This is very cool

Ken:

This is very cool ... and unless I miss my guess, Ken's going to develop a full-blown MRH article on how he did this, starting with how to use SketchUp to develop a design and then sharing the final results.

Yes-sir-ee, this 3D printing stuff is going to radically alter the hobby - and create a burgeoning new market for 3D design files and 3D projects you just buy from places like Shapeways.

Very cool indeed!

Article

Correct.  I just need to get the time to work on it - I've got it started, and I promise I will make it happen.

I also want to take some better photos, but that'll take a little work and time.  At the moment, it's time to go to work again!

Definately cool

Just wish I had some change to develop some parts,etc....    well, time will bring more and more stuff available.....hopefully it will replace the detail companies when their well dries up because of the RTR stuff.(which I hate to say, but will probably come sometime in the future....Luckily, I have a couple hobby shops worth of parts downstairs)

RTR does not kill kitbashing/scratchbuilding

I think it's a misconception that RTR is killing kitbashing and scratchbuilding, and therefore the market for detail parts.

What is actually happening is RTR creates more focus on kitbashing and scratchbuilding, not less. My own example will illustrate.

Thanks to RTR, I decided it would be practical to build a larger basement layout modeling the SP in Oregon during the 1980s. Because I am trying to faithfully model the SP prototype on a larger layout, I can't buy everything I need, I have to build the stuff I can't buy. For instance, I need water cars, which are special cars that go behind the locos in the warmer months, by order of the US Forest Service.

Notice, I need 6 of these cars, not just one. If I had to build more stuff because I could not buy so much RTR, then I would not try to build such a large layout, I would have gone smaller and only needed one or two water cars, not six.

Thanks to RTR, my layout project is much larger and I need to build more of the stuff I can't buy. The end result is RTR didn't stop me from scratchbuilding and kitbashing, it simply changed my focus to the stuff I can't buy. I will still need plenty of detail parts for those cars I have to kitbash and scratchbuild.

That's it, I'm in.

I've been following Rene' Gourley's accounts of his experiences with the SketchUp> Shapeways path to special item production, literally this morning, and this wonderful cab clinches it. 

How soon can you get that article out Ken? I'm ready for it! 

Jeff Allen

Shapeways and water car..

Joe,

Could you post a pic or two of the water cars?  And I'll be looking forwards to the shapeways article; I was happy with the cab I ordered also, thought I had to get one that was already designed, and modify it.  (The material is harder to work than styrene.)

hm, I can see one thing changing

I see the definition of scratch building changing some.  If I can order a bunch of parts printed to fit, then I don't have to do sheet styrene to build them.  I see one issue coming as well - how to catalog the parts so people can find them.  I mean it is great to have the cab designed and printed, now if I want the same cab, how do I find it to print, without redoing all the work myself?  And how do I determine who has good parts designed and who doesn't?

Is 3D printing scratchbuilding?

I'm working on it, Jeff!  Part of the challenge for me is in slowing down and taking the time to remember and record every step.  There are so many things I do without thinking that it's easy to get ahead of myself.

Actually, I think the NMRA would consider the cab in the photos to be scratchbuilt.  I did all the drafting and design work, and the fact that I used a 3D printer instead of a knife to create it is less important.  I can tell you that I spent a LOT more time getting the drawing right than I would have building the cab, so skill and modeling time are invested either way.

With that said, I don't really care what you call it.  The point for me is to have an accurate cab, and especially to have MANY accurate cabs.  If I scratchbuild them, then 10 cabs means 10 times the time and work.

Shapeways has some work to do on that front, and they know it.  It is (usually) possible to find things if you already know they exist, but just browsing can be frustrating.  Of course, the best way is to keep a file of the models or modelers you know about so you can get to them easily whenever you like.

Same way you determine that for any other manufacturer.  Word of mouth, personal experience, reputation, photos of the products.  How do you determine whether or not a new model is worth purchasing before you've seen it or read a review?

Victims of their own success

Regarding the detail parts manufacturers, I see two distinct issues in my local hobby shop.

The Cal-Scale and Precision Scale racks sit untouched other than when I or a very small handful of other customers pick over them.  There are not as many steam modelers as there used to be, and a higher percentage of the steam models today are good enough as RTR models that there's no need for the detail parts.  When the best you could get was a Bowser, Mantua, or Roundhouse engine, you needed a lot of detail parts to make a really accurate model.

