Fun fact: the part that I was printing for this video required a bit more than 1,3km of filament. If you know how to reliably join ABS filament to make a spool this big, please leave a comment. Have a great day!
Cut the capricorn tube lengthwise, remove thin strip of material to decrease internal diameter from 1,9 to 1,75mm, put it inside a piece of aluminium with two heaters mounted alongside the tube to squeeze it to desired OD and provide sufficient heating. If you think you need to see the joint then make small window in this aluminium block.
It's a pleasure working with these guys and I'm so proud to have them as a sponsor! Their ABS prints great and I really would like to use the C8 for some upcoming prototypes. Awesome that you've visited their factory, I'd love to do that too one day!
Been with you from the beginning of your dream, outstanding work my friend don’t let anything stop you. Life are period of joy and heartfelt moment’s with the dreams threw out them, keep go, there are no failures my friend there are only ways no to do that more again.
@@properprinting Daar sluit ik mij bij aan. Woonachtig in Duitsland, is het altijd een genot om af en toe weer de taal van mijn "Landsleute" te horen. Ga door zo!
There's going to be literally, a ton of tensile stress toward the bottom of that wheel in the weakest direction that fff fails moat often. There's no way that print will hold up. But, in the off chance that it does, it's going to be EPIC.
I use MEK (methyl Ethel Ketone) to post process certain ABS parts. I might have a 2 piece part that dove tails together. I then use and eye dropper in the seam, it wicks in and makes the joint forever. Were I building a tool for this, I'd want a tube only about .1mm bigger the the filament. I would cut both filaments at same time in jig so angle is perfectly matched. I'd slip together. Drip in some MEK through a side hole. Then allow it to cure for a few hours. Whatever method you choose, do a test of 20 each 1/2 meter pieces joined together and run a part to consume it and see if you can get all 20 to work properly. I would do a vase mode part so you can also examine the joints there they come out for any flaws. You might do this with alternating colors so you know where to look.
I found that method as well and I thought about doing it that way. The reason why I didn't is because I want to use different materials than ABS later on, hence I want to find a way that works for other materials too. Thanks for your suggestion!
@@properprinting For gluing ABS use Methyl ethyl ketone (MEK) it truly fuses it together and sets fast. I have used it and it's stronger than the prints themselves. Acetone is actually a far worse solvent for ABS because it takes forever to set.
I have used a similar method. Instead of squishing the filament though, I used a small file to get the diameter down to within tolerance after melting and pressing the filament ends together to make a good bond. My method only works about half the time. You might consider using a Prusa MMU2, so that when it senses that one reel has run out, it will switch to the next
Thanks Kenneth! I'm so happy that the printer is printing this reliably. This was my biggest concern and joining filaments will be something I'll figure out ;)
For joining filaments search on Aliexpress for "Filament Welder" or something like that, there are jigs available for little money for that exact purpose. The other suggestion would be... don't join the filament! Use a filament run out sensor that pauses the print whenever it is empty and swap the filament manually. Yes, still means having to babysit the printer, but at least you don't have to be sitting next to it the moment it runs out. And one more recommendation for the rim itself: How about a two part rim? These used to be more common in the past, and you still see them on some vehicles and things like wheelbarrows. And if you then use an inner tube you also don't have to rely on the rim being able to take all that air pressure. Awesome video by the way, looking forward to see more!
Mosaic Palette 2S Pro splices filaments together very reliably. Just do some spice tuning to make sure the splices are a good diameter and wont break and just use it in join spool mode.
Hey do you know that you can resume an interrupted printing !!! Just measure the Z height of you printed part (precisely) , delete the part of your G-code after the header up to the Z height (find a Go Z ...) you just have to be careful with the E axis, you should set it to the first next value (find the next E.... and set the origin here with G92 E....) your G-code should look like that: G92E ... G0 Z ... You also should remove any G28 in the header to prevent homing right in your part... the printer should not be turned off or it will loose its position This can take more than one try but can save you hours and a lot material !
That's a great suggestion, thanks! The only problem I'm facing (which is also the reason why I don't use a run-out sensor) is that I cannot open the heated enclosure without letting cold air in. I'll keep this in mind though because it might come in handy :)
Joining the filament is something I have always wanted to learn. I have used a PTFG tube and a heat gun and pushed like you did and it didn't work. As soon as I saw you try it I just heard "OH NO!". If you do find a way to join two spools of filament with out it getting stuck can you please make a video showing how. Keep up the wood work love the content and hopefully this rim will come true and hold up under load.
We used to join fiber optic cable with lasers. The fibers would be laid in a slot separated by a few thousand microns. The laser would heat up the fibers to transition point then press them together automatically thus fusing them.
This looks so cool as a project. I just found 2 little challenges... 1. If you manage to fit the tire, will it hold the air pressure? 2. The heat produced by the brakes will affect the rims, causing them to deform.
You can do it! I predict you will invent the best filament splicing tool for your needs. This will be the offshoot that makes your invention the one thing every 3d printer will have in the future. Good luck, you can do it.
It would suck, but if there is no sollution that works 100 out of 100 times then your best bet is to install a filament sensor and insert the new rol..
Hoi ik gebruik een slimme filament checken van big tree. Als het filament niet meer door de sensor gaat de print stopt upload het rest filament. Vraagt om nieuw filament. Heat up de nozzle and voert het nieuwe filament in tot voorbij de nozzle. Dan geef je het commando printen en gaat hij verder waar de printer gebleven was. Deze sensor haat ook voorkomen dat de printer voor niets in de lucht zou printen want hij registreert de beweging van het filament. Dus ook als het filament niet op is maar ergens verstopt zit tijden het printen stopt de printer en kan je opnieuw invoeren en verder printen. Groetjes en uitkijken met hard rijden met die velg. En een geweldige printer man.
