Author Topic: Early 2015 3D Printing State of Technology Discussion  (Read 2673 times)

0 Members and 2 Guests are viewing this topic.

lashedup

  • Crew
  • *
  • Posts: 879
  • Respect: +108
    • Model 160
Early 2015 3D Printing State of Technology Discussion
« on: January 08, 2015, 04:48:59 PM »
0
So I had to go to the Consumer Electronics Show this year and decided to take a detour through the 3D printing display area. This is an industry only show filled with media, buyers, exhibitors and lots more - over 150,000 people attend and it is in the top 3 biggest trade shows that come to Las Vegas every January.

The 3D printing exhibits are all in one location in one of the display halls and draw most of the small to large players. Everyone from FormLabs to EnvisionTec to Autodesk are on display with most of their sales, marketing and technical staff on hand. I'm not an expert on 3D printing outside of my interest here and what I've gleaned doing a lot of online research. So I was able to ask a lot of questions and view a lot of machines and samples. Overall I came away with a few solid ideas and impressions and still have some minor confusion regarding some of the ways specs are reported and how those correlate to the output.

The 3D printer market is exploding. Once relegated to high-end prototyping with just a very small handful of companies making printers (for big $$$), we are now seeing a ton of low-cost units starting to reach the market. Technology and volume is changing and expanding very quickly in this segment with the resolutions and build volumes (size of things you can print) going up and costs (printers and consumables) coming down. Overall it will be interesting to see what exactly a general consumer will do with a 3D printer. Even large companies like Dremel have introduced consumer-level 3D printers now that you can buy at big-box stores like Home Depot. I could see some of the general public buying these, printing a couple things and then letting it sit in a corner because they don't really have any practical use for it. Hobbyists like us however see some big potential and low-cost manufacturing could become a niche thing as well. We'll see what happens.

One of the more amusing things I saw was the commercial amusement use (for lack of a better term) where people would get a complete body scan like this:



and the company makes a 3D model based on that scan that looks like this:



Just a little creepy not to mention what you do with that lovely statue of yourself is beyond me. One company would even take a 3D scan of your face and put it on an action figure that comes in a commercial box and everything.

Anyway, the printers on the market are divided into primarily two kinds of 3D printers. The flood of recent low-cost units use a plastic filament (looks like weed whacker line) that gets fed into a print head, heat up and builds up layers like a baker piping frosting on a cake. This technology generally produces 3D prints that are a little rougher, have obviously/visible layers and aren't typically used for things requiring very high detail (engineering prototypes, jewelry, etc., etc.). The filament material is cheaper than resin and generally these printers are a bit quicker due to their lower "resolution" (higher layer thickness).

Filament printer:



Large-scale filament printer:



The other type of 3D printer uses stereolithography (SLA) which uses light (most commonly laser or projector) to fuse a layer of material in a liquid resin bath. The SLA printers typically have higher resolution (smaller layer thickness measured in microns) thanks to the very narrow focus beam of light from a laser or projector. Because the layers are much smaller, the print times take longer. Resin is also (right now anyway) more expensive than the plastic filaments.

SLA printer:



Based on most of the samples I looked at, you would need a printer capable of making 3D prints at 30 microns or less to get an acceptable amount of smoothness. The specs on these printers are a bit confusing in terms of how they relate to what you see in the final print. There are a number of variables involved that include the type of resin you use, the stepper motor for the platform, the type of projector or laser and the amount of beam focus just to name a few. Usually the specs report an X/Y access measurement (in microns) and a Z axis measurement in microns. To add more confusion they will also sometimes list a "Laser Spot Size" or "Minimum Feature Size" as well.

Overall though in the samples I saw, the printers that claimed a 30 micron resolution or less produced the least amount of stepping getting closest to a smooth surface. Generally the 25-35 micron printers cost between $1500 - $3500. The few companies making 10 micron printers cost between $5000-$6000 dollars.  At 10 microns the pieces are by all outward appearances almost completely smooth. Or put another way, I wouldn't hesitate to use them in a second for model making. The 25 micron printers are actually "good enough" that I would also likely use them. Here is an example from the XYZ Printing Noble 1 (which was announced at CES and won't be available till 3rd quarter this year). This printer costs only $1500, has a decent build volume that could handle an n scale passenger car and prints down to 25 microns.

Chess piece:



close up 1:



close up 2:



For a little perspective, notice in the photo above my finger prints and how deep those grooves are in comparison to the build lines you see on the final piece. If you look at this piece from a foot away you don't see the lines. Further, this is an unfinished piece. Painting it will actually level out the surface even more (particularly the primer you use).

Here is another example. This was printed on a FormLabs 1+ which costs around $3000 (including resin and a finishing kit) and also prints at 25 microns. The black cyliner is part of a mockup of a surgical device and was a slide piece (you can see the slide marks) that inserts into a larger cylinder. This piece was beat up pretty good from use. Again these are unfinished pieces with no paint:



Here is a closeup (sorry this was shot through glass in a display case):



You can see the build layers in the closeup, but even on a round cylinder, this is very acceptable.

