DED Metal 3D Printer - Round 2
HTML-код
- Опубликовано: 29 сен 2024
- Looks like we're starting over..
Development of a DIY metal 3D printer, printing 316 stainless steel. This project aims to establish a cost effective opensource metal 3d printing platform. Subscribe for future developments & announcements.
Forum: www.metalmatters.co
BMAC: www.buymeacoffee.com/metalmatters
Laser Safety Goggles:
www.edmundopti...
Artwork/Music:
HOME - Moire Fringes
/ home-2001
Vivacity - Phello
Papers referenced:
"Measurement of actual powder layer height and packing density in a single layer in selective laser melting"
www.sciencedir...
"Hot-Wire Laser-Directed Energy Deposition: Process Characteristics and Benefits of Resistive Pre-Heating of the Feedstock Wire "
www.mdpi.com/2...
Once again thank-you for sharing these ups and downs. It is fascinating to watch and take in. Your updates are always a high point for my RUclips time.
Incredible work! this is the type of disruptive technology that gets me excited to see the future. I hope I live long enough to see that every desktop has one of these. This is a huge amount of public groundwork accomplished by one person.
As for gas coverage I'd recommend using TIG gas lenses. They provide laminar flow and are dirt cheap. Keep up the inspiring work!
Edit: It seems like 27 gauge needles have a nominal ID of .21mm, which could be of use centralizing the wire as it goes through the nozzle?
I have some 27G syringes and they are very flexible. I'm sure you could sleeve them but you do need a reasonable clearance to avoid jamming.
i am also developing a "cheap" 3d metal printer :-) and your work and energey gives me the energy to go forward with it! keep up the good work you are my "idol"
Do you post the progress anywhere?
@@marijntuijl unfortunately i am not a youtuber :-) but when i succeed yes i will make then a youtube video. (but most probably under a different account not my private account)
Maybe you guys should do some collaboration and share ideas. It might shortcut the process for both of you.
@SnApO what technology are you using? powder based, laser, mig welder? And what material are you trying to print?
@@the_normal_guy9308 I did not see that technology used yet, it is powder based and would allow conductive and non-conductive materials with a melting point
You could use a Handheld Welding System With Auto-Wire Feeder and modify it !!
Impressive work mate!
Always amazing and fascinating to watch you work through your process👍👍
Looks very promising.
Have you ever wondered about the possibility of being called the father of affordable metal 3d printing in the future. How does such a thought make you feel?
I wonder if for the wire feeder you could use a 3d printer bowden extruder such as the Mobius M4 which is mostly 3d printed parts and a few metal bits from amazon for cheap, since it's geared down and has adjustable tension with a spring. You might need to replace one of the metal rods with teeth (since there's two in the kit) with something that allows for a smaller filament size.
0:09 Holy shit is that THREE fiberoptic lasers!?
I wonder if you could use some thin sheet of metal so that it's cheaper to get it off the 3d printing bed and place a different sheet, kinda like using glue to cover the bed so the filament sticks to the thin dried glue instead of the printing bed?
My guess is to try different thin metal sheets, maybe tin foil(I don't know)
Heated Copper would be by first guess.
I think this is right way
I have an idea of how to do this easier and better: have both a MIG welder and a TIG welder set up on a 3D printer scaffolding. The MIG welder will lay the bead and the TIG torch will re-heat the bead so that it looks better. Basically an Instagram trick.
Positivize Pressure caused you to scrap the other model/design? OK, what about using negative pressure with like magnates, would that work?
Where you buy this laser modules ?
I have a pretty well equipped machine shop at home and can manufacture smaller and mid sized parts pretty precisely (+-10um or under). I find what you are doing really amazing and if you are interested I would be available for manufacturing some prototype parts for you for the price of the material and shipping.
If you want to talk about this further youre more than welcome to shoot me a dm.
Hello, man! Can your machine make motorcycle engine parts
This guy only prints nylon but it is an SLS maybe you can adapt it to do metal!
I think your link was automatically removed. Are you referring to SLS4All? It's a nice build, but the price tag is a little discouraging.
Nice
Where did you go?
Still here. Still working on things. My attempts build a wire based metal 3D printer did not work out.
@@metalmatters Glad to hear you're still here! Sorry to hear it didn't work out. Your content is still good even if the project doesn't succeed. You don't have to only make videos when you have something that works.
