...

I think it looks quite nice indeed!!

A little working on the masters as they come off with sandpaper, or enough paint on those surfaces, and it'll be smooth.

Plasticard

Damian,

I'm not sure how difficult this would be from the design standpoint, but for large flat surfaces (like the top of that model), how much of a challenge would it be to design an inset into the casting? 

My thought is you could than place a thin piece of styrene in the inset, resulting in an already smooth surface. A touch of green stuff around the edges, and voila.

Progress of the Digital Model

The shell doesn't have any detail along the sides or the roof because I hadn't gotten that far when I sent the model to be 3D printed.  I had the shell made so I could compare the shape of the nose to that of the real thing.

I'll since continued working on the digital model.  Here's my current state of progress...

Right now, I'm working on the front and back pilots.  After that I'll finish off the sides.  I plan to do the roof last.

Thanks,

Damin.

T-pillars

You may consider putting a vertical column along the inside of the shell on either side, where there is room, and if it's needed, to provide inner support so the wall doesn't bow in.

Would you be able to divulge the scale, cost and what program you are using?  It's looking really very nice.

re: t-Pillars

Hi Benny,

It's N scale.  I'm designing the shell to fit a Kato F40PH mechanism.  I'm not too worried about bracing the interior because the finished design will fit snuggly over the mechanism's split-frame.  This version was just a test, with the foamcore block acting as a substitute for the mechanism. 

As for software, I'm using Autodesk Maya.  I export the model from Maya as a VRML file, clean it up in a program called VrmlPad, and upload it to Shapeways.

Thanks,

Damin.

Would you be able to divulge the scale, cost

I see the scale is N.

I also am wondering cost to you from Shapeways.

re: Cost

Hi JustSteve,

Shapeways charges based material and volume.  The material I choose for this shell was described as "Frosted Ultra-Detail" (FUD).  The price is $5 setup plus $3.49 per cubic centimeter.  The price for this shell was $20.38.

Thanks,

Damin.

$130.00 in HO

So up-scaling to HO would cost 8 times that, all things being equal, or around $130.00*.  Hmm.  Pricey little piece.  The grain might be more tolerable in the larger scale, but fine details might be compromised.  Seems to me like it's not quite ready for prime time yet, especially in the larger scales.

*reread the previous post and saw I had miscalculated somewhat, had figured $160.00 the first time.

...

If the setup fee is included in that price, then the cubic volume as determined by the cost is 4.4068ccs. For an HO sized model, you'd need 27.4ccs of material, which comes out to $95 and change for the material and $5.00 for the setup fee, hence $100.

One or two, it makes sense when you compare this to what a unit costs if you were to make it out of either Details Associates, Cannon, or Railflyer components.  For compound curves like the nose, it makes complete sense to do this now, even if you just make the nose and then build the rest of the body from strip and sheet stock.  10 or 20, and it may suddenly make tons of sense to simply get a 3D printer for $2000-$3000 and do it at home.

We live in a privileged time!!!

where's my piggybank?

I don't know which magic hat you pulled that volume out of, but that's still $100 a car or locomotive.  For some folks who resent the loss of Blue Box, that would be a small (large?) fortune.  Even I would not want to finance a fleet of ore cars or bulkhead flatcars at that rate.

Possible and practical are not the same thing.

...

I deduced that figure using a standard Volume Scale factor equation; the linear factor difference between N and HO is 1.83, the cubic difference is then 6.219, and based upon his shell cost [without the setup fee, which I assume he included in the total price], he has 4.4068ccs of material.  That comes out to 27.4ccs of material for his model in HO.

For a custom model/shell, this is really quite reasonable.  You only usually see such offerings on locomotives, anyways.

Either way, a couple years ago this was all just a pipe dream. 

Nowadays, Possible and Practical, in this market, are Rapidly becoming the same thing...

http://www.wired.com/design/2012/09/how-makerbots-replicator2-will-launch-era-of-desktop-manufacturing/all/

still vapourware

How about you get back to us once they are the same thing.

Blue Box is not a Fair Comparision

Hi Jorgen,

Prototype modeling is where you see the greatest potential for this type of technology.

If you're freelancing, then you have the freedom to use whatever you want.  So you can go with what is cheapest to obtain.

However if you're trying to duplicate a specific railroad, in a specific place and time with specific types of equipment, then you're NOT going to be able to populate your layout with "blue box" models (at least not without spend a lot of time and money kitbashing/redetailing).