Jon, you have to have so much patience to deal with everything you have. This is impressive... even in your failures. Can't wait to see the rims in use. (It *will* happen. I have faith in this)
I recommend coating the pressurized portion of the rim with 1-2 layers of low viscosity resin (i prefer XTC resin for this purpose). I designed and 3d printed a housing for a prototype miniature ROV (industrial remote control submarine), and without some kind of resin coating, it would leak air thru the housing walls when we put 30psi in it, even tho the wall thickness was 3.25mm @ 100% infill, and I made the printer over-extrude on purpose. USE RESIN!
Great recommendation, thanks! I was thinking of applying a coating, but didn't think of resin. Did you just cure it in the sun? I don't have a large chamber with UV lighting where I can put the rims in.
@@properprinting XTC is a not cured by UV light. It's more of low viscosity 2-part epoxy. I should have called it that instead of resin. You paint it onto the part with a brush (foam or bristles). To clarify my previous comment, 30psi is roughly what a car tire is supposed to be inflated to (mine are supposed to be 33 for the front & 32 for the rear)
@@properprinting Also if you need a bulk roll of ABS: pushplastic.com They have 3kg, 5kg, 10kg, and 25kg rolls. My company uses a 25kg roll. Shipping might be expensive for you tho since it is an american company. If you do get a large roll, avoid fast and/or frequent retractions. Larger rolls have more inertia and will cause under extrusion immediately after a fast/normal speed retraction. Big rolls also collect a lot more dust.
Just wondering, if you use a dual headed extruder then you could use a cheaper filament for the supports which would mean less splicing of the main more expensive filament roles? Or some kind of g code to swap heads on a filament line break ?? Keep at it and you’ll get there.
Try printing something small out of strings of ABS connected / joined in different ways in order to see which way is the most reliable :) just a thought.
For example, take a load of short pieces, connect them in various ways, note down which order they are in and print something small, to see which breaks first and out of which of the join types is the most reliable.
I would look to see if your board supports filament break detectors. They are cheap and easy, and then you don't need to splice. Install, then an easy firmware edit and update and you are in business. When your filament runs out, just reload, and it will pick up where it left off. Forget about splicing. You run your spools until empty and reload. Saves money. Good luck!
Thanks for your suggestion! This has been asked and suggested by others too, but the problem I'm facing is that I don't want to open the heated enclosure letting cold air in. I'm afraid that this will cause the print to warp and fail. Especially when I'm going to use different materials like Nylon of Polycarbonate. Maybe I'm overlooking something. Fortunately 3D4Makers will send large spools soon so this shouldn't be an issue anymore.
@@properprintingWhy you don't put the filament sensor on the outside and place a PTFE tube to the extruder? You will retract the rest of the filament anyway, so you should be able to grab it outside the chamber and feed the new one in down to the extruder...
@@r0311 Because the head is moving in all three dimensions which makes it difficult for the tube to go everywhere. I also don't know if pausing a print can eventually cause a weak point. Printing in one go would be my preference which shouldn't be a problem if I receive big spools or if I can get the hang of joining filaments. I prefer the first of course xD
For testing filament joints, I suggest making 1m long piece of filament that is full of as many joints you can fit. Then you can test them quickly than having just 1 between each spool. Otherwise maybe it would just be safest to install some kind of filament runout sensor that can pause the print until you swap it. Excited to see how you solve it. I have tried to splice filament a few times but often it caused jams that made me have to take apart my hotend so would be great to discover a reliable way.
Awesome printer and project. 11:30 - you might want to increase your extrusion multiplier a little? Some of the beads around the outside of the hub are barely touching. You want them fully joined to give you maximum radial strength! Good luck!
If you can redesign the center of the wheel to use a preexisting inner and outer barrel like HRE wheels did with their 3d printed wheels. That would eliminate the need to seal the wheel. It would also allow multi piece wheels to be made. That way, if a particular piece failed. That part(inner barrel/outer barrel/center) can be reengineered to fix the problem without having to print a completely new wheel. It would also break down printing into sections that can be bolnded together and make print failures not as catastrophic.
First of all, Amazing work on the printer!! Would love to have one for a similar reason :) joining the ABS: 1. Wouldn't a Palette be of help? You would probably get it sponsored and you can use 4 spools and it automatically joins them or the multimaterial selector from prusa as the open source variant.. my own version would be to make a second jig with razor blades,since they can be bent to a certain degree, 3 of them in sequence should shave down the diameter of the joint to reliable 1.75 or you run the joint through a set or two of extruder gears that are set quite tight.. the PTFE tube has an inside diameter of 1.9-2mm and is stiff on the inside and can absorb some squish on the outside,so it isn't ideal the way you pressed it. Hope it helps get a lightbulb going..cheers from germany
I think you should just use the larger rolls of filament and change them out as they get low. It would be easy to calculate about how long a roll of larger fillament would last and then you would know when to stop by and make the change. Do not fool around with joining the ABS filament.
Less joins and a filament runout would be best IMO, it would allow you to start it back where it left off and a lot less annoying that trying to join so many spools. Eventually filament weight will become an issue for the extruder too, hence why most print farms max out at 2-4kg spools. Think runout is definitely the best best IMO.
