Which machines for an Open Source FabLab?


(Diderik van Wingerden) #1

Hi fellow-Fabbers!

I am in the process of making machines for an Open Source FabLab/Makerspace.

Which Open Source Designs do you believe are best to use?

Overall requirements:

  • Standard components commonly available, sourceable locally in a normal hardware store, electronics shop, home improvement store (as much as reasonably possible). So no kits.
  • Robust, dependable, proven designs/machines (safe to use and somewhat idiot proof)
  • Low in expected maintenance/calibration/repair work
  • Good price/quality ratio (“perfect is the enemy of good”)
  • Good documentation: for the build, use, troubleshooting, etc.
  • An established and active community (no abandoned projects)

My ideas at the moment:

3D Printer

Design: Prusa i3 Hephestos

With a frame made from standard aluminium profile available at home improvement store. I have made one of this in a workshop in 2016 and it has been great. As I am no expert, it would be a good start for a first printer for the FabLab as I can peek over to my existing printer.

Once I complete the Hephestos and get more experience and confidence, what would be a good option for a second 3D Printer? With heated plate perhaps?

CNC Mill

I found these designs using my favourite online search engine:



http://goodenoughcnc.eu/hybrid-cnc/

Any you would recommend or advise against? Any other design that would be better?

The size should be large enough to put a ‘standard size’ large sheet of plywood: 244x122 cm, so we can also make furniture with it.

Lasercutter
Design: Lasersaur

After an online search this design looks like the only existing mature/proven model out there.

Or are there any other options that would fit the requirements?

The main goal of the FabLab is to make functional things that can be used in everyday life.

Would you recommend any other machine(s) besides these 3 to build?

Finally, a question maybe to put in a seperate topic as well:

Are there people in the BeNeLux area, including western Germany perhaps, who offer workshops on building Open Source 3D Printers, CNC Mills or Lasercutters?

It would be great to get up to speed following such a workshop, instead of figuring it all out using online resources only.

Be great!
Diderik


(Peter Troxler) #2

Check out the http://fablab.hochschule-rhein-waal.de/laserduo/?lang=en (and that particular lab)


(Diderik van Wingerden) #3

Very interesting project, thanks for the link! A huge machine by the way. And a huge FabLab!


(Stefan Strappler) #4

Hi Diderik,
we are running a FabLab in Innsbruck/Austria with many open source machines since almost 3 years.
3D printers: 2 x delta rostock printers from a company which is run by friends http://construction-zone.de/ and one ultimaker original +. The delta printers are in use quite a lot and there is tons of online ressources in case anything goes wront
Lasercutter: We have one Lasersaur with a 100W tube and we are quite happy with it. We had some smaller and bigger problems with the machine, but once you are in the lasersaur google groups community, you get help from wonderful and very motivated people all over the world. I can really suggest Lasersaur! You can have a look at the google group, sometimes people sell their Lasersaur.
We also have a small China Lasercutter, which is used for engraving. They are quite cheap, but our machine works really well…we are confident if anything breaks or doesnt work anymore, then we can take our knowledge from building the lasersaur and fix the china laser.
We had to replace the tubes from both machines once.
CNC Mill: We also startet with a shapeoko mill (it was an older version like the one that you plan to get). The mill worked well for soft and/or thin materials, but it was not stable enough for hard materials. We designed and built our own machine, which is quite a beast. It runs on arduino with a very useful and handy software https://www.estlcam.de/
Check out our website to see some pictures of the machines: http://fablab.spielraumfueralle.at/reservierung/

If you come to Austria in winter, you are very welcome to visit our lab! :smiley:
Best
Stefan


(Diderik van Wingerden) #5

Hi Stefan,

Thank you for all the information!

A few follow-up questions:

  • Delta Rostock Printer: do you have a ‘home page’ link to the printers you are using, with information in English? I am guessing you bought them completely assembled from Construction Zone, so perhaps I should ask them.
  • Lasersaur: AFAIK this machine is not available as kit, so everyone who builds it, sources generic parts locally and gets the specialised parts from the Lasersaur store. Is that correct? Did you document/share learnings of the build you did?
  • CNC Mill: did you create documentation of the machine you built yourself? And/or is it based on an Open Source CNC Mill from which you got started?