On the other hand, the Details West and Detail Associates racks turn over surprisingly well.  The problem is that there is such a wide selection, and the parts are so specific, that you have to maintain a huge inventory to satisfy more than one or two modelers.  It costs a lot of money to manufacturer a run of those parts, and then distribute and display them, only to sell a couple dollars' worth at a time.  I would not be at all surprised to learn that they're at best a break-even deal for the shop.  It's not like they can have a bin of 15-20 different parts and expect to sell them to every kitbasher who walks in the door.  There's probably 8'-10' of wall space taken up by Detail Associates alone.

So, while the steam detail parts may well be dying out because of RTR models and a waning interest in steam, I think the modern parts manufacturers are victims of their own success.  They have made it possible to model a specific car or locomotive on a specific day, but in doing so they have caused more modelers to want to do exactly that - meaning they have to make a wider range of parts than they did 20-30 years ago, but sell to roughly the same number of modelers.

Hopefully 3D printing will help the situation.  While the per-part cost is significantly higher (that cab cost me $20), each model will only need a few unique parts, so hopefully the cost will not break the bank.

SP Water cars ...

Here you go ... First a Bob Zenk model (in May 1982 Mainline Modeler, this is the real model that I got to photograph):

And some prototype water cars on the Siskiyou Line, CORP era (late 1990s):

(Click to enlarge)

I just couldn't wait...

I became impatient waiting for Ken to finish that article so I signed up for a Shapeways account, exported a collada 3D file from SketchUp and ordered up a small chair. $15 delivered, and if I'd specified a cheaper material it could have been half that.

The whole process took almost 8 minutes. 

Now, I've had some experience with 3D modeling, but if the physical results are as credible as Rick and Rene's experiences seem to indicate then I must say that the "manufacturing" end of this process is nearly a non-issue. I used to work with 3D Studio and the old gmax, and am becoming familiar with SketchUp. These tools are not conventional model railroading media, but they are central to the railroad simulator world. The learning curve is pretty steep, and the experience is not interchangeable with traditional modeling, but I predict it will complement the heck out of it. 

Wow, this is very cool. The mind reels.

Jeff Allen

The run of 10

Joe, or was it Ken hit the nail on the head: the sweet spot for 3D printing is not the singleton item, but when you want a short run of them.  The length of that run is getting larger as the technology is getting cheaper and better; ultimately, it will overtake injection molding for most of our purposes. 

One of the keys that accelerates this trend is that you can skip manufacturing steps using the technology; for example, there is no need to install a bunch of stand-off details if you can print them in place.  I have an ore hopper car design, which I haven't test-printed yet, that includes all the brake gear and grab irons.  They would be a challenging multimedia kit, if ever offered; as a 3D print, you basically print, paint and letter it.

On the other end of the scale, I think once Ken gets used to the process, he will find he is faster with computer modeling than with styrene.  However, I find that time spent in front of the computer, or waiting for the printer, is less satisfying than time spent with a syringe full of MEK and my Olfa knife (okay, that sounded bad).  When I want ten of something, like my switchrods, or turnout controls, then I will reach for 3D printing.  When I want one of something, I'll go old-skool.

Great to hear that Ken is updating us on the newer techniques with Shapeways and SketchUp!

Rene'

Scratchbuilding

I actually submitted my passenger car for judging at the NMRA national convention a couple of years ago to explore this very question.  They gave me part marks, and later the AP went back and forth on whether 3D printed parts count as scratchbuilding if you are not the machine operator (a perplexing and not particularly clear demarcation in this world of virtual services).  Currently, I think they're accepting them as scratchbuilt even if you never see the machine.

For me, the question of scratchbuilding is not about the tools or skills a modeller has employed, but one of design.  A model is scratchbuilt if the builder is the one who figured out how to solve all the problems posed by the model.

But really, who cares?  I just like to see nice models.

Next steps on your cab

Ken,

Thanks for the comments, Rene

I agree, Rene.  However, because of my work schedule, I have a lot of down time spent in a hotel room, and it is far easier to carry and use a computer than it is to carry and work on a model project.  I've tried both.  So, for me, the advent of digital media and 3D printing means I can get some useful modeling time in, even when I'm not at home.

I'm the same way, provided that I have the skills to make whatever the part is.  There are times when I can draw and print a part which is simply beyond my ability to make.  On the other hand, something like the cab above is clearly within my ability - I've scratchbuilt several in brass and styrene.  I chose to print one as a test.  I can compare it directly to what I am capable of building, and evaluate the cost vs. time equation.

I did both for the same reason.  I printed the decorate half-round bead which surrounds the door & window opening as part of the doors and windows, rather than as part of the cab wall.  That way the cab wall is almost completely free of detail (there is a decorative bead around the edge which I could not practically make a separate part), which should make it much easier to clean up.

