3D Modeling; The Future

[ToniBabelony]

Cool, got any classified informations?

 

Btw, don't you also need a license from the train maker?

 

There is no classified information, as there is nothing decided yet, but I can perhaps reveal these four letters: 'KRDE'.

 

The license depends on the maker of the rolling stock I think. Some may not even care at all. I'll probably leave that to the expert once I'm confident enough to put my products on sale.

[tossedman]

Sounds great Toni. Just watch that your hobby doesn't become too much of a job that it's no fun any more. I've seen this happen quite a few times in the world of radio controlled trucks.

 

Cheers,

 

Todd

[katoftw]

Are these models being designed to use with a specific motor unit?  Other than the middle double doors, Toni's train look very similar to NDCs/Y-DC125s/anything from vol.19 railway collection?

 

Edited April 7, 2015 by katoftw

[ToniBabelony]

Sounds great Toni. Just watch that your hobby doesn't become too much of a job that it's no fun any more. I've seen this happen quite a few times in the world of radio controlled trucks.

 

Thanks. I think this will be nice, even if it becomes a job. I already have one hobby (illustrations and comics) which became my profession, so I have to do taxes and all with that as well. It didn't really take the fun away honestly ;)

 

Are these models being designed to use with a specific motor unit?  Other than the middle double doors, Toni's train look very similar to NDCs/Y-DC125s/anything from vol.19 railway collection?

 

Yes, these will use the Tomytec TM-08 motor unit, but will be compatible with any 20 meter Tomytec unit anyway. The Railway Collection 19 trains are 18 meters and have a slightly higher body design than contemporary Kanto Railway trains, but are all Niigata built.

 

A little update on the preparations:

 

 

Changed the layout to save some space and adjusted the inside of the body to be able to clip into a Tomytec power unit.

[ToniBabelony]

Yeeeeah!

 

DMM accepted my entry (no mistakes in the data) and has calculated the printing price at ¥4,747 in high-quality transparent acrylic.

 

 

I see there are also options for printing in other materials, like silver, gold and other metals. My model is too big for this option, but it certainly is interesting as a future option (but damn expensive). This model can however be printed in Inconel (a kind of nickel) and Maraging Steel (nickel and copper) for 10x the acrylic price. Maybe this is good for making moulds for production models.

[kvp]

The printing price seems acceptable for very small scale production. For moulding, the usual process is to get a base material (steel/aluminium/copper) and a cnc machine, that cuts the mould from two parts. Flat items (like the kit above) are much cheaper and easier to get right and copper moulds are the cheapest, but only last for a couple of hundred items. The next, better version is aluminium, often used for larger production runs that don't need a durable steel tool. For prototyping and small runs even large companies (like lego) use the cnc and copper method. Getting a proper mould is much better than trying to use rubber forms and resin, not to mention the result is a high quality (greenmax grade) kit, instead of a fragile and hard to glue resin cast.

[ToniBabelony]

The printing price seems acceptable for very small scale production. For moulding, the usual process is to get a base material (steel/aluminium/copper) and a cnc machine, that cuts the mould from two parts. Flat items (like the kit above) are much cheaper and easier to get right and copper moulds are the cheapest, but only last for a couple of hundred items. The next, better version is aluminium, often used for larger production runs that don't need a durable steel tool. For prototyping and small runs even large companies (like lego) use the cnc and copper method. Getting a proper mould is much better than trying to use rubber forms and resin, not to mention the result is a high quality (greenmax grade) kit, instead of a fragile and hard to glue resin cast.

 

That's interesting to hear. I can see why mould needs to be durable, but for low production runs, I think having 3D printing is ideal. You can create moulds with it and do low runs of a few hundred productions before the quality of the output drops. After there is still demand, you can put out a new mould and create more.

 

If this turns out good, I'm also thinking about creating coloured 3D prints, as they also seem viable and very cheap. The problem is that I don't know what the detail quality is compared to acrylic prints (I presume low). The upside of coloured printing is that you can already provide the models with complete livery, numberings, etc., so only assembly is required. Maybe even assembly with click-in, so there is absolutely no need for glue anywhere!

[kvp]

 

If this turns out good, I'm also thinking about creating coloured 3D prints, as they also seem viable and very cheap. The problem is that I don't know what the detail quality is compared to acrylic prints (I presume low)

Shapeways has colored prints, but it looks more like colored sandstone than plastic. There is a way to print in color, but you either have to choose a single color and use that (similar to basic injection moulding) or there could be alignement problems on color boundaries as the machine switches heads. Imho the best way is to print in a single base color, then airbrush the resulting parts with a stencil mask. This is how the greenmax and tomytec precolored kits are made. You can even do pad printing on them later.