Great camera work, great editing, good music, amazing technical skills and content. But most of all very entertaining, the way you show failures and fix them is also very impressive. Niets dan lof! Keep up the great work!
what if you clamp two pieces of aluminum together, drill at the seem the correct wire diameter then chamfer it on both ends. Then while your joint is still hot, pull it through the drilled hole? More or less just making a removable extrusion? Cool stuff!!
Have you calculated the force that hit the rim when you put the tire on, pressurize it and maybe steer around? Don't get me wrong, the stuff you do is insane and you build a hell of a monster of a 3d printer. But I think this will not work. Nevertheless: Please keep on going. Your videos are awesome.
Why cant the filament be swapped? Could the hotbed not just stay hot, the head move away and the filament replaced? It kinda sounds like it's solvable with firmware and configuration.
Try to mount a filament Sensor, sure you will need to swop the Filament but joining spools will always have issues. Or send your spool to your sponsor and let it fill up straight from the produktion line 😉
Great project, it will be quite a ride but Ican't wait to see you rollin' on printed rims :D You can check out filament connectors on amazon, the issue came from using 1,9mm PTFE tubes instead of 1,75mm diameter ones
I'm a little concerned at the gaps between your perimeters. That will likely be a structural weakness allowing more flex than you'll want in the finished product. You might need more overlap in your slicer. Of course, I could be totally incorrect in my concerns. Can't wait to see the continuation of this project. 💪👍
Wheel looks cool as heck, but the outhermost ring parts will probably delaminate and pop open when you presurize it, unless you are going to fiberglass/CF the interior part or something.
What about making something that can clamp over the filament, the two ends straighten the filament, and a blade, or several blades in the middle shaves it down to the right diameter? Slide it up and down the filament to trim off any excess?
Congrats Jon! The beast finally ate some nice amount of filament 😄😄. Why not use a filament runout sensor, pause the print, replace the spool and continue printing?
mostly causes layershifts same as shut off and and resume never as strong as one whole print... because the layer before already cooled down if it was not for something as structual as rims it would be fine i guess
@@LarryClay exactly, thank you. There are many workarounds. As soon as I saw Jon trying to bond filaments I knew it was going to be point of failure. I know @ProperPrinting has a card up his sleeve and that’s why he did that ♠️
The main reason for me to not use a filament runout sensor is that I have a heated chamber which is at a stable temperature of 60°C. The print will fail if I open the door letting cool air in. I need that printer to print in one go. Trying to push filament through a reversed bowden tube hoping that the direct drive extruder will grab it is too tricky I think xD
Is there a way you could add a resume print or change filament function to the printer? Then you could plan changes when it suited you, you could stop when a layer had completed so it would be easier to continue on top of the flat plane left when it stopped.
You really need a filament movement sensor. A basic one is just a roller with a photo interrupter. Filament feeding causes the signal to pulse every Xmm, if controller (or even OctoPrint) doesn't see a pulse after Xmm should have been extruded, it assumes the filament has jammed and pauses the print.
Using a dual extruder, you would be able to manually change the filament in an alternating pattern, which won't slow down the print, but you would have to change the filament during a certain time window. Not the ideal solution, but could work?
Could you make and use a split die with a 1.75mm bore through the centre? Made from aluminium or other metal, it clamps over the filament and is heated from the outside and then removed after cooling. The main difficulty would be seeing when the filament ends are touching.
I can't remember when/where but I remember a video about splicing filament and the guy had to run the filament through a series of bends (up/down/left/right and back) to recondition the plastic to flex normally after being heated and cooled. I'll see if I can find it in my history.
I'm curious because I indeed had this experience. The second attempt was with a metal tube and the filament became less brittle after bending when it was cooled down.
@@properprinting I can't find it, might have been on work computer at previous job. He used a hair curling iron (with two flat copper plates pressing together) with ptfe tubing. He then created a jig that had pulleys that he threaded the joined filament on and then ran the spools back and forth -forcing the the filament joint through a series of bends a few times (starting slow and building up speed). I think he had a heat gun somewhere but I can't remember if it was mounted or hand held. He used very sharp angles on the cuts to increase overlap and twisted the ends together while heated. Not like a cigarette being put out, but enough to force a good interaction between the surfaces, not just hard contact. He also stretched and repressed the joint multiple times to help the plastic mix together while molten. Apologies for not finding the link but that's the bulk of what I remember from it.
3d prints are notorious for being not being air tight I think this may be your next big issue I hope you can solve the joining issue. Maybe a two part metal die to size the filament after that can be run across the joint after its colled a little might help to get a consistent size.
To save you a headache and a lot of material / money. The spokes of the rim need to be a lot more solid. The generative design looks good but will only be strong enough if printed out of metal. Other then that spot on, looking forward to the next video!
Good question. I don't want the printer to pause and have to swap the filament manually. I have to find a way to do that without opening the door in that case because it's a heated chamber. Especially when printing tougher materials I'd like this machine to continue until it's done.
I have to agree with the others, you are trying to solve a self made problem. 1) I would redesign the rim to be assembled from smaller parts, you can just reprint those in case of print failures or later failures while using the rim. 2) Use a second material for supports to save the expensive main material. 3) use a splicer like Palette 2 to splice your filament together (and don't do this yourself) or even better use a MMU2 to swap between rolls (and the support material), Marlin 2.0 supports it now. 4) Getting bigger rolls is far the easiest solution for your problem. Go ahead, I am waiting for the final result on your car!
I agree with the self made problem I try to solve, but what I'm hoping to achieve is to be able to print large things in one go. It's not just about printing a rim, this is an interesting thing to use to explore these kind of challenges and learn from them which will help me, and possibly others, with other designs. I'll take a look into the MMU2 of the Palette, thanks!