I also left the doors & windows separate to allow a choice.  I modeled open and closed versions of them, which will allow me to mix and match to suit a specific model.  It only cost a few cents more to include multiple versions of the doors and windows, and this way I don't have to make separate drawings for all of the possible permutations.

I have noticed that many 3D printed models, especially those with a lot of detail on what is supposed to be a flat surface, look pretty obviously 3D printed.  The layering is frequently visible and distracts from the overall appearance.  One of my goals with any model is to make it look as close to the prototype as possible, rather than to be obviously a model built using ABC method and XYZ material.  For 3D printing, that means designing the print so that it can be sanded smooth if needed.

On this cab, there are so many rivets that it would be virtually impossible to clean up the surfaces around them.  The time invested in doing so, and in avoiding accidentally removing the rivets in the process, would be far greater than the time spent applying rivet decals.  Since the goal is to have a good model, rather than to show off 3D printing, I will happily mix materials and methods as I see fit.

Another advantage of 3D printing

One of the things I really like about Shapeways is that they make it easy to sell your models.  Take this cab, for example.  Once I am satisfied that it's worth selling, I can simply click a button on the web site and let people buy it.  I select how much markup I want to make of of each one, and let it go.  When someone else wants one, they go on the web site, order it, and Shapeways prints and ships it.  They take the money, and send the markup to me

(And yes, I do pay a nominal fee to MRH for the right to talk about such products here.  So far, I've paid a lot more than I've made, but it supports MRH, and that's a good thing.)

I've done this with a few items already, and I hope to do so with this cab.  While I'm not likely to get rich selling model railroad parts, it's a nice way to add a couple dollars to my modeling budget.  It also means that people who do not have the time, skill, or tools to design their own 3D models can still get the parts they need.  One of my future projects (for which this cab is a test run) is to design an entire HO scale locomotive to be printed at Shapeways.  Making the locomotive-shaped blob isn't difficult - the challenge is in figuring out how to do it in such a way that the resulting model is of high quality.

Rivets

This thread is my first ever exposure to 3D, and I'm enjoying it. Gotta ask:

You mention "time spent applying rivet decals." Are the decals 3D printed? If that can't happen, could rows of rivets be 3D printed separately, like your window detail, and fitted into channels that are part of the body? It seems this could be fairly easy to clean up.

I look forward to your article.

DrJolS 

DrJoIS

Take a look at the MRH sponsors page for Archer Decals.  They do rivet decals that have some 3D quality to them.  Very realistic for rivet detail.

Rivets

Water slide resin rivets work great. IF .. you were to use say .. Tichy rivets you can 3D print a guide hole. that allows you to sand the print and when finished run an appropriate size drill through the guide holes to fit the rivet shank. I've done that several times successfully with prints.

Here's a turbine i had printed at Shapeways. The rivet locations were printed like I said with 'guide holes' .. so all I had to do was clean them up a bit before inserting the rivets. I didn't even sand the turbine body as it was fine as it was (IMO) .. FUD often prints very well.

"The problem is that there is

"The problem is that there is such a wide selection, and the parts are so specific, that you have to maintain a huge inventory to satisfy more than one or two modelers.  It costs a lot of money to manufacturer a run of those parts, and then distribute and display them "

   I think the solution is for the detail part companies to sell online so shops don't have to stock the parts. Online ordering and cheap mailing due to the light weight should make it quite handy. I'd much rather buy molded parts than shapeways parts since th equality is higher for most types of parts. Shapeways is great for one off or experimental stuff but the price for part and the surface finish is way behind  good molded parts.  Ideally there would be online detail molded part suppliers, online decal printers, online photo etchers, and shapeways all available on demand. ..DaveB

Online photo etch

DaveB,

You should have a look at emachineshop.com.  I've never used their service, but they indicate they do photo chemical milling (perhaps not brass - one would have to check).  They have a similar model to Shapeways in that their value proposition seems to be making it super-easy to get the thing you need.  The price or quality might not be the same as if you go to someone who does only photo-etching (again, I don't know), but they've put a ton of investment into convenience.

Cheers,

Rene'

Rivet decals

Yes, the rivet decals are 3D resin bumps on a decal sheet. You can also get louvers, weld seams ... all kinds things.

See the August 2011 issue of MRH:

http://issuu.com/mr-hobbyist/docs/mrh11-08-aug2011-ol?viewMode=presentation&mode=embed

And Archer Fine Transfers, an MRH Sponsor, sells these ...

Thank you, skiloff, eTraxx, and joef

Now I know two methods to get good looking rivets.

My brain has been ossified around Bill Clouser's die and pencil method of 50+ years ago. Now I expect to make the new knowledge work.

Thank you all,

DrJolS