Hey! This thing is still a Work in Progress. Files, instructions, and other stuff might change!

StrongPrint - The DIY Metal 3D Printer

by kolergy, published

StrongPrint - The DIY Metal 3D Printer by kolergy May 15, 2014
0 Share
Download All Files

Thing Apps Enabled

Order This Printed View All Apps

Thing Info

13075Views 5387Downloads Found in 3D Printers
Report Thing


Design & Built with the support of the Artilect Fablab in Toulouse France

Affordable DIY metal 3d Printer based on TIG process

For more details in English: http://reprap.org/wiki/StrongPrint
Pour plus de details en Francais: http://reprap.org/wiki/StrongPrint/fr

Article on 3Dprint.com: http://3dprint.com/4787/3d-metal-printer-600-euros/
For the french speakers there is an article on 3D Natives:


It consists of the Enlarged Rostock modified with a magnetic effector & coupes to a cheap 160Amps inverter MMA power source.

Spool Holder available from Iron_Momo: http://www.thingiverse.com/thing:335662/#files

Wire feed was one of the most difficult item to get right (and at the time being is not fully stabilized). What seems to work the best is to rapidly extrude & retract the metal wire to make sure it only melt where you want it to melt

Wire is 0.6mm steel. for MIG.

This machine is still very experimental & will evolve.

Setup & test phase activities:

  • Done: Arc ignition, process to reliably ignite the arc with the lift arc technique
  • Done: Arc Power & speed control to identify adequate settings to perform surface fusion of the baseplate
  • Done: initial metal deposition identify wire nozzle angle & position
  • Done: initial metal deposition identify speed wire feed quantity.
  • Done: Initial metal deposition identify preferencial directions
  • Done: initial few Layers stack-up test
  • On Going: identify adequate speed & power profile for layers stack-up
  • On Going: manage layer height
  • On Going: large number of layers stack-up test

Printer design & make:

  • Working: Printer movement mechanism, cinematics from Enlarged Rostock with added magnetic bearings. (a slightly larger magnet size could be beneficial)
  • Working: Fixed torch & nozzle holder
  • Working: Wire feed motor (Extruder)
  • Working: Wire nozzle holder & positioning device
  • Working: Tig torch modification
  • Working: Power source coupled with TIG torch
  • Working: Shielding of all wirerings to prevent electromagnetic interferences
  • Working: Wire roll holder including brake & wire damping mechanism
  • Working: Shilding of the control board
  • Working: Implemented Ignition GCode command & automatic lift arc ignition
  • Working: New 160A power source 80A was clearly to weak
  • Working: Arduino controlled Solenoid valve for argon
  • To Do: Arduino controlled power source
  • To Do: HF ignition system

Design Details to be provided on http://reprap.org/wiki/StrongPrint
(Page still in construction)


The Strong print Printer share the same Frame structure as the Enlarged Rostock

Assembly of the machine is as simple as an IKEA furniture:

For the frame:
1 - Build the three columns:

  • Assemble the Iddler End: http://www.thingiverse.com/thing:143186 with the beraing & the bolts
  • Assemble the Motor End: http://www.thingiverse.com/thing:71687 with the stepper motor & the pulley
  • Assemble the Magnetic Carriage: http://www.thingiverse.com/thing:303637 with two R-15-06-06-N magnets, insert two KB-12-WW Linear Bearings in the housings & tighten the M3 bolts
  • Assemble the Column:

    • Fit the two Z-12-1000A smooth rods on the Motor End
    • Insert the Magnetic Carriage with the magnets facing towards the motor side of the motorend (fix them with Epoxi not hot glue). The linear bearings shood fit smoothly on the roods with no play.
    • Fit the Iddler End with the bearing on the same side as the puley on the Motor End.
    • Ensure everithing is straight & square with no twist & tighten the M3 bolts of the Iddler end & the motor end
    • Fit the T2.5 belt to the tension nut from the carriage, Attach the tehsion nut to the carriage with two long M3 bolts & nuts. pass the belt arount the puley & the bearing and fix the end to the adapted bit on the carriage sith a slight tension, which can then be ajusted with the long M3 nuts.
    • That is it you now have three columns fully equiped.

    2 - Frame Structure

  • Bolt the reinforcment bar to the printer Bed & Top: http://www.thingiverse.com/thing:74119
  • Screw one by one the MotorEnd of the Three columns to the printer Bed
  • Screw the printer top on top of the three columns
  • Fit the tension cable to the reinforcment bars while crossing them (the tension cable are optional it rigidifies the structure which is only needed for fast printing with plastic)
  • Apply moderate tension to the tension cable ensuring no twist or deformation is created in the frame .
  • The structure is now fully assembled

More from 3D Printers

view more

File Name



All Apps

3D Print your file with 3D Hubs, the world’s largest online marketplace for 3D printing services.