@@properprinting good idea, palette can do the splicing extremely well and you don't have to prepare it before you print. What about bigger spools? I know that Evan Miranda is using 6kg spools for his (also large) projects.
I'm guessing you're going to need a pretty clean weld, I think sanding is going to leave a rough finish that can get caught up even at an acceptable diameter, like it did the second time. You could maybe try something like a bandsaw blade spot-welder configuration. Cut a hot-end tube down the centre, with a diamond-wire handsaw or using a slit saw on a mill if you have access, then use the two halves with current applied to make a heated clamp which you can press the two filament pieces together with to give a smooth finish. Hopefully that makes some sense, can send a drawing if it sounds like any help at all haha!
Interesting approach and this didn't occur to me, thanks for bringing this up! The only problem I can think of is that the whole tube gets hot so I need a cold zone. Otherwise the filament on the outside will melt too causing a collar. I got this problem when using a steel tube to melt the filaments together.
Proper Printing Avoiding the collar will be difficult like you said, you’d need to keep the heating part as small as possible and spring load the mechanism so it only heats as you press the two parts together. Maybe instead of an electrical solution a small precision flame would give more localised heat on the tube you use rather than heating the whole tube with a current. It may also be you first method with sanding will work if you expose the sanded filament to a naked flame briefly to smooth out the area. Either way I would make a meter of weld tests and just Tun that through the hot-end to really test the method you use. Good luck with it, looking forward to seeing the rims!
Just an idea. Take a printer nozzle, carefully drill it out with a 1.75 mm drill, screw that in a hot-end and push through the filament. No idea is this will work, but it seems logical.
You could have used one of those clamp hair irons to fuse the filament together. Much neater that cutting up a hotend. I know temps works with pla at least.
Have you considered drilling a 1.75mm-ish hole into a metal cylinder/block and used that instead of a PTFE tube? It's not flexible at all and using a proper drill bit allows you to get even closer to 1.75mm accuracy. For heating, possibly putting it over a flame would be good enough too. I think your main issue there was using a PTFE tube and trying to squish it smaller, cause really you're just basically turning it into some kind of polygon and it's not perfectly round and it can be wider in some areas than it should. You'd best try to get a solid thing to hold it and have the filament go through extremely snug to make it accurate enough. Though I really am not sure. :P Best of luck!
It's going to be not 1.75mm-ish, but exactly 1.75mm ;) twitter.com/properprinting/status/1332036689368330254 I've got a fun and experimental idea to get this done. Hopefully it will work!
Yay! Long time no smile,Jon! I must say that I foresaw the filament breaking, but I’m sure you will find the Proper way! Oh, I miss the usual intro,mate. Looking forward to the next video though. 🤘
Fun fact: the part that I was printing for this video required a bit more than 1,3km of filament. If you know how to reliably join ABS filament to make a spool this big, please leave a comment. Have a great day!
You could try using the palette, it has a "Multi-spool mode" that can automatically splice multiple rolls of filaments together
AliExpress solds a filament joiner tool , it costs almost 7,5 euros and its full metal. Never bought it because of the price but deserved take a look
I would just buy bigger spools. Check your local manufacturers and see if they are willing to make 2-5kg filament spools.
Cut the capricorn tube lengthwise, remove thin strip of material to decrease internal diameter from 1,9 to 1,75mm, put it inside a piece of aluminium with two heaters mounted alongside the tube to squeeze it to desired OD and provide sufficient heating.
If you think you need to see the joint then make small window in this aluminium block.
How about with acetone?
The thing I like about your video's is you don't hide the failures, and eventually you succeed with perseverance.
Keep up the good work
Is it just me or is the video quality just getting better for every time PP uploads a new video... Such a cool project !
it is :D
Love the 3D4makers people, visited their factory two years ago, was fun to see their machines. Facilan C8 is really amazing to print with
That temperature resistance though.... do they have anything similar but heat resistant?
It's a pleasure working with these guys and I'm so proud to have them as a sponsor! Their ABS prints great and I really would like to use the C8 for some upcoming prototypes. Awesome that you've visited their factory, I'd love to do that too one day!
Been with you from the beginning of your dream, outstanding work my friend don’t let anything stop you. Life are period of joy and heartfelt moment’s with the dreams threw out them, keep go, there are no failures my friend there are only ways no to do that more again.
Thanks for these words!
@@properprinting Daar sluit ik mij bij aan. Woonachtig in Duitsland, is het altijd een genot om af en toe weer de taal van mijn "Landsleute" te horen. Ga door zo!
There's going to be literally, a ton of tensile stress toward the bottom of that wheel in the weakest direction that fff fails moat often. There's no way that print will hold up. But, in the off chance that it does, it's going to be EPIC.
I use MEK (methyl Ethel Ketone) to post process certain ABS parts. I might have a 2 piece part that dove tails together. I then use and eye dropper in the seam, it wicks in and makes the joint forever. Were I building a tool for this, I'd want a tube only about .1mm bigger the the filament. I would cut both filaments at same time in jig so angle is perfectly matched. I'd slip together. Drip in some MEK through a side hole. Then allow it to cure for a few hours. Whatever method you choose, do a test of 20 each 1/2 meter pieces joined together and run a part to consume it and see if you can get all 20 to work properly. I would do a vase mode part so you can also examine the joints there they come out for any flaws. You might do this with alternating colors so you know where to look.
