Achieve true 3D printing with non planar slicing

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
what if I told you that your 3d printer was really only working in two-and-a-half dimensions well technically that's correct and in this video I'll teach you how to do true 3d printing firstly this video is not clickbait although your 3d printer has movement for XY and Z three axes it pretty much never uses them at the same time each layer is stacked on top of the last with only x and y movements used to form the geometry in between layers the z-axis moves up and the x and y process continues yes I know that Zed hop lifts the Zed up and down during each layer but it's not really for the purposes of creating geometry the only thing that's close to true 3d printing is when we use spiral or VARs mode 3d printing for the vast majority is planar and that means it's done in flat planes stacked up on top of each other in the CNC world we would call this type of machining 2.5 D in 3d printing it produces the layer lines and step surfaces we see in our final objects to overcome this effect what we need is called non planar where the printer is moving at the same time in XY and Z to create a truly 3d tool path as it extrudes plastic and creates geometry in this guide I'm going to showcase the excellent work of others and if you follow along like me you'll be able to experience true 3d printing it's early days and there's still a lot of issues to overcome but the future is exciting so without any further delay let's begin there's a good chance you've seen this video doing the rounds because it definitely went viral it features an Ultimaker to printing in three dimensions check out the 3d perimeter around the outside of this shape and then it matches it with 3d infill on top like most people I was really impressed with the results and I thought I'd definitely have to give this a try some prints just aren't suited to 3d printing but this technique looked like it could overcome that I googled non planar 3d printing and I found this hackaday article from 2016 it was explained in a lot of detail with lots of technical terms and diagrams I figured since it was created before the most recent viral video I'd give this one a go first I followed the link to their github and I noticed the script at the top that does the actual post-processing I then followed all of the instructions from the github installing strawberry Perl on my Windows machine and configuring my installation of slicer I did actually get the script working and I was able to drag the resultant G code into simplified 3d to preview it it was pretty interesting too because it had non-planar layers the whole way through you can see across the side of the wing that actually bent and really compressed at the nose and tail I stopped testing at this point simply because I wasn't smart enough to understand all of the parameters and how I might tweak them to progress back to the original video then in the description of that video we had a link to a page with more details the video is embedded with the large before and after image it's here that we learned this work was conducted as part of a master thesis at the University of Hamburg this document is linked and is very very detailed it's clear that an immense amount of work has gone into this and for them to make the results public I'm very thankful there's a huge amount of calculation involved but as we can see in the evaluation section the final results are pretty special I was sold so I headed over to the github to discover that instead of being a plugin this was a whole custom version of slicer the instructions were pretty short to compile it but it was only for a Linux machine I headed back to the original video and to my despair saw in the comments that the original author was unable to compile it for Windows I went to the main slicer documentation and saw a link for building on Windows this page however was stupidly long and detailed but I figured I didn't really need to understand what I was doing I just needed to be able to follow the steps which steps to choose however was difficult there was about four or five different versions on how to do it on this page there was also a lot of links to other pages where we had to compile dependencies - I jumped headfirst down the rabbit hole on the Windows command line copying and pasting the commands from the instructions I got part of the way through but eventually I ran into errors and it just wasn't skilled or knowledgeable enough to be able to troubleshoot them I remembered that I had a Linux computer in the raspberry PI's I used to run octave rental I connected via SSH putty and started to follow the instructions again except it was stupidly slow raspberry-pi is not really intended for this type of heavy lifting and it wasn't long until I ran into errors therefore I needed a dedicated Linux machine from here on in is my process to get this thing working and it starts with setting up a virtual machine that means I ran up an emulated Linux computer inside my Windows 10 environment I've done this in the past but needed a refresher so I followed this guide that gave me a step-by-step on the settings needed in the software to set this up it's linked in the description for you to follow as well one nice thing is that this is completely free the VirtualBox software that we use is free as well as the Linux operating system and I chose to use Ubuntu once the VirtualBox was running I opened up the github inside Firefox which is inbuilt and then pasted the commands into the terminal one at a time the good news is everything here worked on hundred percent as it should have no errors and after not too long I was able to successfully compiled the version of slicer that I needed here we see the moment when I enter the last command and the slicer GUI opens up ready to use as you can see until you tweak things the window that you're working through is tiny pretty much unusable so I closed up slicer and navigated to the devices menu of VirtualBox and inserted the guest edition CD this will auto run and you simply keep clicking forward until the scripts are installed when it reboots all of a sudden you'll find that the interface is a lot bigger and now all of a sudden it's quite usable the next problem is getting files in and out of the virtual machine I found this great video tutorial on YouTube which I followed step by step to set up a shared folder that could be accessed in Ubuntu as well as my Windows host machine this video like all the other guides I'm showing is linked in the description after one more restart I was able to create a folder and as you can see all of the files are accessible on both a bun to a virtual machine and the Windows 10 host after slicing I could now whack them on an SD card to print from my Windows machine and that brings us to the next problem that needed solving and that was picking a printer with sufficient clearance the majority of 3d printers these days have nozzles that don't extend very far plaster shrouds and this was true for all of the printers at my house in the rare cases that the nozzle did protrude there was a probe nearby a fan shroud or in the worst cases both my end of three was the only printer that I had that was suitable for this and that's because I'd modified the hot end to have the Hiromi fan duct which is modular disassembling was really straightforward the part cooling fan simply pulls out and I had already wired energy pod connector so I unplugged that to release the fan completely next the fan duct is held in place with an m3 bolt on each side I simply took my allen key and loosened both of those before simply sliding it down the bottom and out of the way exposing the nozzle for the process that we need next to remove