I know what everyone's getting for christmas this year! (barnacles) outro music by a friend of mine, check him out! / badamericansofficial elevator music: bensound.com yakety sax ripoff: toolofgod.com/m...
With robotic MIG welding we typically program a pause at the end of a weld bead where weld settings change to less heat, more wire feed to fill the end crater that develops.
I think industrial grade 3D printers that utilizes this kind of arc welding method always integrates with a CNC milling machine, so after building a few layers the material it can be milled to very high precision.
That half-sphere was really impressive for some garage project built on ender 3. I think you proved that hobby grade metal 3d printer is totally doable, hope we will see them on the market some day.
From what ive seen people say that an argon filled environment helps a lot. So maybe a case full of argon made of tinted glass/acrylic could be helpful
you can also just crank the esteps to up how much it extrudes per mm, for example if you double it, it will move the extruder double what it should be. or instead of using fan pins theres laser, and cnc support in marlin using different gcode, so you could wire the the trigger from the welder to those using a relay. Or something like that.
@@cranktowncity Welding machines for doing TIG push a stream of Nitrogen or Argon directly onto the weld. These can be had for MIG as well which is what he is doing here. This is simply a case of getting a new welder with Gas line and a Cup for the tip which holds the gas around the end. This would save you the headaches of trying to make a full box of gas.
@@seditt5146 There seems to already be a an inert gas line feeding this torch, see at about @8:20 and it is the clear hose coming in from the side. I guessed there was no gas coverage earlier in the video, with the crappy looking aluminium weld-beads, but that could also been from trying to use argon/co2 mix (which is for steel) instead of argon or argon/helium mix.
Metal Matters is trying to do an actual powder laser based system. But these welder based systems are really interesting, as it should be quite simple.. its just a lot of dialing
As a practical machine that was.... erm.... y'know.... erm..... well.... not so good. But as a proof of concept.... a right hum dinger! Keep up The Great Work.
Well, the example you based your idea off of was much larger. Maybe this is one of those things that has to be done in a larger scale? The G code is set up for plastics to heat up and cool down in as short of time as possible but still maintain a certain rigidity. Steal would need more time to cool down so on a larger scale with the same settings this would make metals more manageable. Or option B would be to make your smaller prints in a much colder environment like in a refrigerator, a walk in freezer or Antarctica. Next problem would be how to pull your build off of the build plate. With PLA or ABS it cools down and is easier to pull off an aluminum plate. With your set up you would have to cut off the piece. So for the thumb printing, you would need to cut it off or add bridges to raise it off of the platform...but still have to cut it off. But it may need less grinding work that way? Love the maverick style of inventing your way through a challenge by using stuff already sitting around! Would there be any advantages to having an X and Y base with a Z nozzle or would it still have the same outcome? Would curved surfaces like your sphere be better if the G code were to cool down the material once the welds were in closer proximity to prevent the molten metal mishaps? Again, love the content! Thank you for your channel!
In the big example, they have developed software that calculates all the weird heating/cooling distortions that take place and adjusts the printing accordingly. So its not the size scale that magically makes their tech work. There's much more to it.
This was fantastic! I just thought to myself, "I wonder if anyone's hooked up a welder to a 3D printer?" So I do a search and bam! Here it is! Great job!
Rulof from Italy sent me here. He made me see a MIG welder on a bedslinger, tracing what looked like the first layer of a 5x scaled benchy. I was impressed to say the least!
The wire releases gas as it melts, which forms a bubble around the metal to keep the oxygen away from the weld (he said this in the video). If you try to use a fan, it would blow away the protective gas and weaken the weld.
Great concept. My thoughts are stick with Mig, it's going to be much easier than Tig. Then work on being able to control your Amps and wire feed. Ramping the Amps down once it gets running should be possible as it's already carrying heat. Maybe to start with do it man to prove concept. I can see a cnc welder/plasma cutter in the making. Stick with it, it's gonna be excellent.
I think that tig torch would be better way to go ,and then make a wire feeder for it like mig has.EWM has a auto wire feeder on some of their machines.Maybe thats the way to go.
Seems like Tig would be the better process here. Maybe less build up but better puddle control. Also, +1 for mentioning the oft ignored M-codes. As your past President would say, "...not a lot of people know about M-codes."
@@kundeleczek1 True. TIG is a manual process like soldering using a separatley hand fed rod of material. What I know not even a million dollar industrial robot can TIG weld because of the nature of the process being impossible for any machine to handle. The wire being fed in separatley from the side without touching the electrode that needs to in turn be around a max distance of 0.2 mm from the workpiece is something only a human can do. I TIG weld a lot myself at work at the moment. It a really fun but also really tricky thing to do and I'm very happy to master it good enough that the result in the end looks very profesional. Its a true skill to master TIG welding.
Have you tried doing a tack weld? I have done a progressive tack style that looked like a tig bead. kind of like a line of coins. clearly that continues bead is too hot. Set up a pulse cycle of, say one second where 2 tns. is weld and 8 tns. is cool down. Mess with it that way with different combos. You are way smarter than me, I can't even set up a regular printer, but I've welded the seam on a power-saw muffler, very thin, and it came out nice using Mig. Just a thought.
Great video man gotta post to mess with the analytics for ya, I hope your at or past 10k by Christmas, you may as well live out my dream getting paid to mess around with stuff!