When you first mentioned the joining of ABS filament, my thoughts were 'with acetone' immediatly
Good idea, clean 90 degree cut + acetone joining. Never tried it myself though and it would work with the PEEK he wants to use for the final version
I found that method as well and I thought about doing it that way. The reason why I didn't is because I want to use different materials than ABS later on, hence I want to find a way that works for other materials too. Thanks for your suggestion!
@@properprinting Can you superglue the ends together?
@@ps3ud0r00t3r don't you think that the super glue would eventually clog the nozzle?
@@properprinting For gluing ABS use Methyl ethyl ketone (MEK) it truly fuses it together and sets fast. I have used it and it's stronger than the prints themselves. Acetone is actually a far worse solvent for ABS because it takes forever to set.
I have used a similar method. Instead of squishing the filament though, I used a small file to get the diameter down to within tolerance after melting and pressing the filament ends together to make a good bond.
My method only works about half the time. You might consider using a Prusa MMU2, so that when it senses that one reel has run out, it will switch to the next
My second try worked better and two joints were successfully printed, now I want to search for an easy and reliable method
sugested the same mmu2 methode is the way to go i think
For a less DIY approach, is the Palette 2 Pro fast enough to keep up? Would give you four spools at least
Super mooi project Jon!! Ik volg het al sinds het begin en ben blij dat je alle tegenslagen in je video's houdt!
Bedankt Geert! Ja, het is een reis die we samen maken inclusief de tegenslagen die erbij horen :)
Keep it up Jon'!! That machine is awesome and cant wait to see the rim when it is all done!
Thanks Kenneth! I'm so happy that the printer is printing this reliably. This was my biggest concern and joining filaments will be something I'll figure out ;)
@@properprinting I can't wait till you do figure it out so I can do the same and use up my little remnants.
What about using a mosaic palette? You can load up multiple spools and it will automatically switch spools when one runs out
For joining filaments search on Aliexpress for "Filament Welder" or something like that, there are jigs available for little money for that exact purpose. The other suggestion would be... don't join the filament! Use a filament run out sensor that pauses the print whenever it is empty and swap the filament manually. Yes, still means having to babysit the printer, but at least you don't have to be sitting next to it the moment it runs out. And one more recommendation for the rim itself: How about a two part rim? These used to be more common in the past, and you still see them on some vehicles and things like wheelbarrows. And if you then use an inner tube you also don't have to rely on the rim being able to take all that air pressure. Awesome video by the way, looking forward to see more!
End goal is utterly ridiculous, but the journey is amazingly fascinating, thank you for your videos
Send them your spool and ask them to fill it. Keep up the great work. Cheers, Mike
Mosaic Palette 2S Pro splices filaments together very reliably. Just do some spice tuning to make sure the splices are a good diameter and wont break and just use it in join spool mode.
Hey do you know that you can resume an interrupted printing !!!
Just measure the Z height of you printed part (precisely) , delete the part of your G-code after the header up to the Z height (find a Go Z ...)
you just have to be careful with the E axis, you should set it to the first next value (find the next E.... and set the origin here with G92 E....)
your G-code should look like that:
G92E ...
G0 Z ...
You also should remove any G28 in the header to prevent homing right in your part...
the printer should not be turned off or it will loose its position
This can take more than one try but can save you hours and a lot material !
That's a great suggestion, thanks! The only problem I'm facing (which is also the reason why I don't use a run-out sensor) is that I cannot open the heated enclosure without letting cold air in. I'll keep this in mind though because it might come in handy :)
You'll never how close you really are to the finish line until you've crossed it.. you are on the edge of greatness, keep pushing!
Joining the filament is something I have always wanted to learn. I have used a PTFG tube and a heat gun and pushed like you did and it didn't work. As soon as I saw you try it I just heard "OH NO!". If you do find a way to join two spools of filament with out it getting stuck can you please make a video showing how. Keep up the wood work love the content and hopefully this rim will come true and hold up under load.
This video shows the inspiration behind your belt driven extruder!
We used to join fiber optic cable with lasers. The fibers would be laid in a slot separated by a few thousand microns. The laser would heat up the fibers to transition point then press them together automatically thus fusing them.
This looks so cool as a project. I just found 2 little challenges...
1. If you manage to fit the tire, will it hold the air pressure?
2. The heat produced by the brakes will affect the rims, causing them to deform.
1. We'll find out ;) 2. I hope I will ever get to that point xD
Your channel is everything ive wanted in 3d printing content. Thank you for doing this and keep up the great work !
You can do it! I predict you will invent the best filament splicing tool for your needs. This will be the offshoot that makes your invention the one thing every 3d printer will have in the future. Good luck, you can do it.
For fusing filaments maybe have a look at something such as the pallete 2 and try and work something out with its splice core?
How does this guy only have 19k subs.......this should be 10x!! Keep it up ;)
Super cool project
Awesome work Sir
Amazing dedication and outlook, excited to see this finish!
It would suck, but if there is no sollution that works 100 out of 100 times then your best bet is to install a filament sensor and insert the new rol..
Hoi ik gebruik een slimme filament checken van big tree. Als het filament niet meer door de sensor gaat de print stopt upload het rest filament. Vraagt om nieuw filament. Heat up de nozzle and voert het nieuwe filament in tot voorbij de nozzle. Dan geef je het commando printen en gaat hij verder waar de printer gebleven was. Deze sensor haat ook voorkomen dat de printer voor niets in de lucht zou printen want hij registreert de beweging van het filament. Dus ook als het filament niet op is maar ergens verstopt zit tijden het printen stopt de printer en kan je opnieuw invoeren en verder printen. Groetjes en uitkijken met hard rijden met die velg. En een geweldige printer man.
