This may get me back to Shapeways

[HuskerN]

Ok, I will let you in on my secrets of 3d printing.  There really isn't any, that I know of?

I think I am getting good results because I am designing relatively small models, versus a locomotive shell for example which is larger than even my largest model, my livestock trailer.  I seem to have pretty good consistency with my printing results from Shapeways, but there is variance due to printing orientation, and I have had to get a couple items reprinted due to quality issues.  Overall, I am quite pleased with what I am seeing.  I use a white primer from a rattle can, and some models get a heavier prime than others to help give me a smooth finish.  Here is an example of a model that required virtually no clean up other than the solvent batch, and some minor toothbrush rubs and a quick file here and there.  A couple coats of primer, and it is ready for color.  If you look close, you can see some lines on the side of the cab, but some of that is actually the lower polygon count on the complex curves of my design rather than print quality.  And as far as my models lacking detail?  ok, maybe not rivets, but I am pushing the limits as far as detail goes on an N scale model that will fit in the palm of a child's hand.  Just check out the lug nuts on the front wheel hubs. 

[peteski]

Thanks for sharing HuskerN!  So, except for couple of bad printouts, all you have to do is to prime the models.  That is good to know.

Funny, if you look carefully, the sleeper cab (on both the blue and white models) shows what looks like wood grain. An artifact of 3D printing. But on the blue model, I had to look really hard to notice it.  I guess that what's acceptable for surface quality is really in the eyes of the beholder.  But neither of these models (or any of your models for that matter) looks anything like some of the bad examples of FUD printing I've seen posted here.

[3rdboxcar]

Thanks for sharing HuskerN!  So, except for couple of bad printouts, all you have to do is to prime the models.  That is good to know.

Funny, if you look carefully, the sleeper cab (on both the blue and white models) shows what looks like wood grain. An artifact of 3D printing. But on the blue model, I had to look really hard to notice it.  I guess that what's acceptable for surface quality is really in the eyes of the beholder.  But neither of these models (or any of your models for that matter) looks anything like some of the bad examples of FUD printing I've seen posted here.

Unlike Husker most of my models are full size coaching stock and do have the print lines on the sides sometimes worse than others which is all down to machine calibration, my NFI buses look really good with virtually no clean up required and my latest CNSM battery loco the same but there are very few large flat surfaces on these models but if you look through magnifying glass they are there.

My larger flat surfaces get a coat of tamiya fine surface primer then a carefull rub down with wet and dry.

Alexander

[bbussey]

I've had far more success using a combination of media and taking advantage of the strengths of all.  The ProJet material is outstanding for intricate detail, and poor on smooth flat/compound surfaces (regardless of using Shapeways and the medium-resolution FUD, or another ProJet contractor which provides the high-resolution XHD).  So I use etched brass or stainless for the smooth surfaces and either surface-mount the ProJet details or have them poke through cutouts in the etchings.







I also use the ProJet to make entire pilot models for verifying dimensions and for performance testing purposes.





One of the few exceptions for (nearly) complete models is the G32C gondola, because it has corrugated surfaces all-around inside and out and no smooth surfaces.  In this case, the etchings are the detail parts rather than the reverse.



So yes, the ProJet parts work for me, but not as being the only material used in finished models.

[robert3985]

I've had far more success using a combination of media and taking advantage of the strengths of all.  The ProJet material is outstanding for intricate detail, and poor on smooth flat/compound surfaces (regardless of using Shapeways and the medium-resolution FUD, or another ProJet contractor which provides the high-resolution XHD).  So I use etched brass or stainless for the smooth surfaces and either surface-mount the ProJet details or have them poke through cutouts in the etchings.



So yes, the ProJet parts work for me, but not as being the only material used in finished models.

Excellent post Brian.  I've come to the same conclusion, but haven't taken it to build finished models yet. 

I'd love to see high-res photos of your finished models using this technique.