Thanks for the invitation to come to Innsbruck! We are not planning on a skiing trip this winter, but coming over is a nice idea any time of year. Let’s see what the universe will come up with :wink:


(Stefan Strappler) #6

Hi Diderik,
Printers: We built them in a kind of workshop together with our mates from Construction Zone. You can buy them as a kit to build yourself or buy an assembled one i think.
Lasersaur: Yep thats correct, we bought lots of parts from Misumi. When we ordered the parts, we could download a Excel file from the Lasersaur Website and upload it directly at misumi. Some other parts we bought from different suppliers, but you get good help from Lasersaur website or the google groups community. We did not document the building process, but we have posted some questions in the google groups community. There is a perfect 3D file for the Rhinoceros 3D Software with all parts organised in layers to build the lasersaur step by step. Kinda felt like building a lego spaceship with the perfect instruction :smiley:
CNC Mill: One of our team drew the machine completely from scratch. He has worked in different companies with lots of CNC machines and is quite a genius when it comes to milling. So there is a perfect 3D file for the CNC mill as well. Couple of months ago we built a smaller machine for a friend of us, who uses it to mill for architectural models. Our mill has a work area 90x90x12cm, the smaller brother was like 40x60x10 as far as i remember.
Best
Stefan


(Diderik van Wingerden) #7

Hi Stefan,

“Printers: We built them in a kind of workshop together”

Yes, that is a really good way to learn and build. I did that with my first 3D Printer and would love if it would be possible for other machines as well.

Lasersaur: [etc]

All sounds fantastic!

CNC Mill: One of our team drew the machine completely from scratch […] a perfect 3D file

Can you send a link to the designs and documentation you created?

Be great!
Diderik


(Nicolas de Barquin) #8

Hi Diderik,
I’m in Belgium and I mainly have opensource machine also;

After 5 printers, the next will be to build from scratch the prusa i3, the last one.
I tried tevo tarantula, anet A6, anet E10, m3D, one geeeetech, the 3drag and I started with the thing=O=matic from makerbot.
They all have some con and pro, seems to me today that the original prusa i3 solve a lot of problems from clones.
I had some experience with the sigma from bcn3d, fabrikator v2 fro mhobbyking, ultimaker of course and the replicator 2x at some point.
Prusa I3 is still something I have very good review around me. I want to try.

I built also the lasersaur, I use it since 3 or 4 years now.
Again, today, I’would go for something different or design mine (at some point) lasersaur is overkill.
And components from misumi is quite expensive.
Also, I redesign some parts to be sure tube is well align with the alu porfile 4080 https://www.thingiverse.com/thing:1319254
I replaced the tube, then had problem with electromagnetics interference from the tube with the beagle bone.
And I’m not satisfied with the guidance system, even Vslot rail and wheel should work better than just bare bearing on aluminium.
And, yup, the last thing I want to change is the software. Lasaurapp had some nice features like manual entry of coordinates and tag system, the new driverapp skipped these but add raster which is good to have.
I want to flash everything and use grbl + laserweb which have so much more features. (but don’t have time to shutdown the machine for these kind of experimental core modification)

Then, for CNC. I have a bzt PFE 1010 from germany + linuxCNC

http://www.bzt-cnc.de/index.php?category_id=25&page=shop.product_details&product_id=333&Itemid=97&option=com_virtuemart&vmcchk=1&Itemid=97
I would advice you to buy a machine like that without the controller (quite expensive for nothing, big box of empty with classic chinese controller inside) which is wired to run with proprietary software like mach3 or worst wincnc.
The machine is very sturdy, in steel and not so expensive compared with all the work you need and the heavy lifting to build your own. Aluminium like shapeoko have a big disadvantage of being to soft. So you have to slow down your operations and set smaller depth per pass.
Then I could go for the smoothieboard with grbl as mainboard + 3 drivers leadshine or something, I think you can save 1000€. I remember having bought the bzt controler someting like 1500€. And had to rewire limit switch to use with linuxcnc.

Any way, I started all of these 7years ago with thingomatic and the cnc, then the rest came later. One ofter one.
that’s quite exiting to think about starting with everything from scratch. Quite a challenge.
have fun.


(Diderik van Wingerden) #9

Hi Nicolas,

Thanks for all the information!