App Info Launch App

This App connects Thingiverse with Makeprintable, a cloud-based mesh repair service that analyzes, validates and repairs most common mesh errors that can occur when preparing a 3D design file for p...

App Info Launch App

Kiri:Moto is an integrated cloud-based slicer and tool-path generator for 3D Printing, CAM / CNC and Laser cutting. *** 3D printing mode provides model slicing and GCode output using built-in...

App Info Launch App
KiriMoto Thing App

With 3D Slash, you can edit 3d models like a stonecutter. A unique interface: as fun as a building game! The perfect tool for non-designers and children to create in 3D.

App Info Launch App

lol a mig welder printer lol,,,,,, You know a mig has to follow at an angle to make a solid weld. straight down is gonna look like a bird shit all over your project. You would need to alter your firmware to follow your weld set on at least a 15degree or something. I found this kinda funny a person actually tried it I give you props for that. The first 3d metal printers actually used a Tig welder. They always had a cool down period and flow problem as the layers stacked up higher and higher. Or warped from a hot layer on a already hot layer*

This project use a TIG welder and not a MIG.

Oct 11, 2015 - Modified Oct 11, 2015
LittleCareBear - in reply to kolergy

Oh dear........ You do realize a Mig uses a spoil of wire........ Tig uses rods or electrodes with a gas such as argon or nitrogen, right? Also that is a combo cheap china unit that can be either or. But it clearly has a MIG* tip in it Tig tip has a Ceramic plasma cup on end of it....... But hey what would i know only been welding for 25 years with cnc units......... Here is a link so you don't look so silly : )


Please don't take it personal but, yeah..............

Please visit this site http://reprap.org/wiki/StrongPrint where you will have a lot more pictures & explanation.

Basically it is a TIG torch feed with Argon, powered by a cheap MMA inverter based power source. The wire is fed by a nozzle on the side of the torch you have some close up pictures in the link.

Hi! I like what you're doing here. Just so I know before I get started, is this currently a project that can function once completed? I noticed that you have it listed as still a work in progress, so I just wanted to see where you're at with it and go from there. Brilliant work here.

Up to now I arrived to the point that I can lay-up welds of any shape on the metal but getting up is tricky speed needs to be adapted for each layer... And I do not have to much time to work on it.

Hi Kolgery-

I have all of my parts 3D printed as outlines in the BOM, and am ready to assemble. I am currently familiar with repetier firmware with a Rambo board. Silly question perhaps, but can your Modified Marlin from Jschroll be ported or adapted to Repetier, or would I be better off going with Marlin?

Thank you,


Apr 22, 2015 - Modified Apr 22, 2015
kolergy - in reply to revz

I'm using repeater with no problem it is fully compatible.

I am still using the V0.56 of Repetier because i'm on Mac, I have not tryed the newer version with the cura engine, but i think the new version should be as well compatible with the delta. For the slicer, I have a tendency to prefer Slic3r to the cura engine as it provide more detailed settings.

Mar 12, 2015 - Modified Mar 12, 2015

I have thought about doing this as well. Have you considered that a different angle or push/pull weld might be needed or is this working correctly?

Mar 12, 2015 - Modified Mar 12, 2015
kolergy - in reply to hippyengineer

Ies I have been considering, however it would require one more axis and break the compatibility with the current firmware and G-Code. For the moment it works acceptably well either directions for the low layers. my current difficulty is for the higher layers to find adequate speed & power profiles

Is it possible to drop the firmware? I galley with moving motors.



I have used the Modified Marlin from Jschroll available here: https://github.com/jcrocholl/Marlin with the modifications available here http://www.thingiverse.com/thing:399434/#files

Marlin Update for Strong Print
by kolergy

Hello, can give us yourselves more information onto the length of carbons rods. Opencad gives 432mm. Is it good of axis with axis of 2 balls? What are the parameters to be gone into the Marlin firmware?
In any case, very beautiful realization, congratulations.


For the length I need to check the geometry of what is built. The ones that are in the machine are voluntarily cut too long because at the beginning I did not want the chariots to go down to much in case of weld splattering, however TIG welding is very clean therefore the carbon rods can be cut at the optimal length to give maximum build height, and I have not calculated the optimal length given the new geometry (too long is ok you just loose some max Z but not the opposite).

