the future?

[jimmo]

I tried toothpaste (Colgate) before. Didn't seem to do much at all except make the parts smell minty fresh.

Gels don't work Chris, plus you gotta scrub 'em a bit with a toothbrush.

[DKS]

Gels don't work Chris, plus you gotta scrub 'em a bit with a toothbrush.

And floss, too.

[Chris333]

It was paste, but just plain Colgate and no whitening, tarter, etc. I also bought an air eraser that didn't seem to do anything to FUD, but it was hard to see white grit hitting clear white plastic.

[randgust]

It's not just the process, and the orientation, but the lack of control of orientation, and that's purely a Shapeways call from what I can see on the axis.  I have printed a small batch of my little ATSF phone booths, and none of them appear to have been done exactly the same.  You really have to study it to see it, but its there.  The resolution isn't the issue, heck, my phone booth has a perfectly legible doorknob on it.

What makes this so difficult for 'production' is that you never know what your customer is going to get, and that's why for now, I'm limiting myself to structures, etc. for sale.  Haven't given up yet on using the concept for resin masters, but I've been advocating the brass add-ons from the start as that's what I already do in resin - the entire CF7 project was developed that way.  If you can't solve handrails you may as well not bother, and metal etching can beat that easily.

[Catt]

As much flak that Shapeways gets has to do with orientation of the printed part they are making for their "CUSTOMER" you would think that Shapeways would smarten up and start listening.

[bbussey]

There are additional issues with FUD items from Shapeways, but they've been discussed ad nauseam in other threads over the past three years.  Bottom line: find a rapid prototype shop with ProJet3000+ equipment that will render the parts as desired.

[DKS]

If you can't solve handrails you may as well not bother, and metal etching can beat that easily.

And herein lies another point not stressed often enough, IMO: too many 3D printing designers try to do every tiny little bit in FUD, all as one piece. This creates at least five issues: the "shadow effect" and other artifacts; problems when trying to clean up the model--working around all of those handrails and such can be a risky chore; oversize parts that look fat and ugly because of the minimum rendering dimensions; difficulty painting; and problems just handling the model--the delicate parts are easily broken. The best-looking models leverage the advantages of different media. Sure, it does mean assembly is required (aw, shucks, do I have to?), but it also means stronger, finer, much better-looking details.

[ljudice]

As much flak that Shapeways gets has to do with orientation of the printed part they are making for their "CUSTOMER" you would think that Shapeways would smarten up and start listening.

There are some people here who have done amazing stuff with Shapeways - but bear in mind the vast majority of their business is in arts/crafts....  In a cat figurine these printing issues may not be an issue.  They may even add texture which is desirable.  They were funded by the same VC that funded Etsy and their message to investors is that they are the "Etsy" of the future.  Emphasis on selling a lot of stuff,  not being able to reproduce the contours of some obscure locomotive perfectly.

As Bryan says, if you need perfect, near injection molded quality there are other options out there there to pursue.

- Lou

[Iain]

This has been my approach as of late.  Yes it's awesome that I could print ALL of the piping and railings on my WM 2-8-0, but they had a negative effect on other surfaces around them.  When you take stock of the surprisingly large number of etched brass parts that are out there, the best thing to do is to draft up the mounting holes for those parts instead of trying to print them.  The baby faced baldwin that I am working on is absolutely minimal in regards to small details.  No grills, hand rails, or sunshades.  All will have their mounting locations marked, but these items will be added by the modeler at will

Actually, I drew up all the etched parts, too.  That's why I don't have the proper cab, I didn't take the folds into account so my cab was too wide.

[nscaleSPF2]

Ok, so here is how to overcome the orientation issue with Shapeways.  I stumbled on this while developing the Pennsy signal head.  The entire topic is here:  https://www.therailwire.net/forum/index.php?topic=28968.0  Don't be put off by some of the early posts.  The final product turned out ok, if not perfect.

The reason that the technicians at Shapeways reorient your part, is because they are trying to print as many parts at the same time as possible, and will place your part in any convenient orientation for them.  If it is long and narrow, like a car side, then it could be placed along any edge, for example.

Here is the trick.  Construct the model so that it takes up a cube of space.  The technician then has no incentive to reorient your model.  In my case, I used a 2x2x2 model, consisting of 8 signal heads, but other shapes may require a different number of parts.  The parts do not need to be connected with sprues.  In the model, the tops of the parts do need to be oriented up. 

Have faith.  Every problem has a solution. 

Except for the printer's resolution.  Silk purse from a sow's ear, and all that.

Oh, yeah.  This was my first attempt to design a part in 3D.

Jim

[peteski]

Here is the trick.  Construct the model so that it takes up a cube of space.  The technician then has no incentive to reorient your model.  In my case, I used a 2x2x2 model, consisting of 8 signal heads, but other shapes may require a different number of parts.  The parts do not need to be connected with sprues.  In the model, the tops of the parts do need to be oriented up. 

Jim, I remember that thread. But out of that array, how many parts end up being usable? The ones on top and maybe on the bottom?  But all the other printouts will have undesirable artifacts (and are unusable)?

[bbussey]

... Here is the trick.  Construct the model so that it takes up a cube of space.  The technician then has no incentive to reorient your model.  In my case, I used a 2x2x2 model, consisting of 8 signal heads, but other shapes may require a different number of parts.  The parts do not need to be connected with sprues.  In the model, the tops of the parts do need to be oriented up...

Except that it gets VERY expensive is you're rendering something like passenger car sides and have to submit a 6³" part.  I'd rather use a contractor that will output at a tighter resolution and in the orientation I desire.

[PiperguyUMD]

Speaking of using parts to finish 3D printed shells....  I bought some of this stuff a few days ago to try as grills on some of my Baldwins, and it is AWESOME!

http://www.ebay.com/itm/300909479310?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649

Its a little tricky to work with because unlike fabric, it flexes under pressure.  However, it is easily cut with a sharp exacto knife!

[Dave Schneider]

Again,  I am not trying to turn this into a Shapeways rant. They have their business model and that is their right. I understand that they are trying to fill a print cube to maximize their capacity. I am disappointed that they didn't look at my file and choose to put it into their print cube in a way that would give the best result. They did offer to add a note requesting orientation if I chose to reorder. I am deferring that at the moment and researching other options that allow orientation to be chosen and possibly at a higher resolution.

Best wishes, Dave

[cnw mike]

If you REALLY want it, and 3d printing seems the only way, sometimes you have to eat a few parts until you get the orientation you want. I know, sucks, but they do make it easy to print stuff.