17
$\begingroup$

I would like to print multiple parts continuously (non-interactively), so I can leave the printer alone for a longer time. So after finish, parts could be moved somehow out from the printing area, so the next can start.

Are there any methods of achieving that with standard desktop printers without having to use multiple printers?

$\endgroup$
4
  • 6
    $\begingroup$ 3D printing is not a very productive production technique for mass manufacturing (if not the least productive one). Creating several moulds from one master print is an example for a process that's simpler, faster, more stable and likely cheaper process. $\endgroup$ Commented Jan 12, 2016 at 22:05
  • 1
    $\begingroup$ Ultimaker once showed a video where they printed bracelets on an exhibition as give aways. They used the print head to push the finished print of of the build platform. It probably was just some custom G-Code,.. $\endgroup$ Commented Feb 23, 2016 at 18:18
  • $\begingroup$ What portion of the process are you asking about automation, there can be many steps which require interaction, including: loading filament, selecting model to print, preparation of printing bed, removal of completed prints, and fixing when broken, maintenance, etc. And secondly, what is the purpose of automating this step? Closer to 'mass manufacturing', increased output, decreased maintenance time? $\endgroup$ Commented Jun 13, 2016 at 22:47
  • $\begingroup$ You should look into something like Creality CR-30. Its build plate is a conveyor belt. The prints stick normally when on the flat section near the print head, but upon reaching the roller they peel off and can drop into a collection bin. A decent filament changer can be pre-loaded with 5, 8, 12 spools of filament and set in mode of "continue with next spool" once the previous one is exhausted. $\endgroup$
    – SF.
    Commented Mar 30, 2023 at 14:48

5 Answers 5

13
$\begingroup$

The only thing I can think of off hand is an old mod for the early MakerBot machines. It first was released for the Thing-O'-Matic I believe, but is compatible with Replicator 1 machines (and its knock-offs). Here's the Thingiverse page, but look up Automatic Build Plate.

Essentially, you can use the Replicator G slicing program and there is a setting for "ABP" or Automatic Build Plate. This will basically tell the ABP to run its routine after the controller receives the response that the printing program is done and roll the finished part off the edge of the build plate, then start the same program over again.

Drawbacks:

  • I don't think it's easily compatible with newer machines/slicers. But, it's open source
  • Pretty sure you have to use Replicator G, which is outdated now and may make your machine sound like it's going to fall apart (I know from experience)

Going off of @Pete's answer about solenoids. It reminded me that someone integrated a solenoid "ejector" (aka Boxing Glove) for their machine.

Update (06/08/2016):

Forgot to mention that if you choose to create your own "Boxing Glove" or conveyor belt, some software such as Octo-Pi and Repetier-Host allow plugins. So, you could interface with your hardware via customized code and integrate the functionality directly into the slicing application for the full closed loop operation.

$\endgroup$
1
  • $\begingroup$ The hardware design of the ABP is open source, but the method to use it for 3D printing was patented by Makerbot after they hired on the inventor (Charles Pax) so anyone who wants to use it (or any other conveyor-based build plate) needs to check local patent laws in their country. $\endgroup$ Commented Apr 12, 2016 at 18:04
6
$\begingroup$

An option that might be feasible for some situations (depending on your setup) is Sequential Printing - a feature provided by some slicing software, for instance Slic3r.

In short, this allows you to print multiple objects one at a time rather than simultaneously. This has some obvious benefits, but also some downsides:

Benefits:

  • Each object is finished individually, and you will therefore not have multiple half-finished prints when something goes wrong.
  • No particular printer or bed swapping mechanic is needed.

Downsides:

  • With most consumer printer, print area is somewhat limited, and sequential printing requires you to place objects so that the hot end can move freely between the finished objects, potentially restricting the effective usage of your print volume.
  • By printing objects one at a time, you limit the natural print cooling that happens when printing multiple objects at the same time. In particular for smaller prints, you might not want to print sequentially unless your print cooling solution is up for it.
  • Requires some setup

You specifically mention that you would like the finished prints to be moved outside the print area. In it self, sequential printing does not do this for you; however, if you fit your printer with a large, motorized print bed, you might achieve the same effect without moving into unfamiliar technology!

$\endgroup$
2
  • 1
    $\begingroup$ I like the "when something goes wrong" lol Because if you own a home 3D printer, you'll understand! $\endgroup$
    – tbm0115
    Commented Jan 14, 2016 at 15:17
  • $\begingroup$ @tbm0115, indeed! I normally remember to use sequential printing only after failing a multi-print job! $\endgroup$ Commented Jan 14, 2016 at 16:07
2
$\begingroup$

It would be possible and not too terribly difficult to rig a servo or solenoid with a push plate like a plow, on top of your build plate.

However this would require another processor via raspberry pi or arduino to control it. Scripting execution when the print has completed wouldn't be terribly difficult either, but it isn't readily available and certainly isn't part of a mass produced printer.

My issue with the concept is if you push a bunch of prints off the build plate, what guarantee do you have they wont become damaged in the process?

$\endgroup$
1
0
$\begingroup$

I've recently seen a video of this being done successfully with almost no hardware modifications to the printer.

Unfortunately I can't find the video itself any more, but the basic idea is to use the print head/gantry to knock the finished prints off the table, likely by adding some custom commands to the end of the sliced G-code. After the print is finished, the bed moves to the Y endstop (this was done on a bedslinger printer), the head is lowered to about half of the parts' height, then the bed is slowly moved so that the part crashes into the head, gets detached and knocked off the table, then the head is lowered even further and sweeps the part away.

The only mechanical modification, namely a sheet of paper stuck to the table's edge, ensures that the part will roll clear of the bed rails, and pushes the pile of finished parts further from the printer once it's off. In the case of that video there were three parts being printed at a time, so the motion was repeated for each of them, even though most of the time all three got successfully detached by the gantry.

This approach is likely far from universal, and probably only works with parts that are tall enough and have a small enough footprint to easily detach from the bed; I certainly wouldn't try it if the parts have really good bed adhesion and require more than a slight force to pop off. But it looks like if the requirements are satisfied it should be quite effective.

Also it's likely to be problematic on printers with box frames where it's the bed that moves in the Z direction, as the printed parts are likely to get trapped between the bed, print head and printer frame. CoreXY designs where the bed is static and the head moves in all three axes should be fine, however.

$\endgroup$
-1
$\begingroup$

I Don't really think that it is possible without hardware modifications, or maybe some small parts that will fit in the bed of the printer all on the same time

$\endgroup$
2
  • $\begingroup$ Dan - what type of modifications do you believe will best serve the 'automation' process? $\endgroup$ Commented Jun 13, 2016 at 22:50
  • $\begingroup$ For example long foldable bed $\endgroup$
    – Dan Boyko
    Commented Jul 1, 2016 at 23:59

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .