"3D Printers" rapid prototyping - for model trains?

I don't know if any of you have seen one in action, but to reproduce a small object, such as a bullet, it takes HOURS. Imagine how long it will take to print a locomotive shell!

A cute concept, one that I've seen presented a number of times before on-line, but far more complex in application for real-world model railroading purposes than suggested or implied here I'm afraid. Perhaps the biggest stumbling block in the process would be obtaining the "master" to be replicated. If one could be bought, such as a freight car, for say $20-$25 from the LHS, then what is the advantage of having the $10k-20K machine? Similarly, who is going to supply the high-end brass locomotive to scan to create a plastic shell for you? Truly original items, otherwise unavailable commercially, would undoubtedly require a programer, or pre-written software from someone...who undoubtedly would require considerable compensation for doing so.

While one day an example of these machines may be found in every home, I'd venture that by that time the hobby of model railroading will have long since met its demise.

CNJ999

This isn't really meant for manufacturing... When you make scale models, you keep your costs down by being able to manufacture thousands of models all at once from inexpensive materials. Mass production through injection molding, for instance. These "3d printers" are really meant for rapid prototyping for engineering purposes. In other words, this is a great way for designers to test a design before they even consider investing millions in new tooling and manufacturing. Model shops used to build "prototypes" for designers as a way to test designs, this cuts out that process by having the machine "print" one, even something like a wrench with moving parts, as in the video.

I could be wrong, but I do not believe the process is as simple as scanning a trolley car and printing a model of it. Even if it was, the model would be very crude and would not have the fine detail that scale modelers are accustomed to. What's more, the price per model would be astronomical.

Still a cool video!

-otto-

Of course as it is now, this technology is far from perfect - especially for something like this, but with time I see it evolving to the point where the kind of detail, etc we want would be more than doable.

I for one yearn for the day where these things have enough precision where a company could make 1:600 scale model planes that rival the Schuco/Schabak brand in detail and affordability. :D

I agree. I think that the important thing to be noted here is that this technology has been discovered. As is the case with many new technologies, they first arrive large, expensive, and "not for the common man" (cell phones, microwave ovens, ad infinitum). Over a relatively short time, however, they evolve quickly, becoming smaller, cheaper, and common. This product has so many applications that I predict that it's evolution will be put on the fast track. Our view towards it now may be the same as a 1970's guy watching Star Trek and marvelling at Kirk's cell phone.

I am monitoring this subject.jpg
Shapeways can make excellent 3d prints
check it out:
http://www.shapeways.com/forum/index.ph ... 10&start=0&


the challenge is to produce a 3D cad drawing. I have been playing with Google Sketchup, but I have along way to go.
I'm trying to design an N scale 0600 car body.

CJN's post is kinda confusing, all the talk of getting masters to be replicated, etc- considering when I see 3D printing, I see a process for developing models very different than the current practices of using molds, physical prototypes, etc, unless my understanding of the process from design to 3D-model-printed-with-a-3D-printer is very different than what it is... meaning, I figured you designed it with CAD software, and the software used those specs, etc, to "print" whatever it was using the printer, in which case designing the parts, etc would need people with the skills, and the specs for what is being replicated.

Travelsonic wrote:CJN's post is kinda confusing, all the talk of getting masters to be replicated, etc- considering when I see 3D printing, I see a process for developing models very different than the current practices of using molds, physical prototypes, etc, unless my understanding of the process from design to 3D-model-printed-with-a-3D-printer is very different than what it is... meaning, I figured you designed it with CAD software, and the software used those specs, etc, to "print" whatever it was using the printer, in which case designing the parts, etc would need people with the skills, and the specs for what is being replicated.

No confusion, at all. Just watch what goes on in the video accompanying the initial post. A "master" wrench is scanned with a hand scanner and a replica supposedly is produced directly from that. There is no CAD software writing, or designer interaction, indicated with the process shown. That's what I allude to in my post when I ask where will the master come from to be scanned, as in a highend brass loco? Granted, most 3-D rapid prototypers base there creating ability on specific written programing, but that is not what is inferred by the process demonstrated here.

Regardless of the exact methology employed, I absolutely connot see devices of this sort in the hands of typical modelers for decades into the future. The Star Trek flip-top communicator, cited earlier in this thread, took 40 years to become a reality...so I'll say again that devices like the one illustrated in the video are unlikely to become mainstream until well after our hobby has ceased to exist.

CNJ999

Sorry, but I just don't see our hobby "ceasing to exist" anytime in the next 40-50 years. I'm betting that some modelers will be able to obtain these printers in 20,
The thing about cell phones was that they needed an entire satellite + cell-tower network in place before they could become practical for everyone, so that was a matter of not just the phones/phone-technology themselves, but the support system as well. With the printers, the CAD programing and computer support exists already. It's just a matter of demand (which I think will come round soon enough) to drive the printers themselves into the small entrepreneur's affordability range.