 

A relatively new technology is to use a white base material and inkjet paint with UV activated paint. This allows printing even on curved surfaces and combines airbrush and pad printing into a single step. The result is durable and fully UV resistant, while even allowing photographs to be printed on almost any surface.

 

 

Maybe even assembly with click-in, so there is absolutely no need for glue anywhere!

This mostly works for Tomytec, Bandai and even Maerklin. The trick is to design each part to be clickable, print main parts in different base colors, then stencil paint and pad print them. This usually involves printing the window strips with a clear transparent material. Pad printing can be replaced with rub on decals for a cheaper technology and easier to change car numbers, but many people (for example me) don't really like it.

[cteno4]

a lot of the colored printing is powder and binder printing. works well for larger scales but like kvp says it tends to look like sandstone. lots of architecture firms are buying these printers as they are fast and pretty looking to do architectural models for clients. but they have none of the charm of old chipboard models that give that sort of idea of the design, i find these new models too literal and lack any imagination and actually look a bit clunky and heavy as a result. but when clients pay like $5k for these if you can whip it out in your printer for a few hundred dollars print run from your cad plans, its become a cash cow!

 

yeah the 3D inkjet printing is pretty neat. the print is modeled over the 3d surface to get the right wrapping of the image to the uneven surface. we have a thread somewhere of it even being used to paint the side of the busses and trains!

 

tabs are great but really require great engineering and moulding to work well. bandai i think does the best I've seen in any models with their 3 stage fit. first gets you lined up, second its snug, then the third it clips into a really tight fit. they also do nice rabit joints to make the seams be sealed and supported well.

 

turning this into a full production process may be quite the chore! as todd said, careful not to suck the fun out of this unless you think this can really be a business for you that you want to do!

 

will be interested in seeing how your first model turns out!

 

cheers

 

jeff

[ToniBabelony]

UV activated painting/3D inkjet printing seems the way to go in the future, but creating painting moulds based on the original print (perfectly matching every line of the to be painted object), has been on my mind as well. Printing on the object directly sounds awesome to me, since I'm really not a fan of decals in any way. If I find some reasonable service for this, I'll definitely consider it for my models.

 

I took a look at some colour prints and they look too rough to work with in N-gauge. It looks cheap (it is about half the price of acrylic printing) and far too coarse.

 

Anyway, I just ordered my first print from DMM Make; a high-detail acrylic print. After I'm satisfied with this, I'd like to order a non-kit like model, where the car body is already assembled and only the roof and other details need to be added. I wonder what the difference in quality between a kit and fully printed model would be. I have the feeling the fully printed 3D model will have warping issues...

 

In the attachment below, I have included the very simple and rough dimensions of a Tomytec TM08R power unit in .stl (basic 3D format) for those who are interested.

TM08.stl.zip

[katoftw]

 

Far off this stage?

[HantuBlauLOL]

ありがとうございます！

 

Can you also make a Kato B train chassis 11-105 in stl too?

[ToniBabelony]

c1180230-img236x280-1417943142bloneu12352.jpg

 

Far off this stage?

 

Not really, but I'd first like to experiment a bit before I decide to go for the classic flat-spray mould look or the modern 3D look. xD

 

ありがとうございます！

 

Can you also make a Kato B train chassis 11-105 in stl too?

 

I don't own any Kato B-train chassis, but I do own Bandai B-train chassis, which I might do later.

[Atomsk]

After I'm satisfied with this, I'd like to order a non-kit like model, where the car body is already assembled and only the roof and other details need to be added. I wonder what the difference in quality between a kit and fully printed model would be. I have the feeling the fully printed 3D model will have warping issues...

Warping is not a big issue with passenger cars.  It's more a problem with "2D" models.  I've only had warping problems with my flat car. 

 

I only had passenger car models warp on me when I made the walls too thin.  On some of the larger models, I put temporary 1.5mm "struts" between the bottom edges of the body ( I usually model the floor as a separate piece ).  I can remove these easily, when assembling the body.  Cars with interior partitions, like my Pullman sleepers, should be rock solid, with no additional support.

 

The acrylic material is somewhat flexible, so if a flat part (like a floor) warps, you can still straighten it out during assembly.

[HantuBlauLOL]

Atomsk, how much they charged you for making your flatcar?

 

I think 18$ for selling price is already cheap enough for 3d printed model designed by someone..