Jon, you have to have so much patience to deal with everything you have. This is impressive... even in your failures.
Can't wait to see the rims in use. (It *will* happen. I have faith in this)
I recommend coating the pressurized portion of the rim with 1-2 layers of low viscosity resin (i prefer XTC resin for this purpose).
I designed and 3d printed a housing for a prototype miniature ROV (industrial remote control submarine), and without some kind of resin coating, it would leak air thru the housing walls when we put 30psi in it, even tho the wall thickness was 3.25mm @ 100% infill, and I made the printer over-extrude on purpose.
USE RESIN!
Great recommendation, thanks! I was thinking of applying a coating, but didn't think of resin. Did you just cure it in the sun? I don't have a large chamber with UV lighting where I can put the rims in.
@@properprinting XTC is a not cured by UV light. It's more of low viscosity 2-part epoxy. I should have called it that instead of resin.
You paint it onto the part with a brush (foam or bristles).
To clarify my previous comment, 30psi is roughly what a car tire is supposed to be inflated to (mine are supposed to be 33 for the front & 32 for the rear)
@@pass-123 clear, thank you! This seems like the perfect solution!
@@properprinting Also if you need a bulk roll of ABS: pushplastic.com
They have 3kg, 5kg, 10kg, and 25kg rolls.
My company uses a 25kg roll.
Shipping might be expensive for you tho since it is an american company.
If you do get a large roll, avoid fast and/or frequent retractions. Larger rolls have more inertia and will cause under extrusion immediately after a fast/normal speed retraction. Big rolls also collect a lot more dust.
@@pass-123 thanks for the info! 3D4Makers is working on delivering bigger spools which will solve the issue.
Just wondering, if you use a dual headed extruder then you could use a cheaper filament for the supports which would mean less splicing of the main more expensive filament roles? Or some kind of g code to swap heads on a filament line break ??
Keep at it and you’ll get there.
Try printing something small out of strings of ABS connected / joined in different ways in order to see which way is the most reliable :) just a thought.
For example, take a load of short pieces, connect them in various ways, note down which order they are in and print something small, to see which breaks first and out of which of the join types is the most reliable.
Awesome video, you are becoming my best 3d printing channel
how about a filament sensor like the duet one? it measures if the filament moves or not
Ondanks de tegenslagen, blijft ontzettend vet❤️
Bedankt! Zonder tegenslagen zou het te makkelijk zijn xD
Keep up with the fantastic "dreaming big" projects. Your videos are very entertaining and inspiring. Love, from Australia.
Jon, you are both creative and innovative. I really enjoy your channel. Thank you!
Another excellent video. Thank you
I wonder if you could get just the splicer from a Palette 2 for the joining.
I would look to see if your board supports filament break detectors. They are cheap and easy, and then you don't need to splice. Install, then an easy firmware edit and update and you are in business. When your filament runs out, just reload, and it will pick up where it left off. Forget about splicing. You run your spools until empty and reload. Saves money. Good luck!
Thanks for your suggestion! This has been asked and suggested by others too, but the problem I'm facing is that I don't want to open the heated enclosure letting cold air in. I'm afraid that this will cause the print to warp and fail. Especially when I'm going to use different materials like Nylon of Polycarbonate. Maybe I'm overlooking something. Fortunately 3D4Makers will send large spools soon so this shouldn't be an issue anymore.
@@properprintingWhy you don't put the filament sensor on the outside and place a PTFE tube to the extruder? You will retract the rest of the filament anyway, so you should be able to grab it outside the chamber and feed the new one in down to the extruder...
@@r0311 Because the head is moving in all three dimensions which makes it difficult for the tube to go everywhere. I also don't know if pausing a print can eventually cause a weak point. Printing in one go would be my preference which shouldn't be a problem if I receive big spools or if I can get the hang of joining filaments. I prefer the first of course xD
For testing filament joints, I suggest making 1m long piece of filament that is full of as many joints you can fit. Then you can test them quickly than having just 1 between each spool. Otherwise maybe it would just be safest to install some kind of filament runout sensor that can pause the print until you swap it.
Excited to see how you solve it. I have tried to splice filament a few times but often it caused jams that made me have to take apart my hotend so would be great to discover a reliable way.
Awesome printer and project. 11:30 - you might want to increase your extrusion multiplier a little? Some of the beads around the outside of the hub are barely touching. You want them fully joined to give you maximum radial strength! Good luck!
If you can redesign the center of the wheel to use a preexisting inner and outer barrel like HRE wheels did with their 3d printed wheels. That would eliminate the need to seal the wheel. It would also allow multi piece wheels to be made. That way, if a particular piece failed. That part(inner barrel/outer barrel/center) can be reengineered to fix the problem without having to print a completely new wheel. It would also break down printing into sections that can be bolnded together and make print failures not as catastrophic.
First of all, Amazing work on the printer!! Would love to have one for a similar reason :) joining the ABS: 1. Wouldn't a Palette be of help? You would probably get it sponsored and you can use 4 spools and it automatically joins them or the multimaterial selector from prusa as the open source variant.. my own version would be to make a second jig with razor blades,since they can be bent to a certain degree, 3 of them in sequence should shave down the diameter of the joint to reliable 1.75 or you run the joint through a set or two of extruder gears that are set quite tight.. the PTFE tube has an inside diameter of 1.9-2mm and is stiff on the inside and can absorb some squish on the outside,so it isn't ideal the way you pressed it. Hope it helps get a lightbulb going..cheers from germany
Oh my God!!...... DUDE!!!..... at 9:05 my heart almost stopped beating when I saw that
I think you should just use the larger rolls of filament and change them out as they get low. It would be easy to calculate about how long a roll of larger fillament would last and then you would know when to stop by and make the change. Do not fool around with joining the ABS filament.
