Large Format 3D Printer Is A Serious Engineering Challenge

When you want to build a large format 3D printer, you can’t just scale up the design of a desktop machine. In an excellent four-part build series (videos after the break), [Dr. D-Flo] takes us through all the engineering challenges he had to contend with when building a 3D printer with a 4x4x4 ft (1.2 m cube) print volume.

For such a large print volume you won’t be printing with a 0.4 mm nozzle. The heart of the printer is a commercial Massive Dimension MDPH2 pellet extruder, capable of extruding ~1 kg of plastic per hour through 1.5 mm to 5 mm nozzles. To feed the extruder, [Dr. D-Flo] used a Venturi vacuum system to periodically suck pellets from a large hopper. The system is driven by compressed air, which can introduce moisture back into the carefully dried pellets. To reduce the humidity levels, the compressed air passes through a series of vertical aluminum tubes to allow moisture to condense and drain out the bottom.

At 8.4 kg, it needs a powerful motion platform to move it. [Dr. D-Flo] went with a stationary bed design, with the extruder pushed around by seven high torque NEMA23 motors on a large gantry built from C-beam aluminum extrusions. A machine this size needs to be very rigid with well-fitting parts, so [Dr. D-Flo] made heavy use of CNC machined aluminum parts.

To allow dynamic bed leveling, [Dr. D-Flow] made use of a Quad Gantry Leveling (GQL) scheme. This means that each of the four Z-actuators will dynamically adjust its position based on inputs from the leveling probe. The avoid stressing the corner mountings that hold the X-Y gantry to the Z carriage plates, he used radial spherical bearings at the mounting points to allow a few degrees of play.

The build plate consists of an aluminum plate bolted onto the base in 25 positions with springs for adjustability. A massive 6000 watt 220 V heating pad sticks to the bottom, while the actual printing surface is a large sheet of borosilicate glass. One major concern was the deflection of the build plate when heated to working temperature, but with all the adjustment options [Dr. D-Flo] was able to get height variation down to about 0.25 mm. This is within the acceptable range when printing with layer heights of 1 mm or more.

We’ve featured large scale 3D printers in the past, but none are quite as big the University of Maine’s building-sized 3D printer that can print a motorboat in one piece.

14 thoughts on “Large Format 3D Printer Is A Serious Engineering Challenge

    1. $5300.

      Break even is a lot a filament. Not going to get much cheaper if smaller.

      Suggest hacking one up from the barrel of a worn out tiny Injection molding machine. But ‘worn out’ plus possible hydraulic system requirements.

  1. To me it pretty much looks like a scaled up standard 3D printer, except from the pellet extruder, but I’m not going to sit through 2 hours of video

    One big disadvantage of big 3D printers is their size.
    You can’t stow them in a small area, and they are also heavy and you have to pay for all that weight.

    If I was interested in a bigger 3D printer, I would make it in the form of a SCARA arm. Pretty similar to the “3dpotter” potting clay scara arm, but that thing is ridiculously over priced. The materials for building a SCARA arm of that class would be around EUR2000 and for the other 92% of your money you can put in quite some hours to build something nice with those materials.

    1. To get good warp free prints at scale though your scara arm would still need to be inside a insulated tent/box. So its still huge – quite possibly larger depending on which metric you use to compare the size and what the required build volume actually is!

      There is nothing hugely novel about this design, its just a well thought out and engineered (from what you can see) very large format printer that can print from pellets. But the same sort of arguments can be made about any 3 axis cnc platform, they tend to be very very similar as there are not many ways to get that result, and when the work is definitely being done 2.5D some of the other 3 axis options are ruled out as unsuitable/overkill anyway..

    2. There are no SCARA robots with a 1.2 meter vertical stroke.

      If you used a SCARA or conventional 4-6 axis robot, the cost would be much higher than what is seen in this project.

      If you get into hobby-class SCARA like Dobot, you might get the costs down, but what about the rigidity?

  2. Seems like the best approach should be for the hot end to dynamically ascertain its position using laser measurements rather than engineering perfectly rigid frames, etc. This should allow good prints even in the face of wobbly frames and poor calibrations. Feedback can compensate for a multitude of engineering imperfections!

  3. Except the guy on YouTube that prints plastic tanks just scaled up a desktop printer. That’s basically what Dr.DFlo did too, the only thing not scaled up from desktop is the pellets instead of filament. I’m not even sure why he did that, except maybe the novelty of it, considering you can get some huge rolls 1.75 or 3mm filament. I just really don’t see how that isn’t a scaled up desktop printer 🤷

    1. If you struggle with such easy engineering things, don’t step into 3D printing world.
      I’ve been engineering things for the past 30 years, this topic makes me laugh to death.

Leave a Reply to paulvdhCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.