Fully automated 3D printing using REAL print beds!

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
this is my voron 2.4 and like i've said in the videos in the past i absolutely love this thing uh the fact that it can print so fast just makes a lot of projects possible that would take weeks otherwise and now we can print them within one or two days but there is a problem with printing this quickly it also means that your prints are going to finish faster so for all these prints that i've done at most they've taken like two hours to complete which means if you have a lot of parts to print are going to be pampering the printer every two hours you have to you know to give it a bit of attention you have to clear off the bed you have to slice the next print and start it again so that's something that does slow you down over time and the obvious solution is to have some sort of automated print system in this machine but the solutions that are there right now aren't really up to what i would would say is fully usable like without any obvious downsides sure there's the creality cr-30 which is a great system for its application but for normal prints it is not that fast and you also need to print brims and do other tricks to actually get the parts to stick down to the belt and even then for me they often came out banana shaped which isn't exactly what you want for parts that should be flat then there are other systems that use a angled printer with a special bed surface where supposedly your prints just slide off but you know does that really work for prints that are you know less than five centimeters tall and then might adhere to the bed better than larger ones that can just slide off i don't know so the obvious solution for me is to have some sort of a bed swap system something that's automated and something that uses a real flex bed which is why i've ordered five more of these energetic pi coated beds and they're the same material as the stock but they are a bit smaller because i wanted to give myself a bit of room to work around and we're gonna try to build a automated magazine fed bed swap system that runs within the printer runs automated and theoretically you should be able to just you know start one print and the print is gonna eject the bed out the side pull in a new one and start the print again so let's see how well that goes i've already got most of the parts printed up this wasn't even like 24 hours of printing which is fantastic so yeah let's get to assembling all of these i've printed all these parts in the dust filament b plus ptg which is a line of imperfect filaments that are a result of color swaps in their extrusion line for example but they are pretty much perfect for printing prototypes like this and as a bonus you sometimes get this super tasty looking color change effect in your prints [Music] these all used to be 3d printers so i guess it's only fair to turn them back into 3d printer parts right and while i'm building the frame let's check out today's sponsor box sumo box will make 3d printer enclosures that keep the heat fumes and smells inside the enclosures and drafts pets and little hands out i've long had one of my pushers in a makeshift wooden box for printing polycarbonate and abs which aren't materials that really need that extra temperature but this is a lot nicer the boxmo enclosures come in three sizes to fit your printer and you can configure them as plain boxes or fully decked out with light ventilation and temperature regulation check them out at boximowingclosures.com or at the link below [Applause] [Music] [Applause] [Music] [Applause] [Music] [Music] so [Music] [Applause] [Music] so [Applause] [Music] [Applause] [Music] [Applause] so here is that heated bed or that heated bed base temporarily mounted to what is essentially a lifting mechanism for this and you know what this does is basically you know this is already hooked up to the roar and it's already being controlled by to do it as an additional axis essentially this just moves down the heated bed very slowly but with a lot of force if i stick my finger in here my finger is gone i tried standing on this before and the coupler between the motor and the lead screw just popped up but otherwise this can handle a lot of weight and specifically it needs a lot of weight moving down pulling down so what this is going to do is we're going to have our bed surfaces our spring steel beds that we'll need to snap to this magnetic surface and to do that they're going to be pushed in from this side and let me just get these rails in place they're going to get pushed in from this side roughly land in this spot and then the heated bit the aluminum bed is actually going to move up and it's going to snap it's going to center itself wait for it there we go and it's going to snap in place and once it is snapped in place this is going to be super solid this is not going to move anymore that's why we have this entire mechanism at the base here so we basically have a heated bed that is fixed in place just like the stock bed is in the voron so that is the platform that we can print onto and then once the print is done this thing will need to move down again and because this outer frame is resting on these rails or on these release rails here eventually it's going to pop up and then we can push the next bit in from this side basically once that is down um basically push it out of the printer it's going to drop out the other side and we are ready to raise the bet again and load in the next bit so that is the idea behind that now this is a mechanism that is not like particularly simple but it is what i eventually settled on i had a couple other ideas first the original idea that i had is you know why