AWESOME video! You've earned my subscription.
So much work!! Plus editing!
Good luck, even if the wheels don't work out your machine is epic and so are your videos. Hope you keep the sick ideas coming after this.
Thanks Tony! I am still at the beginning of the list of ideas I started with ;)
Great work. Very impressed with you channel and that printer is amazing
Nice as always! I would put my money on a (or several) filament runout sensor(s).
Or ask my sponsor to provide a 20kg spool!
Less joins and a filament runout would be best IMO, it would allow you to start it back where it left off and a lot less annoying that trying to join so many spools. Eventually filament weight will become an issue for the extruder too, hence why most print farms max out at 2-4kg spools. Think runout is definitely the best best IMO.
I have never been so excited to see the result. This is going to look awesome.
if you dont want to spool all that filament you could try a pallette 2. I have one and it works great for long multi spool prints.
What a great project!
Great camera work, great editing, good music, amazing technical skills and content. But most of all very entertaining, the way you show failures and fix them is also very impressive.
Niets dan lof! Keep up the great work!
what if you clamp two pieces of aluminum together, drill at the seem the correct wire diameter then chamfer it on both ends. Then while your joint is still hot, pull it through the drilled hole? More or less just making a removable extrusion? Cool stuff!!
I absolutely love your channel. Keep it up and good luck.......
Have you calculated the force that hit the rim when you put the tire on, pressurize it and maybe steer around? Don't get me wrong, the stuff you do is insane and you build a hell of a monster of a 3d printer. But I think this will not work. Nevertheless: Please keep on going. Your videos are awesome.
Looks promising 😀 can't wait to see the next part.
Thanks, it's quite the journey and I cannot wait to finally have a finished rim!
Why cant the filament be swapped?
Could the hotbed not just stay hot, the head move away and the filament replaced? It kinda sounds like it's solvable with firmware and configuration.
@@owowowdhxbxgakwlcybwxsimcwx I think he might be using a duet
Thanks for these inspiring videos, keep it up.
Try to mount a filament Sensor, sure you will need to swop the Filament but joining spools will always have issues. Or send your spool to your sponsor and let it fill up straight from the produktion line 😉
Why not put a filament run out sensor on it?
You could use the Mosaic Palette 2 to join the Abs ends, it is somewhat like the Prusa Mmu2 but is not tied to one type or brand of printer
Great project, it will be quite a ride but Ican't wait to see you rollin' on printed rims :D You can check out filament connectors on amazon, the issue came from using 1,9mm PTFE tubes instead of 1,75mm diameter ones
I'm a little concerned at the gaps between your perimeters. That will likely be a structural weakness allowing more flex than you'll want in the finished product. You might need more overlap in your slicer. Of course, I could be totally incorrect in my concerns.
Can't wait to see the continuation of this project. 💪👍
Wheel looks cool as heck, but the outhermost ring parts will probably delaminate and pop open when you presurize it, unless you are going to fiberglass/CF the interior part or something.
How about adding a filament sensor?
cant wait to see the end result :D
What about making something that can clamp over the filament, the two ends straighten the filament, and a blade, or several blades in the middle shaves it down to the right diameter? Slide it up and down the filament to trim off any excess?
You are CRAZY.... BUT I LOVE IT!! Good Luck Brother!!!
Congrats Jon! The beast finally ate some nice amount of filament 😄😄. Why not use a filament runout sensor, pause the print, replace the spool and continue printing?
mostly causes layershifts same as shut off and and resume never as strong as one whole print... because the layer before already cooled down if it was not for something as structual as rims it would be fine i guess
@@PascalVos When I am doing this, I wait till the printer is printing supports then there is not a structural issue.
@@LarryClay exactly, thank you. There are many workarounds. As soon as I saw Jon trying to bond filaments I knew it was going to be point of failure. I know @ProperPrinting has a card up his sleeve and that’s why he did that ♠️
The main reason for me to not use a filament runout sensor is that I have a heated chamber which is at a stable temperature of 60°C. The print will fail if I open the door letting cool air in. I need that printer to print in one go. Trying to push filament through a reversed bowden tube hoping that the direct drive extruder will grab it is too tricky I think xD
Is there a way you could add a resume print or change filament function to the printer? Then you could plan changes when it suited you, you could stop when a layer had completed so it would be easier to continue on top of the flat plane left when it stopped.
well... that "trailer" at the begining was epic ahahah
I'd like to say you have been a great inspiration for me to pursue my wishes to modify my printers as well!. Happy to see what you have achieved!
You really need a filament movement sensor.
A basic one is just a roller with a photo interrupter. Filament feeding causes the signal to pulse every Xmm, if controller (or even OctoPrint) doesn't see a pulse after Xmm should have been extruded, it assumes the filament has jammed and pauses the print.
Using a dual extruder, you would be able to manually change the filament in an alternating pattern, which won't slow down the print, but you would have to change the filament during a certain time window. Not the ideal solution, but could work?
Have you tried the pallet or pallet 2, though I think they only work on pla but it might be something to look into.
Could you make and use a split die with a 1.75mm bore through the centre? Made from aluminium or other metal, it clamps over the filament and is heated from the outside and then removed after cooling. The main difficulty would be seeing when the filament ends are touching.
