What 3D Printing Really Looks Like

Interesting.  Thanks for posting

 

Matt, your picture looks like the product is from a low-res hobby 3D printer such as a MakerBot.  Before I retired, at my employer we made engineering prototypes using a Stratasys Dimension 1200ES printer, from Autodesk Inventor files. The quality of the output was amazing.  This equipment is still too large or expensive for most of us, but it will eventually be in our reach and will be a real game-changer for this hobby!

made entirely from a 3d printer

 

Indeed, 3D printing has evolved to a point where in which the printed model is very good.

Of course the machines are quite expensive.  And don't think that the printing is one part at a time.  There are units out there now that can make multiple components at once. 

 

The example by bigdodgetrain demonstrates just how we may all be using systems of this nature in the near future.  The possibilities are endless. 

 

GN Man, your talking about hugely expensive machines and the higher resolution the longer it takes to print. That photo is an accurate representation of the resolution of even the more expensive commercial printers. High resolution printers are out there but they are the exception, not the rule.

this really isn't correct, and I wish you'd not spread such a falsehood. The 3d printers undergrads are using here at school are lightyears ahead of what you posted, and a friend of mine is regularly making interior parts for his car on one at the place he works. His parts are smooth, highly detailed, and ready to use upon completion. 

 

Having reverse image searched your photo, that is the difference between 150 and 250 DPI printings. Some machines now have at least a 1600 DPI resolution. I'm not sure what your agenda is with this post, but it isn't honest to the realities of the technology, which BTW continues to improve.

Why are all you so fixated on "agendas". Like I said, High res printers are the exception,not the rule. So your undergrad engineering students have access to a mega expensive printer to print on and your buddy has access to a high end printer. Your are talking about a huge capitol expense for a company and in order to make it viable you would need to have lets say 10 printers at what 200k each spitting out what, 3 loco shells every lets say 12 hours so lets say a production run of 2500 shells so you've got 10 printers running 6 bodies each per day so 60 O scale bodies per day. That's 42 days for one production run, so cost on a production run for just the printers you need to get 18.26 per part to pay for just the printer on a 5 year capitol expense so realistically more like $100 per body. High res printers are the exception, not the rule and they are extremely cost and time prohibitive to use in the production of toys.

Andrew is right...

Trinkle Trains is getting outstanding models with relatively inexpensive printers at home in his personal workshop not a huge corporate or school environment. I am looking into It and playing with it too. Waiting for the price to come down even more.

Trinkle--- care to add something here?

I guess it all comes down to what you think is expensive versus someone else?

You are right Matt 3D printing was and still can be like you posted but like all technology things change overnight.

Few, if any, of us here can predict with any degree of certainty the future of 3D printing for enthusiasts/hobbyists -- aside from the fact that the technology will improve.  As for whether it's something for the masses remains to be seen.  And even when the technology improves, it doesn't necessarily imply all of us will be producing locomotives and rolling stock with our own personal 3D printers in the near future.

 

Here's an analogy that may help explain.  Nearly 8-10 years ago, professional photographers had the option of printing their own large, wide-format prints and canvases rather than send them off to a lab for printing.  I dabbled with an entry-size wide-format printer and found the devil to be in the details.  When EVERYTHING worked as planned, it was great.  HOWEVER, that didn't always happen, and there were several variables that always made the process challenging -- not impossible, just challenging.  Bottom line... at the end of the day, I decided it was more cost-efficient for me to have a reliable lab do ALL my printing.  Could I do it?  Yes.  Did I want to do it? No.  Admittedly, YMMV.

 

I think the same phenomenon is gonna happen with 3D printing as well.  Lots of folks will dabble with the technology.  But at the end of the day, most model train enthusiasts are still gonna be buying their trains from the big guys in whatever the importing/manufacturing landscape will look like 5-10 years from now.  That's my story... and I'm sticking to it... and least for now! 