 

 

I'm thinking about converting game objects to 3d printable format. Open freight cars seems to suits well. Freeware contents of course.

[Atomsk]

Atomsk, how much they charged you for making your flatcar?

 

I think 18$ for selling price is already cheap enough for 3d printed model designed by someone..

 

I don't mark-up the models I sell on Shapeways.  $18 is the cost to print that car.

 

Shapeways charges by the volume of material in each model, so the cost is based on the size of the model you want printed.  They list all of the available materials, and their costs, on this page.

[kvp]

Is it enclosing volume or material volume?

[cteno4]

It's a combo of the actual material volume and the amount of space it takes in the printer and how it can be packed in with other jobs and a little labor. They have been refining it slowly as they go along. Last I change think was more emphasis on material volume, but not sure I remeber it right from their emails.

 

This is where the some of the art of 3d printing comes it to get the best quality for the shape and material via orientaion while packing and orienting best for pricing, time etc depending in where you are printing it.

 

http://www.shapeways.com/support/pricing/?li=t_menu

 

Jeff

[Atomsk]

Is it enclosing volume or material volume?

It depends on the material.  They always charge you for material volume.  For some materials, they also charge you "machine space", though they can put smaller parts inside of larger "open" parts.

 

They also charge a "setup fee" for each "model".  If you want to print 10x of an identical model, they will charge you 10x the setup fee, but if you upload a model with 10 identical parts, you only get charged 1x setup fee.

Edited April 12, 2015 by Atomsk

[ToniBabelony]

They also charge a "setup fee" for each "model".  If you want to print 10x of an identical model, they will charge you 10x the setup fee, but if you upload a model with 10 identical parts, you only get charged 1x setup fee.

 

I'm not a Shapeways user (heck, we have DMM and a bunch of other companies in Japan), but that's a good tip! I'll probably remain with my 'flat' models in that case, as I'd like to print a bunch of them in one go at some time. This way full train sets (2+ cars) will become more viable/affordable as well! :)

[HantuBlauLOL]

I don't mark-up the models I sell on Shapeways.  $18 is the cost to print that car.

 

Shapeways charges by the volume of material in each model, so the cost is based on the size of the model you want printed.  They list all of the available materials, and their costs, on this page.

Wow, really you don't charge someone for buying your designed stuff? Cool!

[Atomsk]

Wow, really you don't charge someone for buying your designed stuff? Cool!

Yes, since I know I can't make any serious money from the models I upload, it doesn't seem worth the hassle to collect the money from a markup.  I only sell one or two models a month.

 

I'm more interested in bragging rights (the ability to say "look here, I made this!!!") online, or to members of my Ttrak club.

[ToniBabelony]

Got an email from DMM Make yesterday that my model has been printed and is on its way here. I barely can contain my excitement! Let's see how it has turned out and if the data needs adjustments before I continue to develop this model into various other Kantō Railway trains. The next step is to go for the 2-car trains (2100 and 2300 Types) and the 2000 Type that runs on the Ryūgasaki line.

 

Next to these models, I'd like to go a bit into the realm of trams, as I'd like to have a nice and accurate model of the old Sneltram Utrecht Nieuwegein (SUN):

 

 

All there is, is a rather crude brass model in N-gauge, but luckily, I'll be getting sources to quite accurate plans to it soon! After that, I'd like to create a Stadtbahn B80 from the KVB (Cologne) and/or SWB (Bonn), as both the SUN and B80 carry many great memories for me.

[Densha]

Oh god, going into Dutch trams is dangerous. :P

I need to stay away from that; I already have a small collection in H0 scale.

 

How are you going to power it though? I can't think of a Tomytec chassis that has three bogies of which one a jacobs bogie.

 

Btw, this guy on Shapeways makes Dutch trains and stuff, but these particularly may be of interest to you: http://www.shapeways.com/shops/SpoorObjecten?section=11.+Chassis+-+%28All%29&s=0

These chassis are built to support Tomytec power units.

[ToniBabelony]

For the Dutch and German tram projects I don't need a jacobs-bogie set, as I can make the chassis myself. The most convenient and quickest method would be to have a Tomytec chassis with the correct bogie and axle distance for the first half and middle, and then add an unpowered bogie on the other end.

 

If I can't find that, I'll think of a different solution. That would probably be to have the centre bogie powered with the newly released HM-01, add weights and custom bogies on the front ends. The downside is though that there will be no flywheel with this setup that can smoothen out the ride.

 

In the end, I'd like all the models I create to be semi-plug-and-play, where only painting and adding windows is required. No cutting up poor donor chassis' and adding extra parts.