DIY Metal 3D Printer (part 2) 

I've seen worse benchies on regular 3d printers, it's impressive that it even works at all, for "art" it's a really nice paperweight that can also catch on fire so it's even multifunctional, what else could you want more?

dude, the fact you turned an ender 3 into a "completely functional" mig-3D metal printer is pretty damn impressive. I have a year left on my aerospace engineering degree (after working 5 years as an auto mechanic and doing fab work on a couple race cars) and I would say 80% of the other students wouldn't even know where to start on a project like this.

i remember when plastic filament printers were just as bad as this, and look at how good they are now. this only takes a little time and effort and you can make it as good and any other printer, it is already closer than any like this I have seen.

There’s not that many people who could pull this off. I’m impressed.

This is perfect man! This is a tiny part compared to the nozzle diameter, if you print bigger things, it will appear more precise… 👏🏻

Wow that is a brilliant V2 I’m seriously impressed you can see it’s a benchy 😀

You not having at least 250,000 subscribers at this point is absolute proof that the RU-vid Algorithm is broken. I enjoy the honesty and you showing the real world failures. When watching your videos, I feel like I’m just hanging out at the shop working on a project with with my buddies.

I'm sharing your channel with my students for sure. Good design discussions. Love your style. Props to your mathexpert as well.

1. Put your printer i box. 2. Change to 0.6 mm. Wire. 3. fill the box with an argon atmosphere. (Or CO2 maybe nitrogen. Use nitrogen geneator) 4. Print 5. Option add cooling system 6. You can use a split unit chiller 7. Put the inside unit in a vessel 8. circulate the gas from the printer box through a carbon filter. 9. And through the chiller vessel. 10. And back to the printer box.

Love the way you are thinking! -- I think your arduino was just getting some voltage/noise from your welder! It doesnt take a lot of noise for it to wipe out a cheap voltage or resistance driven 5v analog sensor like that. Esp with a long lead, running parallel to the welding conductor, Plus you hooked it to the (grounded?) metal sheath of the welder with a conductive metal pipe clamp! In your defense I think the EMF would still give you a problem even without metal-to-metal. I had a problem using RTC temp sensors a length of 20 ft using arduino. I ended up using a digital sensor that just sends communication back (ds18b20. But it is not high temp compatible ) You may want to look into IR sensing! I just watched a video about "DIY IR camera vs Commercial IR camera". You could probably utilize the cheap 8x8 pixel sensors that cost about 100$, since you arent using it for "video" but for temperature monitoring remotely (without contact). Or maybe there is a useful sensor inside one of the IR temp guns? Work to find something that reads the temp accurately, FLIR, IR non-contact thermometer, or otherwise. No matter what it is, an interface circuit to feed the data to your printer control can be built. I'm subbed now and tuned in closely!! More power to you!

That's crazy. Collectively we're still trying to figure out plastic printing, but when that's done the next hurdle will be metal 3d printing; in that regard, the most advanced devices will be inept, bulky, and prone to failure, with a printable area no larger than a shoebox. Then there's your printer, which isn't accurate but works, and works kinda well, and works over a large theoretical print area

You should check out the tip dressing stations for robot welding. Maybe like a tip dressing thing that flips into the workspace for cleaning, and flips out of the way for printing

Glad to see my boi Ditto around for the start of the vid

Subbed! I think I stumbled across a video of yours a year back or so and then lost track of the what and where. But boy do you do entertaining videos with awesome results for what you have to put into it. So much fun. Keed it up m8 👍

too cool! i appreciate you dedication to entertain us. id be pretty proud of that boat too. what you gonna name her? the printer i mean.

Nice one mate. I am surprised how well that turned out

If you don't have printer's original heaters on it, it will happily run off 12V. The motors won't quite reach the speed so happily, but the holding torque stays the same. Marlin codebase is actually a proper pleasure to read through. Last thing I had to change was the interrupt pin initialisation logic because uhhh I was janking some things into my board which don't belong, so I got in deep, don't you worry.