i built this war on 2.4 in the first place is to have a spinning drum within its envelope that basically had six sides with a you know with a heated bed on each side and then it would print on the top one it would rotate by 60 degrees is it yeah and it would have another bed on that other side but what i realized is these beds would need to be tiny if we actually have a full drum that fits in here and especially once we have prints on it you know they would need space at the bottom too once this thing has done a 180 degree uh revolution so i threw that out it would have been really cool it would have looked spectacular but it would have been absolutely impractical the next idea was to have some sort of a belt feed that would have a couple of these beds kind of in each tank tread unit that were kind of coupled together and then the printer would pull that in um with some sort of sprocket mechanism basically like a chain of beds and then have some sort of a mechanism like this now i could still implement that but eventually i realized well having a sprocket that engages with the tank tread and kind of pulls that through and having some sort of a release mechanism that would peel the magbeds off of the heater platforms it would just get super complicated and it would have too many moving pipes that would need to work with each other so right now the idea is with this kind of simplified practicalized version is this is going to fit inside the print envelope and then just outside the printer there's going to be a stack of fresh beds in their frames and basically there's going to be a belt which is what we're going to build next that will pull them in one by one and then as you just saw ejected out the other side so yeah now that this works i think the hardest part is over so basically we just need to get to building the feed mechanism or a magazine and then we can put this to the test so let's go do that [Music] [Music] [Music] me [Music] my [Music] okay so that should actually be the majority of this build done we've got the belt fed uh bed feeder mechanism that's working we've got the super sturdy bed lift to attach and peel off the magnetic beds now all that's left to do is to add some end stops to both of these movement axis because these are essentially just extra a and b axes and firmer and yeah then we should be able to give this a try that's not an end stop this is an end stop [Music] all right that's how this entire thing is going to fit in here now one thing i already realized is this thing obviously needed to fit through the side of the printer and it couldn't collide with any of the parts of the z-axis but turns out i took measurements off of the cad model of the voron 2.4 which was correct but that model was for the 250 millimeter size 2.4 and this is a 300 so i would have had a lot more room to work with which would have made some of these design decisions a lot easier but hey it is what it is so this thing also fits in a voron 250. now you can already see there is not a whole lot of z-axis space left um obviously this thing is still going to drop down a bit when we take out the original bit but yeah with this contraption the way it is built right now it does take up a bit of z-axis vertical space but this is more of a first try a proof of concept just to see if we can get that action of latching onto and separating the magpads off of our heater platform and whether we can get that working at all so next up i just need to set up these two additional motors as extra axes in firmer and i did plan ahead for this which is one of the main reasons why i went with wrapper firmware because it is so easy to set up additional axes you just copy paste the config for x and y and there you go there's your a and b axes and with the expander board we do have nine independent stepper drivers in this machine and we need nine so basically these are just plugged into the two spare drivers on the expansion board and they are ready to go so let's see if we can get this moving [Music] [Applause] [Music] so [Music] so that appears to be working just about perfectly um i did run a couple cycles on this on the swap cycle and every time it grabs the bed perfectly it does also remove it from the bed pretty well sometimes though it looks like the bed is too far forward to actually slide into place and then this side doesn't touch down fully so i'm just going to add a bit of a spacer up front here so this bed sits further back the frames also work really well for holding the mag beds and for allowing me to remove them again so this is pretty much ready for use so basically i need to hook up the bed i need to add some aluminum tape to these rails these frames are pusherman asa these rails however are ptg and i'm kind of afraid they're going to melt so aluminum tape just for some extra insulation there and then we're ready to hook up the bed put it in the printer and try a couple prints on this [Music] [Music] so that was just a bit of figuring out where the probing points would need to fall to make sure that the gantry tramming squaring still works because i am still probing uh the bet after every swap just to make sure that you know if one end if one corner ends up like a tenth higher or lower we are compensating for that um but yeah the swap cycle works the homing and probing works i've tested the heated bed that also works perfectly thermistor reads well so i guess all that's left to do is to slice a couple of parts and get this thing rolling make sure to get subscribed if you like what he's seeing all right so that is six prints done we got a small vase we got flawless sticks ring we've got a new fan shroud for my prusa mark iii we've got a table hook a jig for doing decking and