Smaller wire means less heat. Ceramic or fire brick material bed clad with tungsten, with active cooling like what is used in steel mill furnaces. I would also enclose the print volume with shaded materialfor uniformity of heat, eye protection, and shield gas retention. That allows even more cooling below the bed. Swap the rubber wheels out for ceramic or perhaps high temperature silicone? Drive belts will be an issue due to heat as well. A flexible temperature resistant material, or consider rack and pinion or other linear gear setup to replace the drive belts. I would also recommend relocating the stepper motors further away from the print area. The heat will likely destroy them quickly. Just extend the linear gear portion to mesh with the gear mounted to the stepper motor further away from the heat source. I'd do the same thing for the limit switches. And absolutely if you can control wire feed via the stepper, I would. I bet the drive rolls from the welder could be adapted to fit the motor that normally feeds the extruder. Mine already has a spring loaded tensioner . The drive rolls for metal wire I imagine need considerably more force than plastic to accurately feed without slippage.
I think the speed has to be higher, also there has to be a good fan to cool the soldering filament instantly there shouldn't be this red macula after it is soldered, while this red means that it will deform with the next layer , if it cools instantly it will not deform, the heat has to be only at the welding point it does not have to expand to nearby points, that is just my think and your idea is genius
Pretty cool. I wonder if you printed objects at the maximum size for the bed if that would let them cool more. The wall thickness would be a smaller fraction of the overall objects size and probably wouldn't retain as much heat.
That or that blower fan on the object. Instead of the board. It should help but it will blow away the shielding gas... Which could be fixed by fluxcore maybe
if you going to do a metal 3d printer then set the machine up in a purge chamber.. assuming its steel that you trying to make the parts out of then fill it with argon. now this is the expensive part.. you need to get your self a double pulse mig set up or even better a fronius cmt welder.. welding with a flux core/innershield wont work as the porosity and slag layers will prevent the part from actually becoming whole in the grain of the material so any part will be next to useless in a serious application. your also need to use a water cooled set up as the torch will not hold for long on the duty cycle of a air torch.. not to mention your need a machine that will do the needed amount of amps for 100 percent duty cycle otherwise you going to torch your machine in the long run.. welders prefer to run at 60 percent duty cycle or higher.. the 20-30 percent are only there for the odd job. hope this helps you on your quest
Please keep trying and don't give up. Maybe try a less powerful welding machine. I think the problem here is the high temperature with which your welding machine melts or works.
How about current sensor loop to hook up to your arduino controller and run the welding wire through it, add a voltmeter between the welding wire and base plate. Power = Current * Voltage. Integrate (smooth) the power and use that for feed control. I have no idea how much current and power fluctuate during welding, but imagine both do a lot, so both are required. Fill the cabinet with Argon? The Russian built Titanium Submarines inside Argon filled buildings.
You might have the skill to retry this on a modded 4 axis with rotating printhead (theta is the fourth axe), that might solve the issue with complex models where the welder hits the prin- the melted bead. And it would be the bedslinger version of a 200k bucks worth actual welding robot! (those are 5 axis actually ok)
Have a daenger 👍. Why not welding with spots and a little pause when there will be some pressurised air blowing onto welding area? Just curious what's going to be in the next video.
I think you can get some control of the feed rate without effecting the gantry movement with the filament thickness setting in the slicer. And I just noticed there is a part 2, so you probably solved it already. Oh well, I needed a comment and ain't deleting it
I think it might be easier to implement if you could modify a non cartesian style printer where the bed stays stationary like a delta or core-xy style printer. With those it'll be much easier to design cooling and maybe even an enclosure as well since the footprint of the printer stays the same. And also a printer with linear rails instead of v-rollers for axis travel will be a lot more durable since there wont be any plastic/rubber wheels to melt. If you do go this route maybe check out the Flsun SR I've seen the prices drop to high 300s at times and its got all the things i mentioned.
Also building a feed system for a delta or core-xy would be a lot easier since most of them just have a feed system where the filament has a straight travel path directly from the top of the machine to the hotend
Problem is he’d be disassembling a over 1000 printer all to take it apart to pieces and severely modify it, That’s not a cheap job and doubt this guy wants to spend that on building this. Tho you make a decent point.
If you set the z seam to random in your slicer it will put the beginning and end in a different place each time which will mitigate the issue you're having
Awesome work man, hope to see you keep trying at this, would be awesome to see you pushing the diy metal printer industry :D, also try cura slicer, its free and it has SO MANY settings to play with, whether it be speeds, pauses, where you start each seam, etc. Good luck bro
😅 Man ✌🏻 i was looking in that topic for 2weeks now and thought about prozesses like using a plasma cutter, a laser or tig welding. I saw your 3d tig THING. Thats the best way to go for now. BUT - for most the parts we like to print, its like 3d printing!!! So we need MICRO TIG TORCHES like wp9p's, a TIG WELDER WITH ADJUSTABLE small AMPERrates & PULSE FUNCTION stuff, a BETTER WIREFEEDER like for Soldering and so on. I already 3d modeling Holders and stuff to use on a 3d printer, to visualize. LESS HEAT, THIN WALL THICKNESS is the way to go. It sucks heaving such ideas 🤦🏻👍🏻 SUBSCRIBED
Well the result at 8:53 is HOT, and it looked faster than I remember an Ender3 could be! I am wondering why the benchy and not a vase. A vase without bridges possibly. This is one of the coolest...most surprisingly things in 3D printing I've seen so far! Google listens and heard that I just got into a welding class and dude, this content hit me hard! Edit: at 9:06 you were clearly checking layers adhesion. Don't worry, at worst you could turn off the fan 🤣🤣🤣
Amazing.love it. looking for anycubic printer but i hope these will get cheaper as its way too much for me, dont even have the tools dont know abouw welding either