In my case, I'm interested in modern-era commuter (Metra) and intercity (Amtrak) passenger operations in to, and out of, Chicago Union Station in N scale.  And in order to do that I going to need an F40PHM-2, a prototype for which no commercial model is available.

Bear in mind that this is just for the shell.  I still need to mount it on a mechanism (which will cost me between $75 to $100).  And then I'll need to install SMD LED lights and a DCC decoder (probably another $30 to $50).  And after that, I'll need to paint and decal it.

So this project will easily amount to over $200 per locomotive.  That's not cheap, but necessary for the prototype that I've chosen.

Thanks,

Damin.

Better way to do it

Make one master. Make a rubber mold. Make many castings for cheaper than having them printed. The tech is great for masters but costs are high for quanity.

Bernd

suitable use

I appreciate where you're coming from, and I would consider this approach for certain things.  The thing is that most of the people who are using this technology are in N or Z scale because the costs increase geometrically with scale.  HO is just out of reach for anyone without deep pockets, and frankly I'm not certain the finish is acceptable yet if you are going to spend that kind of money.  The fact is that if this were truly viable, people would be selling models made this way as kits.  It isn't, so they don't. 

Resin Kits vs 3D Printing

But some people are selling kits.  Obviously there are several sellers/designers on Shapeways.  But I recently received an email from Island Model Works, an established resin kit manufacturer, saying that they intend to shift some of their products over to Shapeways.

Thanks,

Damin.

Using this for complex masters

The immediate use I see

in HO is for parts that can be used to customize a kit or maybe too complex for basic scratch building. Truck side frames, roof details, whatever you need.

size matters

Making small parts is a different kettle of fish than making an entire car in HO.  It's the size that matters; anything the size of an HO car would pass the hundred dollar mark.  I think I heard somewhere that Railyard Models used rapid prototyping to make their masters, which is why they were so precise.  The parts would have had to be cleaned up before being used to make rubber molds, though.

Again, there are places where 3D printing is useful, but making HO car models is not that place.

Sweet Spot

HO cost isn't the scaled cubic volume of N

The cost is for the material used, not the volume of the product. Wall thickness is much the same. Many HO and OO trolley bodies are being done this way already and somewhere in tje $50-80 range, which is comparable with purchased molded resin equivalents.

Credit: Pictured linked to UK author's forum blog

As to just using a $2,000 home printer. The resolution, inside space and material choices for the very cheap hobby printers are far less, so it's not a fair comparison option. I suspect Shapeways machines are those professional ones in the $20,000-40,000 dollar range.  When hobby machines work dependably down to 0.001" "pixels", then I'll be getting one.

practical considerations

Our club has been discussing the possibilities of 3D printing for some time, and one of our members has even been working on getting masters made for one of his projects using that method.  The cost for a 3D printed master isn't such a big deal if you are making lots of copies.  However, I have seen some of the results of rapid prototyping, and frankly, I don't think it's good enough yet.  Combined with the cost, it's not ready for HO.

We would only consider this method for building fleet cars which are unlikely to be made commercially, like CP 40 foot coal gons.  For some reason, that prototype has slid under most modeler's radar, even though they were a pretty common car on CP and CN.  In any case, we would need dozens of them, around 50 or so.  Spending even $50.00 a car for something that still needs assembly, detailing, paint and decals is pretty rough, especially in such quantities.  I think we would be hoping for something more in the $30.00 range to make it viable.  If we only need one or two, we can scratchbuild or kitbash them.  For larger quantities, we can cast them and get away a lot cheaper than $50.00 for 3D printing.

The fact is, I can quite clearly see the possibilites of this technology, and I'm personally ready to embrace it, once it matures sufficiently.  It will come, but it's not here yet.

Thanks

Thanks Jurgen - like you, I'm very interested in this process (even more thanks to the OP for sharing) and what it will take to make it viable. I was under the presumption of < $30, so it's nice to hear that.

(P.S. - let me know if you do make those coal gonds - I'll order a couple from you!)

Shapeways

I am a fairly steady customer of Shapeways products and have been pleased with everything I've bought so far.This loco shell is small enough that several could be printed at the same time (a process that Shapeways encourages) which brings the price down considerably.

Granted my purchases have been in N and Z scale but they have been for things not readily available on the market or at affordable prices.The finish may or may not be perfect but usually a little sanding and priming will take care of that.But then again the N scale track bumpers I bought would not have a super fine finish anyway.

At $30.00 I would buy a pair of these shells when done.