[Philip H]

Bob,
That's the master for ESM's Keyeser Valley caboose shown in those photos.  While I believe they are sold out at ESM, I have one, and the finish is outstanding.  Now to finish assembling it.

[Sokramiketes]

  And as far as my models lacking detail?  ok, maybe not rivets, but I am pushing the limits as far as detail goes on an N scale model that will fit in the palm of a child's hand.  Just check out the lug nuts on the front wheel hubs. 

I should have used more words, and by saying they didn't have rivets I certainly didn't mean they lacked detail.

Railroad models tend to have lots of flat surfaces covered by regular formations of protruding details like rivets.  This makes it very hard to hide striations in RP models through surface prep... either sanding or heavy primer.

The vehicles are great because of the compound smooth surfaces.  Thick primer or judicious sanding usually doesn't affect the overall shape! 

And hub caps and grills and other areas of small footprint, high detail, turn out great in RP... usually not needing much prep beyond wax cleaning. 

[Scottl]

Rivets are so 20th century!

[HuskerN]

Mike, no worries about rivet/detail comments.  I do admit that when I get a Shapeways package from the UPS delivery driver, there is a sense of anxiety about what is inside until I open it up and view what is inside.  To date, the success rate of models is good, but not perfect, and it is usually not about whether the model is suitable for finishing, but rather which way was it printed, and how clean is it, and how much work might it take to get it ready for primer. 

I just wish they didn't cost so dang much to print, so I could print a dozen models, and pick out the best 6 for finishing. 

It is still a great means to fill a major void in our scale, but I agree that 3d printing has some evolving to do.  I consider myself pretty picky, like many other N scalers, and I just hope that printing will soon be able to match the quality we are used to with manufactured models.

I don't know a lot about the technology, but I really hope that other printing services can begin to offer some competition.  This could change our hobby significantly.  Oh, but you guys still need to learn CAD.  No way around that.   

HuskerN
www.nscaleaddiction.blogspot.com
www.shapeways.com/shops/NScaleAddict

[bbussey]

There are a lot of us here who know 3D modeling.  Mike works in SolidWorks, as do I.  We're just waiting for 3D printing technology to evolve further.

[HuskerN]

So what you are saying is that I should go back to model railroading and layout building, and come back to the 3d printing when technology evolves?  That actually sounds good considering I haven't operated a train for almost a year.  As a matter of fact, I sold my DCC system with the Marias Pass and never replaced it yet.  Where are my priorities?  I need help...

HuskerN

[bbussey]

You should keep designing truck models and having them 3D-printed if that makes you happy.

[Alwyn Cutmore]

The talk of 3D printing is a very emotive subject and unless you are happy with what you get it can be a waste of time. 

Recently there have been a numbers of discussions about other topics when the subject of 3D modelling was broached. 3D modelling is part and parcel of a manufacturing system called "Rapid Prototyping".

The design and drawing of products can then be seen in 3D to ensure that the drawing you do is what you want. It was never designed as a process for multiple products as the process by its very nature is slow and costly. Good rapid prototypers can turn out a product that is as polished as a injection moulded product without the layering showing. That is achieved in two ways, sanding and polishing the external surfaces so that you get a mirrored finish or by spraying with a spray on filler and then sanding back. The biggest problem with this process is that you can easily loose detail. 

There is no getting away from it HuskerN is producing some wonderful items but even my poor old eyes can see the layering  on the tractor in plain material. The item could be made that much better if it was devoid of all the fine detail, which could be added later from lost wax castings or white metal castings, and the body was polished to a mirror finish by either of the aforementioned techniques.

That other company who as far I am concerned is unmentionable have done a great job with marketing. I looked at the system they used and was not impressed. Their system and marketing is designed on the principle "You pay for what you get". The more you pay the better the product.