Yust a comment:

It is not really nessecary to use carbon rods. I have build similar machine with Aluminium tubing form local hardware store. They are strong enough and even lighter and cheaper than cabron rods.


This comment has been deleted.

I am very interested in this project and others like it. Both these guys : https://www.youtube.com/watch?v=_Dy-2F81DWA and Molten 3D: http://molten3d.blogspot.com are using cartesian machines for welder based metal printing; however StrongPrint is the only Delta based project I've seen so far that is implementing a TIG welder. I would like to know how I can become involved in this. Though I am fairly new to 3D printing, I am very familiar with the TIG welding process, and I think I have some insights that could help the StrongPrint project. Perhaps I will build one in my metal shop. What is a basic estimate of the time needed to build this assuming good mechanical aptitude and one Delta Robot build already under my belt?

Feb 1, 2015 - Modified Feb 1, 2015
kolergy - in reply to revz

Thanks for your interest and the links the molten3D project is very interesting. The rationale for the Delta machine is first because I had one available (From the Enlarged Rostock) but there are as well some technical advantages:

  • The structure is mechanically cleaner no stack-up of axis and therefore of errors
  • It is easy to make large delta printer
  • It is easy to assemble & set-up
  • Most importantly, the use of magnetic joints allows to tolerate errors (dismantles in case of inadvertent contact or electrode stick)

To make one it is not too difficult everything is either standard material or 3D printed (except the top & bottom plate which are in plywood). You can find the BOM & some assembly instructions on: http://reprap.org/wiki/StrongPrint but take care for the moment I have not yet been able to make a good part (I think I'm not to far but I do not have enough time to work on it)
The main issue is to identify good rules for the evolution of speed & weld parameters as wall height are increased. I manage to get the 4 first layers perfectly with thickness less than 2mm & 0.75mm height per pass but above 5 layers I need to accelerate more or reduce power. in parallel I'm designing an Arduino based device to measure precisely the welder output & set the power to have more flexibility.

revz - in reply to kolergy

Very interesting. The choice of a Delta design makes perfect sense. I just built a Rostock Max v2, and and am interested in making another purpose built machine. I understand that all the 3dp parts are ABS, what % infill are they? layer height?

kolergy - in reply to revz

Yes it is better to do the part in ABS specially the ones close to the printing head. Layer height is 0.3mm for the large frame parts & 0.2mm for the small & moving ones 30% infill 3 solid layers on top, bottom & perimeters. The magnectic carriage is the only delicate part for warping you need to have very good thermal conditions.

Instead of using the wire feed, why not use powdered metal? I know the wire is convenient, you could greatly reduce the weld size and the needed voltage output by using a micro-TIG torch and spray deposition similar to the sintering laser deposition systems out there. To spray the metal, a sandblasting unit could be used with the appropriate fittings to spray toward the arc. I think it would 1: reduce the roundess of the weld (due to no drop formation), and 2: significantly reduce the needed voltage by using a micro torch. I am planning on trying this, but I wanted to share the idea to see if you have any thoughts.

If you are set on wire feed, why not use micro wire? There are micro welding system that lays down very small welds: http://www.fivestartoolwelding.com/Images/micro2-2.jpg

Great project though!

The aim is to be able to build the machine for less than 1000€ which limit to affordable components. however this sounds interesting, for the powder the question is how to get hold of it and if it is safe to handle?
For the moment i'm using the smallest welding wire I can find (0.6mm).

After reading the powerpoint below on WAAM, I see that the deposition process is good for detailed parts and is slower. I did like the idea of a hybrid system that they touched on. It would allow the flexibility of small detailed parts, and the fast buildup of larger parts. I also like that the Hive system they spoke of also has machining capabilities. So many possibilities.

ESD is better than TIG for 3dprint.TIG has too much weakness.

Jan 22, 2015 - Modified Jan 22, 2015
tbaggins - in reply to kolergy

great link! Also, ESD is electron spray deposition. It uses a focused electron beam rather than a laser to sinter metal parts using powder deposition. NASA is using it right now instead of the laser technology.

Interesting. Please explain your reasoning. TIG is, or can be, as strong as the patent metal.

Does this print in a circle range or can it print in a square because I am looking for a headed bed and I was also wondering, if it does print with a circular heated be (for plastic), what the diameter of the circle is. Thanks.

Jan 10, 2015 - Modified Jan 10, 2015
kolergy - in reply to RobotBuilder

The print area is not as per say circular, but a circular bed is the best approximation.