 

And lest anyone think I'm being deliberately down on this new technology, that's not the case at all.  Rather I think folks just need to approach it with both eyes wide open.  I was one of the "earliest adopters"  in the desktop publishing world when Adobe PageMaker hit the Mac/PC world in the mid-1980's and HP Laserjet printers were just being unveiled (and pretty expensive too!).  That was MUCH different though.  It was more "turnkey" (from the consumer perspective) with much fewer variables than wide-format printing and 3D printing.  And as a result, the adoption curve grew exponentially -- first among IT professionals, and then among consumers in general.  I just don't see the same thing happening with 3D printing -- at least not for the general consumer.  Most folks just aren't gonna have the time and patience with it.  But that certainly won't stop college undergrads from having a ton of fun with the technology in the meantime.  Count on it! 

 

David

From what I understand much of the quality is dependent on what you send to the printer. Like any project it's all about the research and final drawings or in this case 3D point data.

Samples I also found online of what appear to be in some cases exceptional and on par with any master tool makers pattern.

Good thoughts Dave..

Very true. For me it's a tool to gain prototypes the big guys will simply not make for me. Darn it.

The prints you can order from on-line sites like shapeways are done with very high resolution printers and can even print in jewelry quality metals.

If you're thinking about using a 3D printer to mass produce models that are as good as something from MTH or Lionel, then, no we're not there yet. It's not a practical solution. But if you 're wanting to do small runs or single items, you can get some incredible looking prints to make whatever custom items you're willing to take the time to model.

 

I've seen some custom N scale locomotives that were so good you couldn't tell the difference between them and an injection molded model.

This is another N scale unit you can currently buy on Shapeways:

S Scale Hopper

 

 

HO Scale hopper body

And it's only going to get better! It's a great time in history to be alive and modeling!

- RICH

Don't get me confused, I am not knocking 3D printing or the technology behind it. Its amazing and a huge advantage to engineers and companies to be able to build a prototype in hours without having to machine or build molds. And the possibilities are endless but a huge portion of the printers most people would have access to are not the super high end printers. I plan on getting a printer in the near future just to mess with, most likely a Makerbot but the fact of the matter is that its not a technology that companies can just start printing off models and even when you do get a perfect part they still aren't perfect and they definitely aren't model quality but some printers can print super nice but the higher the resolution means the thinner the layer you put down and the smaller amount of material you can add meaning the time to print goes way up. I hear people talk about how Lionel or MTH can just print a loco and no molds are needed. The reality is that most printed parts look just OK to good

You would not want to 3D print production runs  of larger items  with current technology. It takes too long and costs too much. Now keep in mind I am thinking of rolling stock not brake wheels or fire hydrants. What you can do is print prototypes that are then used to make casting molds or final model before having a CNC machine cut a die to be used as an injection mold. 

There are 3D scanners that can be use to reproduce lost or broken parts as small as a brake wheel to real rolling stock. With some items there is almost no skill required with others would have to have the skills of a draftsperson / engineer.

 

Imagine importing a file with from a company such as Greenbriar that has the constituent parts of a tank car being able to manipulate the model (to be 3D or CNC ready) and sending the resulting data to a printer / CNC machine.  It is a skill that I do not have, but others here probably do.

 

  

I d not think they will ever replace injection molds with 3D printers only. Mass production still needs speed and injection can do that.

But the surface finish varies greatly on the 3D printer itself. Some are getting a very nice finished surface.  Boils down to how high-tech of a machine you have access to.  Model car guys are making parts with very high detail and almost ready to use.  

From what I understand of Lionel's use of 3D printing, they use the design for the product and use to mock up a prototype to check for possible issue of clearance, parts fitting etc. before they create an actual mold so as to not waste money fixing mistakes in the design afterward.

You CAN get a good model from 3D printing ... and in O scale.

 

Check out the photos of a Chicago Rapid Transit 4000 series car on my blog. Some photos are at:

 

http://ctalayout.blogspot.com/...baldie-in-paint.html

 

Yes, there are some lines, but once painted, weathered and in person, its not too noticeable. Other prints are near perfect.

 

Was it printed in a complete shell? Nope, it was printed in 9 separate pieces then assembled (two ends, roof and three sections per side).

 

Cost? $230 at Shapeways.

 

Design work was all done on free software (Sketchup).

