesigning
Let us know who it goes! I am really interested
But how i can tell my slicer to finish completely the 1st part first and after that go to the 2nd part?
We made a video called “All at once?” Or something like that. It walks you through that on multiple slicers.
ua-cam.com/video/23JI5JVAdeA/v-deo.htmlsi=lTnZihmTGOiYwjmH
@@ChipMonkeys thanks so much first for your video and 2nd for your answer! I never knew that this option even exists! Wonderful :)
Great tips!
Thanks!
I thought you will be wrong... but you were right. Good job. You got sub and like.
Well, thanks for being so thorough! If we ever mess up, I know you'll let us know!
@@ChipMonkeys 😁
Good vid. Thank you!
Thanks for watching!
a few years ago, in 2019, i ran into this issue, i just used the "cut" option in my slicer to separate the part, and thats it. The slicer sees it as a floating piece and will ask if you want supports, just hit no. and BOOM, you have a more solid piece
Hey, what slicer are you using?
@@ChipMonkeys prusa slicer has the cut option and a whole lot more. There tree supports are awesome as well
@@LT72884 Right on! We used the Prusa Slicer in one of our last videos. We will take a look at the Cut option! Thanks for the heads up!
@@ChipMonkeys you are welcome
thanks chip
Good info!
Nice. But, how do you print sequentially? Is there a slicer setting for this?
Prusaslicer under output options. Have to set avoidance height and radius so your nozzle won't hit the last completed part or so your gantry can't crash into one while traveling.
Thanks!