the VL touch the probe of which was far too low to bolts to undo that unplug it and all of a sudden we have the required clearance we need the final step being of course to reflash the firmware back to standard as per the original video I took measurements of 8 millimeters clearance with an angle of 45 degrees back in slicer we now need to configure it for our printer we can hit the COG next to the three options and then go through copying and pasting our settings from our existing slicer into this copy ru slicer for many years back in the day so I found it quite familiar but if you are switching from something else you should find that most of the parameter names match up or at least can be discerned with a bit of common sense one of the most important things to setup is the size of your printer under printer settings getting this one correct will ensure that the object matches in real life what you're seeing in the preview on the printer bed now we're up to the actual settings for non-planar printing we need to tick the box enter our angle and our height from the tip of the nozzle to the next chunkier spot after this save a preset with non-planar in the name so you don't get confused in slicer we can click Add to add our STL to the plate after this we click on preview to automatically slice the object and see the resultant g-code at first the output of the slicer was not generating any non-planar g-code and I was quite confused and if you've got this problem you can switch back to the terminal where you started the script and you'll have any error messages we can see here I have a combination of areas being too small or collisions being detected I didn't really get consistent results until I tweaked my settings I found it was necessary to increase the angle from 45 to 50 degrees and after I saved and re sliced everything started to work as expected performance inside the emulated environment is not as good as a native computer but it'll get the job done like most slicing software we can preview the g-code and we have a slider on the side to go through layer by layer we can see that the fuller majority of this object it builds it up with completely standard processes until it reaches the end in which case it starts layering non planar layers on top to build up to the final height if we're happy with our G code we simply click on export g-code and i save it to my shared folder so I can access it again back in Windows the first print I tried was the aerofoil that you just saw after I saw that the first layer was going down correctly I headed off to bed now there was a lot of hurdles to jump for me but if I had the benefit of hindsight and stuck with the method that did actually work I could probably get this whole thing set up in around about an hour I guess it's a good time to look at my results and in my opinion they're very encouraging as you saw earlier the first test print I did was an aerofoil and from the moment I saw the finished result I was hooked you only need to see it next to another aerofoil printed using planar traditional slicing to see the tremendous improvement in surface quality now last week at school I was helping a student 3d print a component from the major design project and had very shallow tops I therefore imported the SEL and did a back-to-back comparison the main problem here was that without a cooling fan the archway in the middle of the part looked pretty terrible the final result is pretty interesting but probably not the best demonstration of this technique next up wanting to recreate some of the surfaces seen in the original video I jumped onto CAD and made up some special shapes they started life as a simple rectangular prism and a simplified version of my TT logo cut from the middle I then applied additional cuts from the sides giving the top shallow curving surfaces this lack of others was very mesmerizing watching the extruder create the solid infill on top this first one has a single continuous convex top and in my opinion the results are definitely more pleasing next up I up the complexity of my cut making it like a very shallow wave I think this one also looks good but the real benefits were seen in my third test piece for the shape here I did back-to-back testing and I decided to cut the shape from one side and then from a second 90 degrees to the first what that gives is a lot more complex geometry on top even though I've smoothed it all that with some Phillips the difference in before and after here is massive with the non planar version just looking so much more aesthetically pleasing finally I wanted to make a shape similar to that seen in the demo video so I did another before and after of this new one this was the most complex geometry on top that I tried and in my opinion it gave the best results the non planar part looks so much better than the original without any of the staircase effect that we traditionally see with 3d printing if you want to have a go at printing this one yourself without going to the hassle of setting up the special slicer I'm gonna link the g-code down below but be warned you will need to modify your hot-end just like I did to avoid collisions so I'm really impressed by my testing results here but don't expect this to be rolled out into your favorite slice at anytime soon and that's because there's a number of issues firstly 3d printers aren't currently designed to handle this type of technique we would need to have a big shift in 3d printer design philosophy with manufacturers aiming to have much point here nozzles and have them extending far below fan shrouds and probes for auto bed leveling secondly we would need a lot more testing on the slicing side it's clear from the link thesis that a hell of a lot of research went into this and it's not going to be easy for other slices to roll out this type of implementation furthermore compared to the parameters that we usually tweak in our slicing software if you put in incorrect inputs for both the nozzle angle and clearance height it's going to have catastrophic results as everything collides and causes potential damage another limitation is that with current 3d printers the nozzle is designed to extrude sideways across a flat layer but in this instance we have slopes and therefore on the low side it's not going to quite meet properly and this can only really be fixed by being able to tilt the printhead over from side to side like you'd have on a 5 axis CNC machine all of that aside it does seem that this is the future especially when you look at all the positive comments left on the original video I'm certainly very appreciative of the work of the original author and have enjoyed following through and testing out something new hopefully this video gives you the steps to be able to pursue that as well and please let me know in the comments if you do so because I'd love to see your results thank you so much for watching and until next time happy true 3d printing g'day it's Michael again if you liked the video then please click like if you want to see more content like this in future click Subscribe and make sure you click on the Bell to receive every notification if you really want to support the channel and see exclusive content become a patron visit my patreon page see you next time
Info
Channel: Teaching Tech
Views: 1,436,325
Rating: undefined out of 5
Keywords: 3d printing, 3d printed, 3d print, non planar, no layer lines, 2.5D, guide, step by step, slic3r, test, how to, lesson, tutorial, linux, ubuntu, compile, process, future, virtual box, creality, ender 3, true 3d printing, free download, thingiverse, gcode, slic3r profile
Id: gmePlcU0TRw
Channel Id: undefined
Length: 14min 8sec (848 seconds)
Published: Mon Sep 02 2019
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.