Reminds me my mtech project, thin wire is never a good option
I agree. I'm interested to know why you think this though?
@@metalmatters I used MIG power source for 250 um wire. Even at very low voltage, the moment wire touches the substrate, it will get red hot and the integrity of the wire will be lost.
@@kk3074 Ah, I guess the circumstances are a little different then. I would expect there would be some level of threshold power required to gain enough heat to melt the parent material. 250um doesn't afford you a lot.
Your biggest problem seems to be lack of research and trying to improvise too many things based on what you feel should work. This means many problems that should never happen end up holding you up. Also plowing full steam in a totally non-workable direction based on assumptions. Starting with a 10 beam BLUE laser was a total waste of time and money, even a 10 minute google search could have prevented. Different wavelengths are absorbed differently, and blue being very short wavelengths are relatively bad at heating things.
As I pointed out in a previous comment this should be more than doable for under $2k. An F theta lens costs
Ironically, 10 minutes of research would reveal to you why shorter wavelengths are the industry standard.
buddy wh stop work on that galvo SLS?? someone threating you?? or they pay you to NOT?? finish it,?
No. It came down to cost, safety, and burnout. The machine was looking like it would cost upwards of 5k, potentially more, and require a significant investment on my own part. It was becoming more detached from the original vision. I also want to avoid metal powders; I think a lot of people underestimate the hazards involved.
Despite all of this, I have been working on the DED printer. I'm on the third head design since the last video; I've just been embroiled with challenges.
I'm studying to be an engineer or material science- not 100% sure yet, and am new to being a maker/designer. I am hugely interested in what you are doing and am taking notes with everything I see you do. I hope I can replicate your work one day and contribute to the knowledge pool of 3D metal printing.
I can feel your frustration, but I truly, sincerely appreciate the work that you put into developing this project and I really look forward to your progress. Such an amazing goal!
Very nice project !! Im following your work with a lot of interests
Excellent work! Since you have resorted to wires now, and with shielding, please take a look at "coaxial electrospinning needles" they are axially integrated needles. I came across them to attempt powder injection to my hho torch (printing with aluminum/chromium oxide) setup. With wire feed directly through, you can have your shielding gas feed directly over your wire. There's plenty of cheap-ish sellers.
Thanks for that, I'll have a look into it. The very first design used a coaxial feed for both the feedstock and shielding gas (basically the thumbnail with a hole in the base, hence the screws) but wasn't well thought out, so I went with something more conventional.
Never surrender (only pivot)!
This chap needs funding, could spur into something great.
Awesome, can't wait for round 3!
I've got some bad news from the future
if retraction is making the pooling at the end of the extrusion more thats gotta be from the wire puIIing the pooI of metaI during the retraction because unIike with plastic the metaI doesnt break tension with the retraction. The probIem stiII is that its on the nozzIe and wiII be deposited at the beginning on the next Iayer. maybe if you wait Ionger before moving to the next Iayer for it to cool or even extruding a bit of wire then retracting to give the metaI in the nozzIe and the nozzIe itseIf space and time to cooI and alIow it to cIeanIy break off the extrusion.
It seems like a lot of problem can be attributed to instability of the print head setup. Maybe it would be better to move the print bed instead of the head.
Oh this is getting good. it's getting so close to just being like, perfectly usable
A super fast retraction speed can help with ball formation. I'd recommend a high speed recoding to get the speeds at the start and end of a bead just right
Dear sir,
Im sorry but I dont know your name, I following your portal from several years ago, because I'm engineer I find some of your issues in any cases that you have. I give you some advise in below texts to help you and other guys.
Please try to use standard electronics board same Mach3 cnc board I use that and it works perfectly and simple.
Calculate termal extension in any parts that have contact with laser beam.
The powder bed ned to redesign because of termal expansion and heat reformation.
You need to fix the powder bed with fixture technique dont let that to rotate or re placing.
You need to remove air aound the poweder wen its busy with sintering, you know that oxidation dident let the new melted layer to joint on previous layer you can use clused chamber with argon gas or co2 gas, dont forget that you doing Sintering or welding and not cutting.
You need to cooling the actuator and other plastics bodys as well, dont forget to cool down the lens or lenses frames.