 

Its viable if you have an odd prototype you wish to model and understand the limitations. I'm happy with how it turned out. Is it perfect? Nope, but still better than I could've scratch built myself.

 

CTA Fan

Having used several 3D printers, I'm with you Matt.  It's not a production quality solution.  They are great for prototypes, and some one off models, but they cost too much and take too long for anything else.

 

And as an aside, don't get a Makerbot, get something else like a Stratasys.

Here is what I printed out at work for my passenger platforms:

 

 

Sinclair, Ive got a friend who works for Stratasys and even if he gets me a deal I am not sure I can make it happen but I will look at my options

3D printers use a consumable and the consumable will be the key for any 3D printer.

 AG.

 

Its not just the consumables, there ability to print in so many forms of media is incredible including metal. I can see printing parts to use as masters for lost wax castings as viable being that the surface will still need finishing. You can truly make one of a kind items that look good and its amazing. even at the high res the surface look and texture is still not there

What you say is correct. For mock up and master is Ok, but for production is not.

Same whith the printer, is less expensive an offset than a digital. I am not saying is bad what I am saying is the market will be controlled by consumables.

AG.

Or more realistically, an SLA type printer like the Formlab or B9 Creator.  These are also desktop consumerist printers, but use cured resin as opposed to fused filament (like the Makerbot) for the material. Generally, SLA printers are going to offer higher resolutions with thinner print layers.

 

I think the important distinction with 3d printing is between short run/prototyping versus production.  Can printers produce just about anything you can imagine? Absolutely.  Can they produce 5000 pieces in a timely, affordable manner?  Absolutely not.  So yes, 3d printing could work well for creating masters, albeit requiring post printing finishing work.  And no, companies like Lionel cannot pragmatically use the technology to manufacture high volume production runs.  As usual, the caveat applies that current technology has certain limitations and that technology will invariable improve and cost less in the future.

 

The four issues with 3d printing for production at the moment are resolution, speed, cost of materials, and what I'll call value.  The value component is simply the acknowledgement that much better results could be gained if the very best (and therefor expensive) printers could be used.  But for what most of us could afford, and even for most manufacturers, the first three issues are the prevalent ones.

 

Here is a bolster for a modern 100 ton roller bearing freight car truck.  It is printed at 1:48 scale on a high quality Objet printer.  Even using the Objet printer, the final product would still require some finishing work to clean up the surface.  Also, given the small size, the model is quite fragile.  I've damaged parts of it while trying to remove support material from the inside of the casting.  This part was printed by the same 3d printer hub that printed the tiny naval boat shown in Erik's post above.  Note that not all support material has been removed from the inside, but the outside of the casting is the bare plastic.

 

 

 

I've posted this next model elsewhere here on the forum.  It's printed at 1:10 actual size on Stratasys 1200 and 250 FDM printers.  Again, the resolution is okay but not finish quality.  Worse yet from a production standpoint, there are probably 150 to 200 hours of actual print time to create the parts for what you see. I've since added more detail to the castings as well as full brake rigging that will take even more time to print.

 

 

 

 

I hope this helps give a sense of what is possible, but also what some of the limitations are.  Better printers produce better results, but at a significantly higher price. Practically speaking, printers have limited use currently in production, but great value in the prototyping process.

 

If anybody has any questions, feel free to ask.  I probably know just about enough to be dangerous!

 

Jim

I've included this video as it mentions toward the end of the process about what finishing techniques are available and required depending on the degree of finish.  The beautiful fan blade is painted to achieve that super smooth polished finish.

 

One thing the fan blade and the springs in my truck above illustrate is that there are certain types of parts that cannot be produced through injection molding.

I think there is confusion because that's not what you said in your original post. You said "I have seen a lot of printed plastic and the technology is a long way off of being a viable manufacturing option." The title of your post is "What 3D printing really looks like."

 

You made an open-ended statement that sounded like 3D printing is of poor quality, without properly qualifying that statement by noting that commercial higher quality printers currently in use CAN ( and currently do) print high quality commercial items. It was misleading. You photo is not representative of what 3D printing really looks like. Obviously, there are many 3D printed items of high quality currently being marketed, including some medical devices. In fact, Lionel is doing some quality mockups of their new engines with 3D printing.