Great video... I have some ideas. I watched last night and thought maybe thinner wire would help with heat and puddle control. My second idea is much harder but might be the best solution. Move the motors and submerge the entire setup in liquid similar to underwater welding. You could use different types of liquid to control heat and oxidation. This would be hard but maybe the ultimate solution.

An encoder on the wire feeder, or a stepper motor in it, would give you more control. As this is how the plastic extruder works you can use the existing firmware unchanged (the Ender might not have encoder/servo built in but they certainly exist in the reprap firmware world)

It may not look pretty, but it's impressive that it works at all. Well done! I'm looking forward to future improvements.

Holy shit when you attached a servo to a knob, that's something I've always dreamed of doing, but didn't want to get fired at work. You went full send, and for that, I sub.

I think you would have better luck with a Tig torch, and the filler off to the side, no splatter. Also I know cooling might be an issue (maybe run water loops on your steppers), but since it's in a cabinet, you can just purge the whole cabinet with Argon (or whatever shielding gas/mix you're using), then you wouldn't need a cup, and could lighten Z axis.

Wire speed solution, pure genius!

Thats really awesome. I wish you were my next door buddy, kinda like Barney Rubble.

Man, i am realy impressive with what you achieved! After the first video i wouldn’t try again, but you did it! Glad to follow your videos

Whoa i am really impressed. I always wondered if that principal would work.

i like how u show the failures and explain the thought process of the solution.

My gf and I collect benchys and I gotta say this is one of the coolest I have ever seen! I had a thought about doing this exactly, but never took it anywhere. I'm happy to see someone actually doing it! It came about while the gf and I were talking about free hand 3d printing with a mig welder like one would with a 3d pen. That progressed into adding one to a cr-10 or ender3. Which is exactly what you did! This is awesome! Please keep up the great work!

The wife listening in whilst I watch: "is he drunk". Excellent effort! Really pushing it to the limits there.

That was awesome. I love your persistence and positive attitude through the entirety of the project! Definitely going to check out your other vids! Subscribed!

Your channel is educational, creative, and very entertaining. Keep it up man.

That is amazing dude! I got so excited as i could see it working in the end. Keep up the great work, i love the videos!

great build man....keep them coming.. i see you are getting healthier... props to you..

Frankly, that looks and worked a crapton better than I thought it would. The baseline work shows some promise. Worth a sub, at least.

You better continue tinkering with it I can't wait to see what else you come up with, I always wanted to try this idea you are playing with... Hopefully in a couple of years I'll have a place to put a workshop.

Nice work! In that enclosure you've presumably got lots of ozone and UV light rattling around. All the plastic parts, oils etc are going to get chewed up, in particular the belts and bearings, so watch out for that. I was once doing a project with a uv lamp on a motion stage in an enclosure and I kept on having problems like "Why does the oil turn into tar?" and "Why are all the o-rings turning to dust?" before I caught on. If you keep going on this one, you might consider putting an enclosure just around the torch--you'd lose build volume, but then all the delicate bits would be better protected. Also, the thermocouple was probably either wired backwards or picking up emi from the torch. You were on the right track, but the signal from a thermocouple is only ever a millivolt or so (10-40 microvolts/degree C depending on type) at 10-100 ohms.

I'd love to see you do a general shape, with intent to machine. Then machine it down to an actual spec with basically just one quick CNC operation. Thats how industry 3d printing works most of the time and you could actually make some pretty good parts this way.

Love how you keep going on a project and how it dosent let you down. Man I would have given up I wish I had you persistents.

Your resident math expert is hilarious, you should keep her around.

Most epic Benchy ever!

Looks good! That x belt path isn't parallel, so you need to fix that, because right now you don't have linear movements and belt tension will vary across the travel length.

I enjoyed your video. And kind of panicked when you connected an arc welder to thermocouple to arduino to a laptop which is a bad idea as your laptop could blow up if there is a power loop(i have seen someone doing similar and their desktop caught on fire and nearly burned down the lab). Instead i advice using a digital isolator or a bluetooth wireless or an esp32 wifi/bluetooth interface to read the data.