Finaly when you use the powder tacke care that poweder stof have 0,00004 mm as max. sizes please let the machine working more on each layer , you need to use the section layers closeer to each others and dont put blaser beam for long time on one point.
I want to visit your website but its out of access, I dont know whay.
With my best wishes for you and your familie.
Farbod.
Hi Farbod. Thanks for your insights. Many of the above issues have been addressed. The system is semi operational now. The website was taken down due to inactivity. Hopefully I will have together a video for you all soon.
Been following this build for a while now, always excited to see your videos pop up. It seems like you're having issues getting precise parts, I dont know how impractical shipping would be since I'm in the states, but I'm willing to make you parts if needed. I have a 3axis cnc, a mill, and a lathe
Appreciate that
Hello, man! Can your machine make motorcycle engine parts
to get rid of extra ball of weld that builds up on the wire at the end of a weld just snip the wire at the end of each weld. you can put a mechanical sniper in your printer and have it snip its self after each pass. Welding robots do this
Jeeezzze man , freekn guy could build a satalite in his closet
Good job. Keep it up!
Trying to think outside the box, what if you put the printer inside a box? Empty the box of air, as much as reasonably possible, with a vacuum pump. Then fill with CO2, so you don't have to worry about gas flow over the weld. It might be worth it just as an experiment.
You also have to look at the metal transfer. The two basic types of metal transfer you will be seeing, other than at the start of the weld, is either spray transfer or globular transfer. Ideally you will want spray transfer as it will lay down the most consistent bead of metal. Unfortunately, it is difficult to maintain this with your setup. As the power supply heats up, base metal heats, and wire warms, everything will change and the amps and volts that have you the correct settings at the beginning will need adjustment. The best, and most easy way, to counter this is by using a smart welder. Look at miller welding's MillerMatic line of welders. They will adjust voltage, amps, and feed rate on the fly to keep the weld working as intended.
Wouldn't it be better to just 3D print something in plastic, and then, 3D laser image scan that plastic component into a computer, and then, have that computer cut out a ceramic mold which you would then pour metal into... thereby 3D printing a metal part based on the original plastic?.. so basically, it would just create a negative space within a ceramic block that would be the same dimensions as the original plastic but in Reverse so it would be hollow, and then you just pour metal into it..
The mold has to be in 2 parts with the right kind of ceramic, and some kind of anti bonding agent, like sand, in between part, so it separates properly,
The pour of molten metal needs to be one, uninterrupted stream, and the mold needs to be pre-heated, and agitated possibly while the metal is molten, to reduce any bubbles or warping.
When thats hard, you crack the mold open, polish the parts with the 3D printer and a diamond tipped drill bit, so it's the same exact dimensions of the plastic model....
Then, you heat it to a certain heat, and anneal it in an oil bath, if it's hardenable (high carbon) steel.... otherwise you just anneal in a kiln.
That's how you make a fuckin... King Tiger Tank
Great Job! I operate two DED laser wire metal printer in our lab at IIT Bombay. If I could help in any way please let me know.
Congratulations on the work so far. Impressive perseverance
dude, get rid of the gas cup and make a chamber for it. even if it has leaks you'll still use less gas. Use an oxygen sensor to keep it filled. less weight on the system, less gas usage, better gas coverage which will be especially important for variable speed and changing directions abruptly.
once its finish will you print out a metal mario?
Feed co2 into the wire feed to eliminate air leak
Its basically a cnc welder🧐 maybe try cheating by making lines using small beads instead, to make more consistent lines maybe also less strain on parts 💁♂️
Maybe we can adapt this system in miniature form. Friction welding: ruclips.net/video/VZf_eGyxii4/видео.html
Why the lasers instead of an electric arc? Seems like you're using the arc for preheating, so why not go all the way with that?
Very interesting stuff. I'd like to attempt to compute a solution but I'm currently persuing a safer, multimaterial way of printing in resin. I'm looking to add continuous fibres too to reduce the amount of parts that would need the physical properties of metals.
Maybe some kind of atomised metal embeded resin is possible and a $1500 printer + $3500 sintering oven workflow could be a solution.
we need a consistant reliable metal 3d welding printer, for the future of self reliant 3D printing.