UV light bonded LIQUID stereo-lithography can create extremely fine details, not the crude details depicted in the first image.

 

 

Andrew 

Here are some examples of what is available and how a real cottage industry has be started by the technology: http://www.shapeways.com/shops/Eastern_Road_Models

I saw several of the newest 3D printers cranking away at B&H cameras in their main NYC store, just 2 weeks ago (still kicking myself that I didn't take photos of them) and got to handle the prints. The prints were extremely light but not as 'smooth' as I would have imagined. I guess I can assume these aren't the 'fine detail' ones mentioned here earlier? The detail on those was nice, but nowhere near lost wax quality. They all had a look that screamed for some light sandpaper over the surface.

As for the high detail ones mentioned, I'd have loved to have seen one of those in action as well but they apparently didn't have one of them...

If you ever go into the store, the candy dishes they have scattered around the place were all made with their 3D printers. I suggested that they sell those prints, which came as surprise to the employee I talked with as that had never crossed his mind that anyone might want to buy one. Heck, I'd have bought one right there, just to be able to show people. The novelty will be there for a while, I'd think.

Has anybody tried printing scale diesel locomotive pilots?  In converting to scale sized couplers, the 3-rail pilots have too big of a hole/gap left.  Even when 2-rail pilots are available, they usually also have too big of a hole/gap and are often incorrect.  Also, a lot of the pilots were modified by the roads and are thus unique.  It would be great to be able to print a correct scale pilot for each locomotive.

I've thought about it, I was planning on getting a printer soon but I'm not sure the lower end printers I can afford will have the resolution to make them look good

LIke early TV's (now digital),  pocket calculators, digital photography, and any technology you can think of....the quality will improve and the price will go down.  Sit tight, and stay informed.

The quality is pretty good but I still find the wavyness of the flat surfaces to look bad and this is my biggest issue with it. High res printers to a pretty dang good job but at 100k plus that's a bit out of my price range.

You should be able to smooth out that surface with acetone.

But say you have rivet detail or window moldings or other details. Ive wondered if you could somehow index the print so it could be removed have the surface smoothed out and then put back in for the smaller details

"A 3D printed elephant, but one that is just a tenth of a millimeter tall, using nanoscale 3D printing techniques."  That's 0.00393701 of an inch tall, the thickness of the ink deposited on a piece of paper when you make a laser copy!

 

Seems to me that when one makes a statement like "WHAT 3D PRINTING REALLY LOOKS LIKE" they would at least do a little research.  I believe that statement refers to many of todays home 3d printers.  The 3d printing industry is a 500 billion dollar a year business and I guarantee they didn't get there at $500 a machine!  I remember back in the early PC days that there were folks trying to compare an IBM 360 and an Apple II and making the same kinds of noises.  Speed, quality, and materials used are all TECHNOLOGIES not concepts.  3d printing is here and it works, how much detail you want or need 'today' is matter of how much you want to spend.  Will the machines be better NEXT year, sure!  Will the materials printed be better NEXT year, sure!  Ohh and those two statements can be repeated EVERY YEAR from now on.  I love my low res, really slow 3d printer. and despite its low res it really is fun to make your goodies so we can 'PLAY' trains.    Russ

 

Nobody has mentioned the 800 pound elephant in the room.  In order to do 3D printing, you have to have the 3D model created first!  That's the difficult part.  Unless you are accomplished with 3D CAD, you're not going to be making a lot of original parts.

 

For small quantities, I can see 3D printing being cost effective, but for any quantity of parts, you're never going to beat something like injection molding.  Even the fastest printers will take an eternity to crank out 10,000 or more parts!

 

I keep looking at the 3D printers, but I wonder how practical one would really be, and would I get any real value from it.

 

I have been having that same thought John, Making the 3d models has a pretty steep learning curve to become good at it. I still have yet to see any sort of printed model that looks as good as an injection molded part. Everybody keeps arguing the exception, not the rule. That little elephant is an exception, not a rule. Its not viable affordable tech yet. They take forever to print and while some of them look pretty good, they still dont look great.