I can't remember when/where but I remember a video about splicing filament and the guy had to run the filament through a series of bends (up/down/left/right and back) to recondition the plastic to flex normally after being heated and cooled. I'll see if I can find it in my history.
I'm curious because I indeed had this experience. The second attempt was with a metal tube and the filament became less brittle after bending when it was cooled down.
@@properprinting I can't find it, might have been on work computer at previous job. He used a hair curling iron (with two flat copper plates pressing together) with ptfe tubing. He then created a jig that had pulleys that he threaded the joined filament on and then ran the spools back and forth -forcing the the filament joint through a series of bends a few times (starting slow and building up speed). I think he had a heat gun somewhere but I can't remember if it was mounted or hand held. He used very sharp angles on the cuts to increase overlap and twisted the ends together while heated. Not like a cigarette being put out, but enough to force a good interaction between the surfaces, not just hard contact. He also stretched and repressed the joint multiple times to help the plastic mix together while molten. Apologies for not finding the link but that's the bulk of what I remember from it.
Rad idea!
I recommend if you want to respool the filament just use a palette 2 by mosaic
3d prints are notorious for being not being air tight I think this may be your next big issue I hope you can solve the joining issue. Maybe a two part metal die to size the filament after that can be run across the joint after its colled a little might help to get a consistent size.
To save you a headache and a lot of material / money. The spokes of the rim need to be a lot more solid. The generative design looks good but will only be strong enough if printed out of metal. Other then that spot on, looking forward to the next video!
Why you dont just add filament sensor?
Good question. I don't want the printer to pause and have to swap the filament manually. I have to find a way to do that without opening the door in that case because it's a heated chamber. Especially when printing tougher materials I'd like this machine to continue until it's done.
I have to agree with the others, you are trying to solve a self made problem. 1) I would redesign the rim to be assembled from smaller parts, you can just reprint those in case of print failures or later failures while using the rim. 2) Use a second material for supports to save the expensive main material. 3) use a splicer like Palette 2 to splice your filament together (and don't do this yourself) or even better use a MMU2 to swap between rolls (and the support material), Marlin 2.0 supports it now. 4) Getting bigger rolls is far the easiest solution for your problem.
Go ahead, I am waiting for the final result on your car!
I agree with the self made problem I try to solve, but what I'm hoping to achieve is to be able to print large things in one go. It's not just about printing a rim, this is an interesting thing to use to explore these kind of challenges and learn from them which will help me, and possibly others, with other designs. I'll take a look into the MMU2 of the Palette, thanks!
@@properprinting good idea, palette can do the splicing extremely well and you don't have to prepare it before you print.
What about bigger spools? I know that Evan Miranda is using 6kg spools for his (also large) projects.
I'm guessing you're going to need a pretty clean weld, I think sanding is going to leave a rough finish that can get caught up even at an acceptable diameter, like it did the second time. You could maybe try something like a bandsaw blade spot-welder configuration. Cut a hot-end tube down the centre, with a diamond-wire handsaw or using a slit saw on a mill if you have access, then use the two halves with current applied to make a heated clamp which you can press the two filament pieces together with to give a smooth finish. Hopefully that makes some sense, can send a drawing if it sounds like any help at all haha!
Interesting approach and this didn't occur to me, thanks for bringing this up! The only problem I can think of is that the whole tube gets hot so I need a cold zone. Otherwise the filament on the outside will melt too causing a collar. I got this problem when using a steel tube to melt the filaments together.
Proper Printing Avoiding the collar will be difficult like you said, you’d need to keep the heating part as small as possible and spring load the mechanism so it only heats as you press the two parts together. Maybe instead of an electrical solution a small precision flame would give more localised heat on the tube you use rather than heating the whole tube with a current. It may also be you first method with sanding will work if you expose the sanded filament to a naked flame briefly to smooth out the area. Either way I would make a meter of weld tests and just Tun that through the hot-end to really test the method you use. Good luck with it, looking forward to seeing the rims!
Great video. Such a cool project. Mahalo for sharing! :) 🐒
You can do it!!! Best of luck. No idea how to join filaments but I'm sure others will help you get there.
Very cool Project, but even if the rim is strong enough for normal driving loads, will it withstand the heat from the brakes?
Just an idea. Take a printer nozzle, carefully drill it out with a 1.75 mm drill, screw that in a hot-end and push through the filament. No idea is this will work, but it seems logical.
You could have used one of those clamp hair irons to fuse the filament together. Much neater that cutting up a hotend. I know temps works with pla at least.
Have you considered drilling a 1.75mm-ish hole into a metal cylinder/block and used that instead of a PTFE tube? It's not flexible at all and using a proper drill bit allows you to get even closer to 1.75mm accuracy. For heating, possibly putting it over a flame would be good enough too. I think your main issue there was using a PTFE tube and trying to squish it smaller, cause really you're just basically turning it into some kind of polygon and it's not perfectly round and it can be wider in some areas than it should. You'd best try to get a solid thing to hold it and have the filament go through extremely snug to make it accurate enough. Though I really am not sure. :P Best of luck!
It's going to be not 1.75mm-ish, but exactly 1.75mm ;) twitter.com/properprinting/status/1332036689368330254 I've got a fun and experimental idea to get this done. Hopefully it will work!
Yay! Long time no smile,Jon! I must say that I foresaw the filament breaking, but I’m sure you will find the Proper way! Oh, I miss the usual intro,mate. Looking forward to the next video though. 🤘
Probeer een eigen filament extruder te maken.
Dan kan je dat ding voeden met ABS korrels.
Succes kerel --> ik blijf voor je duimen.
You may try to use pallette machine connecting several spools of same filament