This project gives me hope that a DIY metal 3d printer is achiveable soon and also your energy is amazing
Thanks for the time you put in to the vids. A few thoughts in exchange: have read of Invar powders being V good for low-porosity SLS (L-PBF). About the excess metal at the start/end of lines, maybe you could have overcome that by more passes at lower powers, to sort of pre-heat the powder, then final pass.
Another Idea: by starting and ending each layer at a sacrificial/waste piece, to be trimmed off at the end. Or perhaps the start of each line had a something like an overlapping movement, sort of starting along the line a few mm with lower power, then building up to nearly full power at the line start, then back along? Sort of an intuitive thought. Hopefully you return to the project some day (dig it up?).
Re the wire printer: when you mentioned the wire pushing through melt-pool I imagined the build plate being a 4th axis, turning so that didn't happen. Maybe an Invar wire would have interesting properties? Similar to another commenter: why not have a sealed chamber instead of gas flow? Even if not a 100% sealed, there could be a positive pressure from gas to keep out oxygen.
Last thought - I read about Atomic hydrogen welding using an arc in hydrogen shielding gas. Perhaps this could be an interesting method of 1: producing the gas on demand using electrolysis of water. 2. Doing away with the inherent difficulties of working with lasers. Perhaps the arc is created between wires that are fed in? Anyways, this welding can apparently melt tungsten - using that with a tungsten powder in a kind of synthesis could be interesting?
Thanks for suggesting Invar powders. I'll have a look into them. I'm amidst reassembling the printer. I'll have a chance to play around with preheating as this time I'll be using a galvanometer. I have to admit, I'm not overly fond of the idea due to how much heat will leak into the powder bed and how it will affect the metal, but I think it's worth trying.
The wire based effort didn't use a chamber due to the expense involved in purging it and how much additional effort it takes just to do some basic experiments. I played around with an arc based process but found it to be too poorly constrained in terms of heat and deposition control. I don't want to rely on any form of milling as a post processing step.
have you seen the printer that uses electro plating to effectively print metal parts?
As per the second to last comment, yeah. I assume that is what you're referencing.
Hi !
I was wondering which kind of lasers you used in yourdifferents metal 3d printer . I'm interested in building one but i wanted to document myself a bit more on lasers before. If you have any sites and/or companies to recommend it would be helpful ! Thanks in advance.
The lasers in both this system and the other powder based system are diode pumps. You can find them on eBay or Alibaba. If you get stuck, send me an email via the about page.
Inspiring
I would love to progress to a metal printer, but considering all the big boys have gone down the route of SLS of metal powders I think for the hobby level of printing, doing it in resin first to create a mould and then using that mould to create your metal part is the cheaper option that produces higher quality metal parts than you would get from any setup like this.
I've gone back to working on an SLM printer. That should be able to yield a respectable degree of precision. Still a little more work ahead though.
Awesome stuff mate. Glad to see you back into it, it's a shame to see the end of the slm project but if you are flogging a dead horse then there is no point continuing. That square looks really promising. Thanks for not stopping you are an inspiration to many.
Amazing journey pal! I don't know nearly as much as you about the engineering but what about using aluminium wire in a machine with similar topology to the automatic soldering machines ?
how i see it u will haVE more problems how to cool in between layers , then how to heat it
The ball on the wire could be mitigated by pulsing both the laser and the wire feed, intead of always on laser and constant feed rate. some of the trickier TIG welders i know will pulse welds to make them look better (at the cost of strength). if the pulses were spaced closely enough, strength shouldn't be an issue. Perhaps the answer to a difficult to control feed is precisely controlled steps that resemble a steady feed?
Maybe so. I'd like to keep the hardware to a minimum and pulsing those lasers without killing them might involve quite a bit more work. It might just be easier to let the joint cool and separate the wire and weld by treating it like a fuse. There could be problems with this method as well though.
Hoping you don't completely drop the slm project. It would be cool to see the pros and cons of slm vs ded.
I've actually dropped the DED project. Too many compromises had to be made and I still had some overarching issues which I couldn't resolve. I have been working on the SLM project however: ruclips.net/video/qNF3d99Jk5M/видео.html. Dealing with some mechanical issues at the moment. Will have it up and running soon.