Great you never gave up. finally that stupid boat in proper metal !!

Your 3dbenchy is gorgeous, I really liked that

you are lazy and genius, and only lazy genius comes up with super easy innovative solution , I subscribed - love your style

For your content is top quality work, top quality production, and top quality entertainment. The stuff you build is just amazing and unimaginable and so inspiring, Thank you I love your channel

For what you have said you have a problem with heat, try using some cheap water blocks that you can buy on eBay and sandwich them between the build plate and your metal sheet (base)

Keep up the good work. I love this even though I'll never do this

Actually turned out better than expected! Great work!

I just love to watch your videos you're so entertaining ! Congratulations on the 3D Benchy too, not an easy task to teach a printer software to weld !!!

Nice fart (for art) machine man! I remember when hackady posted some crazy guy doing the same thing with a glue gun. A year later ppl were printing with all kinds of plastics and now you can get a solid 3d printer for like 150 bucks! I wonder if someone took the time and effort to develop purpose built lance and feed machine if this could be a thing i an cluple of years … like more than just a f.art thing …

With it in the enclosure, seal that thing off and flood it in argon, you would still need some flow for cooling, but I wonder if it would make a difference or not.

THIS IS WHY I SUBSCRIBED WITH NOTIFICATIONS

16:19 absolutely amazing , you did too much work and a lot of efforts ..... well done ... perfect

Man, this needs more views! This is awesome

Mad props for making that happen!

Consider pumping argon into the chamber. It would probably help a lot with defects and produce better results.

best benchy EVER ! Surprised the x-axis belt didnt just melt :D

As a graduated student in engineering, i am pleased to see some are willing to take on challenges.

I only see a perfect Benchy. This is totally dope!

Nunca vi algo tan bueno!! Muy bueno!! Saludos desde Argentina.

Some vase mode print would work fine probably

Fantastic! Keep pushing at this. Will quickly improve.

Your videos are honestly awesome. This kind of stuff may be a bit janky but the fact that you built a metal 3d printer using a mig head is mildly magical to me. I'm really curious if it can tackle the beast that is mig aluminum?

Art is just a part without the p. Great job

Killer work dawg! I’m curious to cut that in half to see what the layers look like. There is most definitely some practical use ideas. Although, printing metal art parts is also f’n RAD! Big ups!

To be honest that’s better than my 3D prints.. you should be proud man. Awesome job… Big like and subscribed.. 👍🏽😉

Cant tell you how many times I’ve actually thought about this. Glad to see you go through with it. Way better then what I could of ever come up with!

That’s really impressive. Could you use some type of sensor to monitor the metal temp and arc color? I might be wrong but I think the arc color or brightness will change in a measurable way as the base gets hotter.

As a sheet metal fabricator that makes lots of enclosures i was pretty impressed with yours.

Well I was subscribed from part 1, I can just uncheck-recheck the mark and the bell. That intro made me scream. I want a machine like that. Because just yes.

Amazing! I really loved this video, thank you for sharing it :-)

That was awesome! Thank you for making me smile so much.

Nice :) It's all about the scale, imagine the same model printed 10 times bigger, it might look not so bad...

More power to you man! To be honest the real art is in the building not the result.

This is really epic bro, I love the style of your videos definitely subbing.

eye decongestant. great job.

You could maybe use some kind of optical sensor to monitor the welding. Probably behind dark glass. But it should instantly react to increas or decrease in the light which you would think has correlation to the welding.

That is awesome, man!! We’ve been talking about what would happen if you attached a mig to a 3d printer, but you beat us to it! I still want to give it a shot, but I don’t think we will do any better…

LOL, here from Hackaday, and glad I found your channel. Subbed because you're funny, straight to the point, and especially leave the mistakes, warts and all in which shows an important part of any project, learnin'. I think that turned out great, rough? sure. Buts its made of metal with a welding gun, not surprised that there is a bit more slop at this stage. I expect it you wanted to throw a couple grand worth of motors, machined surfaces and some sort of slower welding process, you could probably do better. If you don't want to throw cash away, keep iterating, and it will also get better. They don't weld wood, and they don't glue-up steel works, for a reason. You're kind of breaking new ground, like the citizen-scientists of olde.