The FormLabs guys are actually aware of the model makers, particularly model railroaders. I chatted with him for a bit and he offered to print up some samples if I send him files. Mike Skibbe has some Solidworks models already produced and we'll send them some worst case n scale stuff with rivets, passenger car fluting and maybe a tank car with rounded ends and rounded relief dome on top.

Honestly at this point I may take a serious look at one of these printers if the samples come back at a detail level I can live with. It will probably be a couple weeks or more till we get something back from them, but I think there are huge possibilities in this whole thing.

-jamie
« Last Edit: January 10, 2015, 11:27:35 AM by lashedup »

C855B

  • Crew
  • *
  • Posts: 10860
  • Respect: +2415
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #1 on: January 08, 2015, 05:02:32 PM »
0
Thank you VERY MUCH for the detailed writeup! I agree with your assessment - extruders are out, the Noble and the FormLabs 1+ are where we need to be. That we're seeing this resolution from sub-$5K printers is even better. The wall thickness and smoothness of the 1+-produced part is comparable to injection molding. Interesting that they produced the piece at a slight angle - this was obviously to get maximum smoothness for the cylinder.

Exciting times ahead!
...mike

http://www.gibboncozadandwestern.com

Note: Images linked in my postings are on an HTTP server, not HTTPS. Enable "mixed content" in your browser to view.

There are over 1000 images on this server. Not changing anytime soon.

ednadolski

  • Crew
  • *
  • Posts: 4809
  • Respect: +1756
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #2 on: January 08, 2015, 05:12:05 PM »
0
Very cool stuff, thanks for sharing!

I can't see myself ever buying a 3D printer, not for the resolution I would need for N scale detailing parts.  But I am very interested in learning how to design the parts and send them in to a printing service (like I currently do with 2D photo etch).  What software is there for creating the CAD drawings?   Do any of these manufacturers offer print services?  Or how about sample CAD files, that a user could modify to produce some simple parts?

Ed

lashedup

  • Crew
  • *
  • Posts: 879
  • Respect: +108
    • Model 160
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #3 on: January 08, 2015, 05:21:28 PM »
0
Interesting that they produced the piece at a slight angle - this was obviously to get maximum smoothness for the cylinder.

Are you referring to the black piece? That one was printed building from the top to the bottom longways according to the FormLabs guy. The scratches you see diagonally were made from rough handling and sliding the piece in and out of another cylinder. You can see the layers in the closeup as ridges on the edges. Looking at it though, it does appear that it could have been printed diagonally.
« Last Edit: January 08, 2015, 05:23:55 PM by lashedup »

pjm20

  • Crew
  • *
  • Posts: 1145
  • Gender: Male
  • Modeling the Bellefonte Central
  • Respect: +144
    • My Youtube Channel
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #4 on: January 08, 2015, 05:26:11 PM »
0
Most companies, Makerbot for example, have software that can build very basic models, but really converts your drawings into a file that their 3d printer can print. As for drawing software Sketchup is a great start, and I'm sure their are other cheap programs; I have my eye out for AutoCAD.
Peter
Modeling the Bellefonte Central Railroad circa 1953
PRRT&HS #8862
Live Steam Enthusiast

Check out my Youtube Channel: https://www.youtube.com/user/PennsyModeler

Iain

  • Crew
  • *
  • Posts: 4661
  • Gender: Female
  • Na sgrìobhaidh a Iain
  • Respect: +385
    • The Best Puppers
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #5 on: January 10, 2015, 10:17:49 PM »
0
I use Bricscad, as it is Linux native and very similar in UI to AutoCAD.
I like ducks

reinhardtjh

  • Crew
  • *
  • Posts: 3006
  • Respect: +365
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #6 on: January 18, 2015, 10:46:45 AM »
0
This Kudos3D Titan 1 printer looks interesting also.

http://www.3ders.org/articles/20140515-fastest-high-res-titan-1-dlp-3d-printer-set-to-launch-within-two-weeks.html

http://www.kudo3d.com/products/

The price is about 2/3 of the FormLabs Form 1+ and the build area is bigger.    They aren't shipping yet so it will be interesting to see how that goes.

What's nice about the Form 1+ is the types of resin available.  They have the black for high detail and a burnout resin for making investment castings.

Edited for clarity
« Last Edit: January 18, 2015, 08:36:12 PM by reinhardtjh »
John H. Reinhardt
PRRT&HS #8909
C&O HS #11530
N-Trak #7566

PJPickard

  • Crew
  • *
  • Posts: 205
  • Respect: +27
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #7 on: January 18, 2015, 11:58:08 AM »
0
Thanks here too for the write up. This is really heading in the direction I am hoping for.

jimmo

  • Crew
  • *
  • Posts: 902
  • Gender: Male
  • Representing Willmodels
  • Respect: +6
    • Willmodels
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #8 on: January 18, 2015, 11:16:17 PM »
0
Very cool stuff, thanks for sharing!

I can't see myself ever buying a 3D printer, not for the resolution I would need for N scale detailing parts.  But I am very interested in learning how to design the parts and send them in to a printing service (like I currently do with 2D photo etch).  What software is there for creating the CAD drawings?   Do any of these manufacturers offer print services?  Or how about sample CAD files, that a user could modify to produce some simple parts?