About a 3D Printer:

the next will be to build from scratch the prusa i3, the last one

Do you mean the Original Prusa i3 MK3?
If yes, do you mean to buy that kit and then build it?
If no, then which printer?

About a lasercutter:

Again, today, I’would go for something different or design mine (at some point) lasersaur is overkill.

Do you have any Open Source Design in mind? To me it looks like the Lasersaur is the only mature, well-supported machine with an active community currently available. For me personally I would not be able to design such a machine myself.

About a CNC:

I have a bzt PFE 1010 from germany […]
Aluminium like shapeoko have a big disadvantage of being to soft.

Yes, I am starting to understand that steel is preferred over aluminium if one wants to do heavier work. A requirement for us would be to at least be able to ‘2.5D mill’ large plywood of around 18mm thick, to be able to cut furniture (think Open Desk / AtFAB) and WikiHouse parts.

Is the BZT an Open Source machine? I do not see that kind of documentation on the BZT website (but the website is mostly in German and my German is not that great).

Another point to consider for us: creating a steel frame would require welding (I guess), so that means we would need to get the equipment for that plus learn how to weld before we can build the machine.

that’s quite exiting to think about starting with everything from scratch. Quite a challenge.

Well, yes. It is quite challenging. Our goal is to advance the practice of creating functional medium-complex machines/objects that can be used in every day life, using computer controlled machines. And not only in WEIRD (*) countries, but also in countries where people do not have large sums of money available to buy complete and ready expensive machines, but need to also make these machines from locally and easily sourcable components.

(*) WEIRD: Western, Educated, Industrialized, Rich and Democratic

Be great!


(Nicolas de Barquin) #10

Do you mean the Original Prusa i3 MK3?

yes, the original, from their github https://github.com/prusa3d/Original-Prusa-i3
But not from a kit. We have here actually a lot of parts (motor, rails, …) and the tools to make the rest.

Lasersaur, Do you have any Open Source Design in mind?
Nope, you are right. but from experience, there is room for optimization. And of course, for that you need better understanding of the machine.
I made a post when I changed the lasertube. https://nicolasdb.github.io/post/lasertech1/

It still a good tool and it’s a very good start but… like for the CNC, we are not equipped enough to have the same capability of production and precision as a company.
It could be cool to just buy the hardware and install your opensource electronics, which I think is way easier to use because it’s open. You have better control of it, you have communities to help and it’s less expensive.

CNC
no, bzt is not open. It was even difficult to receive something simple as the pin out of the serial port.
But it’s super solid. Every thing is screwed so no need to solder anything, they sell no kit so everything is already assembled. You only need allen keys :slight_smile:
And they said to me that they can even bought you back the frame when you want to upgrade. It’s heavy steel and they just give a new coat of paint, new rails and it’s like new.
You cannot do that without heavy machinery.
Only problem, same everywhere actually, as soon as you want to have a big CNC for MP pannel 120x240, it’s even as soon as you want to have a larger size than 100cm and go for 120cm, it fell in the professional category and price triple. So bigger machine become expensive. not only in BZT.

Another choice from friends working with these big cnc, is too buy chinese frame and add EU opensource electronics.

need to also make these machines from locally and easily sourcable components.

And yeah, exactly so it’s better to start from these and then learn how to adapt them for other country.

see you


(Diderik van Wingerden) #11

Thanks for the follow-up info! We will digest it for our plans.

Alright, and then the MK3 probably.

Very interesting project! I had this in mind as a next one after getting the experience/confidence from building the Hephestos. I am curious how you are going to build an alternative for the MK3 specific parts like the frame. Please keep us informed and perhaps share your learnings, if you can.

Be great!


(Nicolas de Barquin) #12

Yes, mk3 of course, and I want to test their last multimaterial system.

I will use Trespa for material and the cnc to cut it.
image

Hephestos is sexy but last time I checked, there is no heated bed which is useful to use different materials.


(Diderik van Wingerden) #13

Trespa + CNC for the frame sounds like a great idea. Do you have previous experience of using Trespa for a frame? Which thickness?

For my 1st Hephestos we cut a standard aliminium corner profile and used 3D printed anchors to connect. From experience seems to be pretty sturdy.

Indeed, does not have a heated bed, so limited to PLA.