I will be doing a one piece kit later and the RP will be done in Australia. I can have the RP done in 16 micron or 32 micron layers and it can be done in 3 different finished. Unfortunately I am not a cad operator so the drawings will be done by a professional draughtsman.  If I can get the patterns to turn out as good as no 2 and 3 locos in the attached photo I will be happy. These were made using the same system and with very little cleanup before being cast in Urethane.



No 1 loco is just as good but the focus was not the greatest

I would encourage all those using 3D RP to persist but just remember it is what it is and until the technology grows and costs come down those guys with the printers are the ones really smiling.



Now this is a FUD model. Time wise it is a waste of it. If I received that I would hit it with a hammer.  I will say no more as I think that between the two pic the difference is clearly evident.  Mind you I will say upfront I have never ever bought any FUD items because the photos of the sample items have always turned me off.

Regards

Al

[robert3985]

I think that FUD "shells" and larger items are probably not going to cut the mustard for a while coming out of Shapeways.  The Perfactory process is, however, still RP...but has a much nicer finish and interprets the 3D model differently so that "layers" are much less obvious, and in some cases are not obvious at all.

But, ya gets what ya pays for and Perfactory prints are expensive and nobody that I know uses anything close to Shapeways marketing strategy to generate profits both for the company and for the designers who use them.  I actually think it's a great idea, but the big, black fly in the ointment is the quality of the print on larger items.

However, I do purchase small parts from Panamint Models and Eric Cox who uses Shapeways for his 19th century parts and car kits.  The part that I LOVE is the 1890's CB&Q 5' wooden framed passenger truck, which the UP called the "Q" truck and was under Non-Standard wooden cabooses, Harriman Era CA wooden cabooses (MT Wooden Caboose), CA-1's, CA-3's and CA-4's because of their ride quality.  Some of these trucks made it into the 70's on several of the last wooden cabooses.

Here's a photo that some of you have seen before of these lovely trucks, available in both pin and screw mounts as well as offset for IMR CA-3's and CA-4's:


Here's another with them mounted on a virtually stock MTL "Wooden Caboose" (a really good model of a Harriman Era CA caboose), with MTL Z-scale couplers. They totally change the character of the car IMO.


The reason I posted these photos is to illustrate that not all FUD prints are garbage from Shapeways.  The level of detail on these prints are much more than could be done with injection molding without making these a kit.  The included brake shoes and hangars on the ends of the trucks add a level of detail that isn't available on MTL or BLMA or Bachmann Old-Timer trucks (all of which I have $hit-canned since the Panamint Models trucks appeared) or any other trucks injection molded or investment cast that I've seen.

I believe that "kits" instead of "shells" with two ends, two sides and a roof that get assembled into a car or engine body would solve much of the orientation problems, making the prints with flat sides all separate and oriented the same way either completely acceptable or complete junk...and the subcontractors could easily see the "junk" and re-print without shipping.

I haven't given that concept a try yet, but I plan on it soon, just to see if it works 'cause I'm getting tired of spending 20+ hours kitbashing every UP CA-1 that I need since I need 15 or so. If RP doesn't work, then resin casting will, along with etched ladders, steps and other platform furniture.  It's pretty obvious that prints with five sides (two sides, two ends and a roof) will have, at best, only two sides that are oriented optimally.  That leaves three or four sides that won't look so pretty in the classic "shell" protocol...every time.

Brian's photos got me really thinking along a slightly different angle, so I'm going to play around with my 3D model and maybe use more etched surfaces than I'd planned on.

Verrrrry interesting.  This is what TRW is all about.  My hat's off to the asshats here!! 

[Chris333]

I have these made with FUD:

They are small and sit on a single Z gauge truck. There is slight stepping, but I'm happy with it.
http://www.shapeways.com/model/466148/nn3-v-tipper-x-16.html
BTW there is no mark up on the price, they are that price because all 16 are attached to a "parts tree" that is also printed.

And I have the exact same trucks that Bob showed and I'm happy with them. That hopper shown above, not so much.