How about, for aligning the beams, you take pictures of each dot individually, and compare those images. This should allow you to better define where each beam is pointed compared to the mess you get with all dots on top of each other. Kind of like avoiding the diffraction limit in super resolution microscopy (if that means anything to you).
That's actually what I was doing less the photos. I just wasn't sure if it was happening at the correct focal length. Footage didn't really convey that.
This is such good work. I can't tell you how much we appreciate this.
Niiiic, that's quality hobby research!
If you decide to go back to a center feed could you use a servo or stepper controlled flexture to aline the beams and feed?
I've since redesigned the head, allowing for independent control over each beam. I've been playing with the idea of returning to a central feed design. We'll see what becomes of it.
some of those lines look really promising!
That's incredible work! This is the first I've seen of your channel and I'll be eagerly watching your upcoming work!
It takes a lot to admit that it's time to give it a rest. To admit that the constraints are just too tough to make happen. You're doing some really difficult and complex work.
I'd like to see some resin prints used for flow management of gasses, manifolds, etc. Hmu if you ever want to bounce ideas off someone.
Thanks!
Was just wondering what you’d been up to, so great to see the progress! Always inspired by your level of execution and dedication, thank you for the work you’re doing!
NICE , i alway wait your project ♥
Have you tried using a jeweler's oxyacetelyne torch instead of lasers?
No
Looked into Electrochemical Additive Manufacturing (ECAM)?
Briefly. It's an interesting concept. I think it would be difficult to achieve good results without developing a high density matrix for the process. I notice that Fabric8labs mainly showcase copper parts too.
@@metalmatters I agree, a matrix like what they have seems like the key. I don't know how much effort it would take to create something like it in the open-source community but I'd love to see it.
your music is so loud i can't hear you talk
Keep it mate, that's brilliant!
Have you tried "electrochemical 3D printing" ?
No, but I did see Sunshine's video.
Have you thought about using molten salt (in crystal form after cooldown) to round the surface where the laser is hitting?
The melting point of salt is 801 Celsius, molten salt is really good at heat medium
No, can't say that I have. Why would I want a round profile where the laser is hitting?
@@metalmatters I honestly do not remember exactly why I wrote that, but I think as solution for particles escaping! I did a lot of research lately for my graduation project, not sure what problems I will face. Now thinking to plug a magnetic generator for a metal powder fusion printer!
I'm so glad you're back, I've been waiting for your videos for a year now
This is a story of a stranger who I shall never meet, yet hold the deepest respect for.
Hi there. I love what you are doing and I wanted to check some more of your work, but it looks like your webpage is down. Is it down for good or only temporary?
Thanks! Yes, the webpage is down for good unfortunately. I have done a lot of work since the last video but it has been in vain. I think it might be time to return to SLM.
@@metalmattersNo! Nothing is in vain! Please post your last update at least, your work is passionating. Anyways, thank you.
great work! hows is your project going ? also have you considered using a mig welder to print ? I've seen few example it def has potential
Progress isn't too bad but slower than I would like it to be. I briefly experimented with a TIG welder for the sake of 3D printing, too much unfocused heat imo.
@@metalmatters keep it up! I meant the mig welder not tig, as in the one that already has a wire inside and it melts the wire on contact, it’s got potential but needs alot of experimenting
Great work and thanks for sharing! :D
Thanks for all the work you do on this! It’s amazing to watch and the possibility of a working prototype at the end, even if I only ever see it through a screen, is tantalizing.
How difficult would it be to add some shielding to the components from heat and reflected lasers? I imagine such shields and sheaths would get in the way but would the safety be worth it?
Something that could print pewter or zamak like a standard FDM would already be astonishing.
For this hypothetical setup, in my opinion, the trick to make it happen would be a laser constantly heating the layer on which the metal gets poured. That would be far more cheaper but I understand why you're investing so much on your project: steel would be a revolution.
Another approach could be a powerful induction heater that melts the very tip of a metallic filament and an ink-jet style technique, where micro drops of filament are poured one by one.
This could make a general purpose metal FDM printer cheap and versatile, given that the only setting to change for a given metal/alloy would be the power output of the induction heater. Varying the height from the plane, the power of the laser that heats the layers already made and the temperature of the gas may also be parameters that one may use to tune the printer.
Free public knowledge here, if anybody finds something useful, feel free to use it.