Great stuff man really interesting saw that veritasium vid on relativity space too recently, i have an ender 3 and a saturn sla myself can't wait till metal printers get more affordable tho ,wondering now are there any opensource people working on anything for openbuilds for the additive metal fab side of things will definitely be looking more into this myself now

Considering the layer width/height constraints of the weld bead, I'd call that a very good result. You just need to scale up in size since resolution is very low.

While I can't shake the feeling I am watching a caveman do high-tech. I wouldn't be surprised if you end up launching a satellite. Now for some ideas. Honestly, I have no idea how to actually solve the warp issue. Thermal expansion relief cuts? Active cooling? Metal sandwich (Fe|Al|Fe)? Thermal monitoring. The thermal couple got overwhelmed with EMI. NTC resistor could work, if the temperature was below 250 Celsius. There's some infrared sensors, Andreas Spiess has videos on them. A enclosure allows for Argon atmosphere, however requires intense active cooling, so isn't worth unless aluminium is use. So... What I would do, use a combination of thermal expansion relief with metal sandwich, attach thermic sensor to Aluminium layer. Forget about the Argon (still love you, Argon) setup a mini drill with wire brush for cleaning (something tells me it's not gonna work). Uhmmm... Oh, I can use the voltage drop across the welding cable to measure current, and also voltage... Now I wonder if I can use that plus layer stop and start to use regular argon purges... Maybe it's better to use lasers and metal powder, however the hazards are too high. P.S. ok I would be surprised, just not in a "wait, what?" Way. More in a I knew he could do it, still.

You are a genius, go on friend we want you to perfect the technique

Well done persistence is key! Maybe a change belts with sprocket & chains or ballscrews ( can get cheap Chinese seconds (bent), with a little work they can be used,) How do they position plasma torches on cnc, surely that tech could help here. Keep the up the content, always interestinh

When welding body panels, you can use an air compressor and air sprayer nozzle to cool the metal faster to avoid warping... Maybe get something like that set up to increase quality

I'd love to spend a week with this guy building cool shit.

Thank you! Makers like you will make this work. Thank you!

Total kudos on that one man !

If you change the liner and polarity, you can run aluminum wire, which will run cooler and freeze faster.

That’s so cool dude! You did such a great job with that thing

To fix the extruder issue, you can change e-steps in the printer config interface without updating any firmware. Or possibly you could increase the flow rate in the slicer.

If i could make an observation, i would use a method employed for pipe welding where the pipe is enclosed at one end and then an innert gas is pumped in at a desired pressure. You could essentially get rid of the nozzel on the mig gun and use gas in the chamber to stop all your problems of contamination, let me know your thoughts please, great video thank you. Also just a thought, can i ask what determined height is each pass?, because my thoughts would be that you need a type of secondary functiin to the printer that shaves off the very tiniest amount of weld from the top of each pass to maintain a set height every time, sort of a grinder that passes each the pass and levels it and cleaning it before the next weld pass. Thanks again

7:27 I see a fellow Technic Launcher enjoyer xD

убери воздух из сварочной кабины, применяй флюс чтобы уменьшить шлак при сварке, применяй охлаждение воздухом участков детали и применяй подпорки чтобы деталь не вели. сколько уходит времени на застывание от жидкого до вязкого состояния металла и какая корреляция температуры идет. Идея очень классная у тебя, всех благ и успехов

That's pretty dang impressive, SUBSCRIBED. Maybe part of the issue with the later layers is due to the belts stretching due to heat?

Hey man, when you were testing your amperage fluctuating, if you switch to spray arc or pulse mig it should change from variable amperage to constant current which should be easier to find a consistent slope.

Great job man, I hope to have skill sets as diverse as you some day.

Awesome I love this! Someday when I get a welder I want to make this!

love ur janky contraptions