Ed

I finally got started building models in CAD by following a link on the Shapeways site. I clicked the "Design" button and then "Get Started" where I watched a short video and almost immediately started using a simple (free) program from AutoDesk called TinkerCAD.
James R. Will

bbussey

  • Crew
  • *
  • Posts: 8888
  • Gender: Male
  • Respect: +4712
    • www.bbussey.net
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #9 on: January 19, 2015, 12:11:26 AM »
0
@lashedup — Jamie, what would be good to know on both the XYZ Printing Noble 1 and the FormLabs equipment is if there is a tolerance, and if the printers output exactly to scale.  For example, the ProJet3000 series of printers such as what Shapeways uses for their FUD items must be calibrated to scale the items properly on the X, Y and Z axises.  Models designed for the ProJet contractor I use must be scaled up 0.55% on the X axis and 0.25% on the Y axis in order for the part to render at the intended dimensions.

But this looks promising.  The rook chess piece appears to be more fine that XHD renderings on a ProJet3000Plus, and at a $1,500 upfront equipment cost (which will only drop as time advances and new printers hit the market) is very tempting.

Bryan Busséy
NHRHTA #2246
NSE #1117
www.bbussey.net


reinhardtjh

  • Crew
  • *
  • Posts: 3006
  • Respect: +365
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #10 on: January 19, 2015, 01:05:33 AM »
0
On the 3DSTFC Yahoo group (a 3D focused offshoot of the Steam/Freight Car group) there is a guy named Bill Lane (www.lanestrains.com - yeah, he's an SPF) that bought a Form 1 (not the 1+)  and has been learning how to work it.  He hasn't mentioned any calibration issues, but he is focused on S scale though so his tolerances aren't as tight as ours.  I'm not sure what the exact differences are between the Form 1 and the 1+.  Their web site only says the 1+ is faster and the laser is more powerful (not sure why that is important).
John H. Reinhardt
PRRT&HS #8909
C&O HS #11530
N-Trak #7566

lashedup

  • Crew
  • *
  • Posts: 879
  • Respect: +108
    • Model 160
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #11 on: January 20, 2015, 01:45:39 PM »
0
This Kudos3D Titan 1 printer looks interesting also.

http://www.3ders.org/articles/20140515-fastest-high-res-titan-1-dlp-3d-printer-set-to-launch-within-two-weeks.html

http://www.kudo3d.com/products/

The price is about 2/3 of the FormLabs Form 1+ and the build area is bigger.    They aren't shipping yet so it will be interesting to see how that goes.

What's nice about the Form 1+ is the types of resin available.  They have the black for high detail and a burnout resin for making investment castings.

Edited for clarity

I looked at the Kudo3D printer as well. It is a DLP projection type printer and they mentioned that as the detail level goes up, the build volume goes down. Something to do with the focusing depth. Logically it would seem that the laser-based systems (depending on design) might not have this specific limitation. However the laser itself becomes a limitation depending on how focused the beam width is and how many mirrors it needs to bounce off (which affects the wavelengths a bit in terms of focus).

One other company called Old World Labs is claiming a 1 micron detail level. Somehow I missed them at CES, but I'm going to contact them to see samples and see if I can get a sample printed as well. Their MC1 is supposed to be fairly cost effective (although at that detail level I'm guessing $5k price point).

Either way, the good news is that we're getting much closer and the price is coming down quickly.

-jamie

C855B

  • Crew
  • *
  • Posts: 10860
  • Respect: +2415
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #12 on: January 20, 2015, 02:23:53 PM »
0
http://oldworldlabs.com/product/owl-mc-1

The MC-2 is spec'ed at 100 nM resolution, accuracy of +/-50 nM. They seem to be thumping the subscription/leasing sales model. They also spec both models' build rate at 1 cubic inch per hour, but don't mention at what resolution.

In general, it appears that we have crossed the threshold of acceptability for resolution. The issue, then, becomes build rate. It'll be great to get 0.0005" resolutions, but if it takes 5 days to get a model out of it we have a problem.
...mike

http://www.gibboncozadandwestern.com

Note: Images linked in my postings are on an HTTP server, not HTTPS. Enable "mixed content" in your browser to view.

There are over 1000 images on this server. Not changing anytime soon.

Jesse6669

  • Crew
  • *
  • Posts: 578
  • Respect: +1287
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #13 on: January 20, 2015, 04:16:22 PM »
0
What's the price of this printer?  Are we talking $20-30k or so?  Even if so.. in a couple years I could see consumer level pricing.. 
http://oldworldlabs.com/product/owl-mc-1  ...The MC-2 is spec'ed at 100 nM resolution, accuracy of +/-50 nM.

Baronjutter

  • Crew
  • *
  • Posts: 509
  • Respect: +11
Re: Early 2015 3D Printing State of Technology Discussion
« Reply #14 on: January 20, 2015, 04:31:43 PM »
0
I really hope my makerspace gets a UV based printer.   The "glue gun" style printers just don't give hobby-usable results.