1

Hi makers,

I’m a student researcher at PSU and I spend a lot of time working on 3D printing. I’ve found that it can be quite a hassle to constantly trek back to my lab to remove my completed object and start my next print every 3-4 hours when I’m trying to print a lot of things out. Does anyone else have a problem that could be solved by an “automated” 3D printer? If so, has anyone tried to tackle this problem on their own yet?

We’re trying to tackle some of the problems that the 3D printing community faces, so feel free to post any of your other pain points here as well!

4 Likes
2

@Pot8oSH3D @obelix @BeachLab is this something you’ve tried before?

3

Hey Kevin. There have been a number of attempts to automate aspects of the process. Most of which are in the public domain but it’s usually far simpler and quicker to buy a second printer. Usually a larger one.

May I ask, will you be putting the findings that result from your research into the public domain and making your designs open-source?

Cheers,

AndyL

4

Even though it would be fairly simply to automate a 3D printer to place finished prints in a bin, the main problem is that the process still needs a lot of manual intervention: filament stuck, grinding…

Also it could only be made in a printer like the Ultimaker2, where there is no need to put lacquer in the buildplate. If you need lacquer, things get more complex

1 Like
5

I thought about it for some time and found out that it would lead to many other problems. I was always able to fill the buildplate so that it takes long enough. As @Pot8oSH3D said i think it’s easier and cheaper to buy a second 3d printer, like an Ultimaker 1+.

I had to print out over 100 scanned people, so i filled up the buildplate with 10-12 figures, leading to 200MB GCODE and one or two days of print time.

@BeachLab why would you put lacquer on the buildplate? I get perfect prints with 3m blue painters tape or kapton… If it works for you… :slight_smile:

6

Thanks for replying Andy! Unfortunately as a student researcher, my institution owns basically all of my IP in return for paying for my schooling. However, I am determined to get as much of what we come up with out into the community as possible.

As far as the attempts made in the public domain, do you think you could guide me to any of these? I have yet to find anything that seems reliable/successful

7

Thanks for replying Andy! Unfortunately as a student researcher, my institution owns basically all of my IP in return for paying for my schooling. However, I am determined to get as much of what we come up with out into the community as possible.

As far as the attempts made in the public domain, do you think you could guide me to any of these? I have yet to find anything that seems reliable/successful

8

I was thinking that if you had one printer that worked reliably, by having a device or something that essentially removed a completed print for you and started the next one, you could end up printing all day and all night without loosing any efficiency by having to go to your printer late at night (or early in the morning!) to start the next print, as it would just keep chugging away until you told it to stop or it ran out of files.

Any thoughts? Or am I just being lazy?

9

I was thinking that if you had one printer that worked reliably, by having a device or something that essentially removed a completed print for you and started the next one, you could end up printing all day and all night without loosing any efficiency by having to go to your printer late at night (or early in the morning!) to start the next print, as it would just keep chugging away until you told it to stop or it ran out of files.

Any thoughts? Or am I just being lazy?

10

I tried something like this but the problem was to find the right amount of adhesion on the buildplate. Not too much so it can be removed easily and enough so that the print doesn’t fall off during printing.

11

Interesting, good point. Let’s pretend for a moment though that it’s not an issue or could be solved… Do you think there might be a need for a device or system that could do this? If, say, it could be adapted to work in any traditional kind of FDM machine like ultimakers, rep raps, makerbots etc.

1 Like
12

Until now i have seen two systems:

  • The automated buildplate from MakerBot and some derivates. In order to get normal precision the belt needs to fit really tight.
  • Add-ons to the head so it pushes the print away. Difficult / Impossible if it sticks too much to the plate.
13

Well, Kevin, a few minutes searching turned up these links. Worth looking into.

http://3dprintingindustry.com/2013/10/30/open-source-automated-build-platform-turns-3d-printer-3d-factory/

I guess, other ideas you might investigate could be:

Printing onto a large circular build plate that rotates by some angle after each print.

Dual, or multiple build plates which can be easily replaced. Say four plates on the same plane could extend time between attention to a reasonable 9 to 12 hours.

But it all depends on how long your print is going to take. Maybe you could just slow down your printer and service it every 8 hours or so? :slight_smile:

If I were you and had access to a load of students and academics, I’d ask my colleagues to submit ideas and filter them for gems. In general, the 3D printing community has adopted the concept of open-source and IP is anathema to our ethos. So contributing to market research in order to build IP for someone else isn’t really in line with the values of the wider 3D printing community.

One pointer you might like to consider: How can you make an FDM printer extrude any colour on demand? Crack that one and a whole load of people will be interested.

I wish you luck with your endeavours.

AndyL

2 Likes
14

Thanks AndyL - I’m really excited that you found the first video in that list. Matt Woods, (the creator of it) is a friend, and he has a pretty nifty setup. However it has some limitations that would prevent it from being a widespread solution.

Also, those were some great ideas. I can promise you that you’re at no risk of having your IP stolen by me or anyone at my university. I have a professor who is an avid supporter of open source design, and while it is a minority voice in the institution, he would never go for that kind of thing.

To be honest, I’m here for reasons concerning less about how to accomplish this and more about whether or not it would be accepted by the community of extremely active printers (like yourself) as a useful product. Is it something you could see catching on?

p.s… If I figure out how to reliably print in any color, at any time, I’ll probably just drop out of school :stuck_out_tongue:

15

Interesting. Do you see there being any widespread acceptance of something like this if it were to work with consistent reliability? Something that could be easily attached as an add-on possibly? It would seem to be the most effective in add-on fashion to me personally.

16

I don’t see the need. Small pieces can be fitted multiples times on the buildplate, big prints take long enough. I wouldn’t let the printer run for 5 days straight without passing by.

Maybe if it could check the print result and decide: continue or wait for human help…

17

Here are some thoughts: It seems the simplest solution would be to have two build plates with a magnetic attachment that both secures and centers the plate. Once the print is complete, an automated process would swap the two plates. Then, you can let the first print cool down on the Initial plate and do whatever you have to do to remove the print ( could be automated as well) while the printer continues on. Similar to how a CNC machine changes tools automatically. At this point you might also want to have a quadcopter navigate to your position to bring you your print when it’s complete ;p)

18

I think more interesting that “continuos printing” are “remote configuring printing”. For example: use an APP on my smartphone for select a model, select a printer near me, configure the printer (material, color, layer, etc. like if I stay in front of a real printer), and receipt the piece in a couple of day.

As a printer owner: have an autoconfigurable printer (meaning, capable of change parameter of slicer, color and type of plastic, temperature of hotend and bed, etc.) on demand through internet.

Sorry for my english , I wish you can understand my point of view, and my english. Apologize me please!!.

19

No worries about the english, I understood perfectly. That’s a good idea, and one that would seem to fit well into the framework of 3D Hubs. With those, as well as continuous printing capabilities, a Hub operator could print as many files as he/she wanted, however they wanted them, from anywhere. It would certainly make it more efficient to create things the moment the customer sent them their file.

Changing colors however, is a much more difficult story. If you can come up with an affordable, easily implementable way to change colors automatically (especially during the printing process) please let me know and I will be one of the first to buy it from you!

20

Maybe amazon will be delivering us 3D printed parts with their supposed future fleet of drones!

That’s a cool idea though. Would something that facilitated that kind of continuous manufacturing be of value to you personally?