Love following your channel. Good job!
Very cool project, excited to watch your progress. Subscribed!
If using wire, is laser anymore the best way to go? What would be the benefit of laser, compared to arch welding?
I believe so. In terms of output, better resolution and faster translation speeds from better heat concentration.
@@metalmatters i think laser welders are going to be dropping in price fast, there are already some cheap ones in aliexpress.. those might be worth a look
Wow, seeing all the work you put into the old printer just to have to chuck it in the bin. That's such a hard decision. Your new square looks great however.
I feel bad that you had to ditch the slm project. You had put in so much effort. I'm going to build an slm machine starting next year. Good luck with your ded machine. What do you think of fdmetal, like desktop metal. I think chances of success could be more in that than DED.
Thanks for that. Yeah, I think it is typically a 3 stage process involving binders / pressurized kilns etc. I'm looking for something direct.
@@metalmatters I'm interested in the slm / dmls for the same reason. It is direct and it is superior in mechanical properties. And that's the only thing I've got experience in. Although it's messy, that's ok, i can get functional end use parts. I'm building a high temp fdm for now, gonna make some money with this and start with the slm. It's gonna need a lot of money, even if i use Chinese parts. I think it would take atleast 30,000 usd to build one.
@@247chiranjeevi What volume are you aiming to achieve?
Yeaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
Nice work as always!
There are wire feeders for tig(!) welders with resistive heating, you seem to have outdone those in terms of precision.
Have you considered using a tig torch instead of lasers? This might work better than a conventional mig/mag with your precision wire feeder
Thanks, and yes I have. That's actually how I ended up with the MIG wire feeder and TIG wire delivery kit. There was just too much waste heat at too lower speed for it to be practical.
Another Video!!!!! Love your stuff man, was just checking your channel a couple days ago for an update, what great timing 👍
Lol i though thought the thumbnail was three lense in handcuffs xD Like an artistic statement about the more cameras/or whatever, the more I limit myself🤔
Good afternoon, and the forum with discussions and your project is still closed?)
Hi. I've since taken it down due to inactivity. It's unlikely it will return. I may setup a discord in the future.
@@metalmatters I'll be waiting, I'm already in the process of assembling my sls printer. I want to complete it in a month. And then start refining it for slm. So far, the price can be kept at $ 1000 (with the exception of the galvanometer control board)
@@metalmattersdiscord is a good idea, no cost or investment other than self-moderating it or having community mods do it.
You could look at linear advance forthe bulging (happens with plastic too, although probably for different reasons).
Have you ever tried to hack into those 20-30W fiber laser engraving machines? They work using Galvanometer and do one-layer at a time scanning, just need to build a z axis descending table and work out a way to output each layer scanning control codes and integrate into the z axis movement. Those machines are available 2nd hand about $1000 only.
I actually bought a Nd:YAG laser from an earlier generation of engraver with the intention of exploring galvanometers. I don't think implementing a laser system is the difficult part. For me, the difficulty arose from trying to build a pressurized chamber that was both economical and safe. There are too many dangers around handling metal powders in my opinion.
@@metalmatters Does pressurized chamber necessary? I thought you just need a heated chamber and regulated inert gas inside. I'm looking forward to seeing both printer projects but personally really want to see the powder bed project working to some extent. By the way, your forum is down？
@@pengjinpan8293 To the best of my knowledge, yes. The most effective way to deal with the issues brought about by the plasma plume is to use a pressurized atmosphere.
I won't be working on the powder printer beyond this point for the reasons I've stated here. I took the forum down due to inactivity.
Any progress in the last 6 months? Your videos are awesome
@@tw60407 A lot has changed in regard to the mechanical aspects of the printer, and there is still yet more to do. It's stopped me from working on the process itself and therefore I've little to show for it in terms of output, hence the lack of videos. Things are moving along though. Thanks for the feedback.
Stainless steel needs a mixture in MIG MAG 75 %argon 25%CO2 in this case i think it would be easier in a chamber filled with argon
unreal work mate. Kia kaha!
Very good so far.
I feel that to contain the argon, you'll end up printing inside a can so the Ar doesn't diffuse away.
It will matter more as you get to higher layers.
What if instead, you approach it like welding body panels and run spot weld so the weld pool can cool? This way you dont run into the puddle?