of course a 3d benchy this one actually looks pretty good except for the stringing now these prints were all done without any intervention well not quite i did have to replenish the bad magazine because there is first of all a dead space of one bed between the magazine and the print platform so it always needs one extra bet in between there and also to eject that last print it needs one more bit to push that out but other than that as long as the hopper is fed this did not need any intervention or interaction for me and that is exactly what it was going for now obviously the question you're going to have is like how do you how do you do that how do you get these parts to come out continuously without you having to start a new one every time and the simple answer to how i did this is just copy paste a bunch of g-codes onto each other and just create one long file that has uh the start ng code of everyone and of course the print but the proper way to do that is of course through octoprint um there is a plugin called continuous print that is made exactly for this application for any printer that has a bed that can clear itself so be it a belt printer or like a 3d cue system that rams the prints off of your bed for anything like that that plug-in is perfect but even when you don't need to fill up a bunch of print surfaces with parts and just have them continuously pop out even then this can be pretty useful like for example the way i designed this um is i was drawing parts and while i was drawing stuff it was already printing that the last stuff i had designed i immediately sent off to the printer and then of course i had to clear the beds manually like this i could just export an stl throw in the slicer upload it to the printer and it would be you know once the last print is finished it would be automatically able to pull in a fresh bet and keep on printing now there are still a couple of shortcomings with this design first of all the most obvious one is the beds that i'm using are significantly smaller than the original bed this has a 300 by 300 millimeter print size this has like two 20 by 150ish now that is mostly a result of me having ordered these beds a while ago while i was still thinking i would make it a revolving drum that sits in the printer and not a magazine system but looking at the system again i think it would be possible to actually use the full size original beds like this can still fit to the side of the printer just fine even with a frame and with a different pusher design a different belt pusher design i think we could even eliminate the need to have like one extra spare bit in the queue just to push that bit onto the platform i think that would definitely be possible also of course you do lose a lot of build height with this one i am down to i think 110 millimeters of usable z-axis space i've already lost quite a bit by using the longer mosquito magnum plus and now this reduces that usable space further again prototype first iteration this can definitely be built slimmer i just built what i knew would work but now that i know how stuff works together i think this could be slimmed down significantly also with this sort of an automatic q system you kind of need a filament runout sensor right now i'm just using standard one kilogram spools but if you just keep replacing the beds you know once they get pushed out you pop the print off and you put it back into the magazine um like you will be running out of filament at some point and that would require manual interaction intervention and keeping an eye on it filament sensors are super easy to add and if you want to go way overboard you could even add the palettes you know the mosaic palette which is made for like multi-color printing but in the simplest use case you can just load a bunch of spools into it and it's going to automatically switch over from one spool to the next once those run out so that is something that could definitely solve that challenge in my case i am extremely happy with how this worked out especially the fact that it just works like there's there's nothing that really went wrong yeah there were a couple tweaks that i had to do like adjust the spacing but that was all planned for and overall i think i very much achieved my goal of wanting to have a printer that would print like it has a regular static bed while still having an automatic clearing like not using any special bed surfaces that might have subpar adhesion uh not requiring a belt system or stuff like that but just having something that looks to the printer and functions like a regular static bed and i'm very happy to report that the voron with the goal that i built it for the application that i built it for did perfectly for this so if you like this video as always give it a like get subscribed if you want to see more like it and if you want to support the channel so that i can do projects like these that may or may not turn out well at all um if you want to support me and make sure i can keep doing this sort of stuff uh check out patreon or the youtube memberships and thank you to everyone who is already supporting me there yeah as always thank you for watching keep on making get subscribed and i will see you in the next one bye you
Info
Channel: Thomas Sanladerer
Views: 179,798
Rating: undefined out of 5
Keywords: 3D printing, Tom's, 3D printer, RepRap, Voron, Voron 2.4, automatic print bed swap, belt printer, automated 3d printer, printfarm, 3d printer farm, automated 3d printing, continuous 3d printing, voron 2.4, voron 3d printer, voron 2.4 build, voron 2.4 review
Id: 8O9E9rcH6Us
Channel Id: undefined
Length: 21min 46sec (1306 seconds)
Published: Fri Jul 30 2021
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.