If you want more insight you can look there: https://drive.google.com/viewerng/viewer?a=v&pid=forums&srcid=MTgyNjQwODAyMDkxNzQxMTUwNzIBMDc2NTg4NjQ0MjUxMTE1ODY5OTkBdmZiejRRR2phZjhKATAuMQEBdjI

Around page 15 you should see the print area as well as the expected precision in the different axes

Jan 10, 2015 - Modified Jan 10, 2015
RobotBuilder - in reply to kolergy

Thanks again. I am starting it very soon but I am using metal brackets instead of plastic for the linear rod mounts.

Good idea

Thanks, I am using a lost wax method.

It says on the wiki to use an Arduino Due but I heard that an Arduino Mega is best to use with a RAMPS board. Which one should I buy? Thanks.

And what is the diameter of the carbon fiber rod?

8mm these are Kite rods can be found for cheep in Decathlon

Ok thanks and does it print in a circular pattern?

Hey, I was wondering if I built this model of printer weather or not I could use the plastic print head from the Enlarged Rostock. I was also wondering if this was any different than the enlarged Rostock or weather the head is just different. Could you also include how many of each part you need to print and which ones? This is my first Delta and I would like to not mess it up. Thanks, I am working on it with my dad so any recommendations help.

Dec 30, 2014 - Modified Dec 30, 2014
kolergy - in reply to RobotBuilder

All the bits of this printer except the head obviousely are made do be compatible with a standard palastic printer. The movment systems with the magnets is a big improvment from the enlarged rostock it is safer for the machine if a colision happens it just dismantles itself & do not break anything. and it is much more precise. the only thing to be carefull is not to heat the magnets ie: no hot glue. The printhead for the enlarged rostock dose not ude the magnet system and is therefore not compatible.
The best thing is to modify the "Magnetic_Platform_Tool-V0.3.scad" file to fit your head, it is something I plan to do but I had not the time to do it yet.
The full Bill Of Material including the number of parts need to be printed is available on http://reprap.org/wiki/StrongPrint.

  • To have a set-up to print plastic is not to complicated and can be handled as any other printer.
  • However I do not advise to go for the metal set-up unless you know what you are doing especially concerning the Safety of welding, it is still very experimental & did not yet produced a part, I think i'm not to faf but still not yet there.

P.S.: I realised thet the file "Magnetic_platform_base_V0.2.scad" was missing, it is required t modify "Magnetic_Platform_Tool-V0.3.scad" it has been added.

Also what .scad program do you use? I am familiar with .stl however I do not have a program for scad files and I would like to find a nice one.

I use OpenScad http://www.openscad.org/downloads.html make sure you use the latest development snapshot as it has been improving a lot recently.

It is very simple & editor based so it is full parametric therefore once you have a design it easy to modify it.

Thank you so much. I can not wait to start and this will help my dad and I a lot with the printer. This looked like a great printer and great job with the improvements.

I do insist that arc welding is an activity that requires very serious safety measures for the protection against UV & light and to prevent the risk of fire. Unless you & your dad are familiar with arc welding I would encourage you to have proper training in arc welding before you engage in this project.

In addition I remind that it is for the moment very experimental and as of today I have not yet managed to produce a functional metal object.

Thank a lot. My dad has experience with arc welding and we have an arc welder at my house. I am planing on starting with plastic printing. Is there any way I could purchase the printed parts from any one? I am very anxious to start on this and those are the only pieces I can not find. I would like to thank you kolergy again for all the help you are providing for me.

Depending of where you are there are fablabs, clubs or shops who have 3D printers where you can go and print your parts.

Thank you for telling me about this. If I have any more questions I will come here first, Thanks again.

This looks just like an idea I had! Great work!

beautiful ! congratulations.
I am very glad that your design is made in Toulouse ;-)

Génial !
Est-ce finalement le procédé TIG qui est utilisé, ou MIG ? Si j'ai bien compris, dans le premier cas, l'électrode et le métal d'apport sont distincts, alors que c'est le même dans le second cas... Du coup, le MIG semble plus simple à mettre en oeuvre, non ? Sauf peut-être en ce qui concerne la gestion des arcs...

kolergy - in reply to fma

As it is an English website I will rephrase in english for the non french speakers: Is it a TIG or a MIG process?
My answer: It is a TIG process it is more complex to master but it has the potential to generate finer details less than 2mm as the MIG in my opinion will not be able to go below 5mm wells size

fma - in reply to kolergy

Thanks! I was not aware about TIG vs MIG resolution... And what about speed?

kolergy - in reply to fma

For the moment it is very slow but in theory speed will depend on th power available