Cosmo wrote:Sorry, but I just don't see our hobby "ceasing to exist" anytime in the next 40-50 years. I'm betting that some modelers will be able to obtain these printers in 20,

I'm afraid that one needs only to look over our hobby's current demographics and economics to appreciate that the handwriting is really already on the wall. The vast majority of today's hobbyists are 55 and older. Most of them will no longer be active participants in 10-15 years. There is very little new blood coming into the hobby from age groups below 50, a drastic alteration from the situation that existed in the past. Hobby prices have soared over the last decade. Fewer and fewer major manufacturers remain from year to year and the same is true of the hobby's publications. And today public interest in model railroading, even in its simplest forms, is at an all-time low, with toy train sets and associated materials having all but vanished from the public sector marketplace. No pursuit can thrive in obscurity.

The reality of the situation is that by 2025 the more serious form of the hobby we know today can be no more than a pale ghost of itself, aged and priced all but out of existence. Putting rapid prototypers in the hands of supposed hobbyists 20 years hence will come far too late to make any difference.

CNJ999

Hello Guys,
I have been following this thread to see the responses and will give you my opinions as someone who does use this technology to produce our masters.

First off, machines costing under $30K will never produce anything that is usable. They produce such a rough finish that by time you got it smooth, any detail would be lost. Our masters are made with machines costing $150K+ Unfortunately, we do not own this equipment, we use prototypers to build the parts for us. A typical train costs well over $2000.00 for the parts alone, by time you make the body, chassis, trucks, windows, and any separate detail items. This doesn't include the design work, which, typically can take 24 - 48 hours of actual design time at $30 - $100.00 per hour, depending upon the detail level, and where the designer is located. If you have the technical knowledge of how to produce the model yourself, it allows for far greater control over the design process. We have been designing our own models for the last few years now after paying many thousands of dollars to designers to do the work for us, the caveat, the software that we use cost over $6000.00 Unless you are a business, you can see that these costs add up very quickly. Don't forget the ton of research that is required to produce an accurate model.

Secondly, there are many different processes that can be used to make a model, such as SLA, Polyjet, SLS, and others. All of them have advantages and disadvantages, I'm not going to get into them, if you want more info, do an online search on any of the aforementioned terms. The largest concern with producing a model through this technique is feature size, O scale is the most accurate due to it's larger size, most machines can't handle anything smaller than .014" as it may or may not "build" correctly or at all. The machines that can build much finer detail are limited in size to only a few inches.

If you saw the follow up video, showing the work that was required to make the model complete after the scan, you would notice the work that is involved. The guy made it look simple as he only had to make a few adjustments to the wrench. In reality, it probably took him about 30 minutes or so, which isn't much, but there were only a few adjustments. Think of a train that has a ton of detail, it would take several hours. Plus the scan process itself isn't cheap, you may spend over $600.00, closer to a thousand just for the scan. You would be better off just buying the brass model as it would be cheaper and a better quality product. Where the process is a benefit to a modeler is producing something that currently doesn't exist as this process gives you options, albeit, not cheap ones.

We could design beautiful models with such detail that it would rival the best brass pieces. The problem is the technology, while it's getting better, it has a long way to go to be a rival to an injection molded train. If you saw the parts as they are built or printed, you would run. The process used for the wrench, which is SLS, produces the worst surface of all the other processes. It may work for something like a large wrench that gives a wow factor, it is useless for a model train. This I can tell you from firsthand experience, we discarded our model almost as fast as we received it! Completely useless!

Just my 2 cents.

Joe

I assume you're talking about today's technology, right? If so, I sense the people proposing the use of this tech are mostly, at this point in the thread, projecting where the tech will be eventually - whether 10 or 20 yrs down down the line, etc. Surely by then, much like how computers, cell phones, and other tech evolved to become absurdly sophisticated, so will these printers to the point where we won't need to dream about it and can produce what we want, with the details we want. IMO it just becomes a matter of keeping track of the technology's evolution, and above all, being patient.

[In what feels like no time, for a lack of better analogies, the days of Windows 98, dial-up, low hard drive sizes, and long computer boot up times gave way to faster, bigger, higher capacity beasts - though we still have the annoyance of malware, and easily exploitable operating systems. ]

I've got a reprap sitting in my other office. It's not complete; I've not even melted any plastic with it yet. However, I can tell you for sure that the surfaces it will create are NOT particularly smooth and they are clearly built-up out of layers. You could use it for structural items like a bridge pier. You would be disappointed with the quality of, say, a boxcar or worse, a tank car. The surfaces, being made of melted plastic, are not particularly smoothable. You want your tank car to have rivets on the seams? With affordable 3D printing, you would be lucky to have seams.

Hey folks, Bre Pettis here, CEO of MakerBot Industries. We make affordable 3D printers.

Here's the video that's got me psyched about what railroaders can do with a MakerBot: http://www.youtube.com/watch?v=UIJaxJSZ ... ure=relmfu

Here are some stone houses that have been printed on a MakerBot: http://www.thingiverse.com/thing:3943

It makes me excited that the railroad community is interested in 3D printing and I'd like to see more railroad models made on MakerBots. What would be the best way to encourage model railroaders to explore 3D printing? A train designing contest or limited time discount code come to mind, but I'm open to ideas. Feel free to respond here or hit me up on my personal email at bre at makerbot dot com.