3D production of parts

My problem with 3D printing is the creation of the 3D model.  There isn't currently an easy way for the non-CAD user to generate the models. 

John,

This is  the only limitation that has steered me away from making my own parts with a 3-D print machine. I can even get free set up time. But no CAD programs to run...

Is there a web site for shared data to run parts in O?

Auto Cad is a great tool, but to sit down and be proficient in it, would take weeks...

I'd be impressed if you can become proficient in AutoCAD in weeks!

Autodesk has a simplified, freely downloadable design application called 123D that can be used to create models for 3D printers.

---PCJ

Been a number of threads on this topic.....

I know 2 OGR forum members have 3D printers.

I looked at one....tried to learn the software...Decided it was taking too much of my time to learn...gave up.

And I consider myself very good at 2D artwork to produce model parts and have been for 12 years. This was built using laser cut parts....

3D artwork is a whole different animal.......it is what limits 3D printing. The 3D scanners can't scan inside things or figure out internal part shapes etc.

Let's see something that you've created.  3D modeling is not like drawing something in 2D, any sort of complex part takes a lot of work.  I know it's fashionable to state how easy something is, please read my signature wisdom.

I use UG and Gatia lightly. So it would not be hard to deep dive into AutoCad.... its one of the  Most friendliest CAD softwares to learn.

Agreed.......I tried to create a 3D model of a simple cube with two holes running at 90 degrees.......much simpler than ANY part need in O scale. Never got the two holes to line up 100% correctly......

Now 2D drawing......I feel I can tackle any project. ONE of my many....

Parts were drawn in 2D CorelDraw, cut and assembled.

This was EASY compared to the simple 3D cube I treid!!!

I posted a image of some railing in the scenery thread earlier today and got some "great candidate for 3D printing" so tonight I'm going to download SketchUp and if I can put something together I will seriously think about one of these:

http://printrbot.com/shop/simple-metal-kit/

I'm familiar with a lot of graphics software so hopefully some of that will translate. I do know some 3D guys - I've hired to do work for my professionally (broadcast/animation) and they've all been really expensive and a pain in the arse to work with honestly so I'm hesitant to reach out. Sometimes I feel like it's a struggle keeping current with all this junk and I dont really feel like I can cram yet a new software package in my head but this is for me and not work so I'll give it a whirl...

I do know is that there are tons of 3D files out there of objects that are free or can be purchased pretty inexpensively. 

    Check out this site to see what can be done with 3D printers for model trains.

    Someone from 3D Printer manufacturer- Solidoodle- sent it to me.

http://www.shapeways.com/minia.../model-trains?li=nav

John Knapp

Erie, not Eerie

No question it's a cool technology......the talent to create the 3D model to print is the big hold up.......I know a number of guys that are very good at it......but I have so many things going on I don't have time to learn more new!

Here's the vents for a Seaboard Air Line REA Express Boxcar I just received yesterday from Bill Lane of http://3dmodeltech.com/:

For this type car:

Stuff like this (can't be found anywhere else) is where 3D printing will make the difference.

Just think, you buy a boxcar from XXX company, then find out the model has the wrong doors for your RR.  Simple fix, have them made (as long as there's a file on hand for the software they use).

Broken part, get it made.  Missing part, BINGO!

The only thing Bill said to do was to remove them off the sprues and paint them asap, which I did this morning (not exactly asap but hey).

I'm going to compile a list of parts that have kept me from doing a number of modifications to existing rolling stock, hopefully I can get some of them made.

A guy showed up at the monthly meeting of our museum a few months ago with a couple of G gauge passenger platforms had made on a 3D printer for one of our members. He gave a talk on the possibilities of 3D printing for model railroading. One of the things he talked about was 3D scanning of an existing part or a handmade prototype to simplify the design phase. This guy wasn't quite ready for prime time - but he was on his way, and the G gauge passenger shelters were pretty good. 3D reproduction of parts is on its way - and given the unwillingness of MTH to stock parts, that's a good thing. 

Sketchup seems pretty easy to learn. In about an hour I've been able to create the solids needed to make the part of the concrete railing I'd posted about earlier that would be a pain to cut out by hand. The program appears to be geared toward architectural stuff. They have what they call Sketchup Warehouse for downloadable files. What's sort of crazy is the fact that there are downloadable files for a ton of rail stations. https://3dwarehouse.sketchup.com/

There are Sketchup templates for 3D printing - specifically Makerbot Replicator. They give you what appears to be the volume available for printing.

So far, making a simple geometric object looks pretty straightforward - how this relates to the whole printing process and whatever setup is necessary for that I have no idea.

You can extend the apps functions with all these extensions: https://extensions.sketchup.com/

Program: http://www.sketchup.com/

Tutorials: http://www.sketchup.com/learn

Years ago, so long ago that people were still impressed when you HAD a computer and a lot of people were terrified of them, I told Tom about my toaster-blender theory. When you want to use a kitchen appliance, most of the fine details are taken care of for you; you don't need to tell a toaster how thick your bread slice is or how high to pop it up. I told him then that when computers were as easy to use as a toaster, people would do a lot more with them. I remember when WordStar was THE word processor and people scoffed at Professional Write because you could print a simple page with default margins and 12-point Times New Roman without going through the twenty-seven steps (no, seriously) the original WordStar required to print a page.

That's where 3D printing software is right now. If you're good with software that requires constant thought and perpetual tweaking, you can do a whole lot...or you can be really annoyed really fast. We'll get to toaster status sooner than it seems we will, because we now have experience with what assumptions make software easier to use and how a user's preferences can speed up the process.

Home 3D scanners have indeed hit the market, but word is these first ones are as clunky and irritating as most technology is when it hits the bottom of the food chain where most of us can afford it. Being able to scan the object that isn't quite right, then modify it, will be critical for modeling as well as for full-size parts and molds. After all, what good is a scan of a worn-out bearing if you can't use it to make a good new one by modifying the worn areas?

It'll be a lot easier to take a scan of a boxcar door that is wrong in, say, the bracing, but is close to correct in length, width and overall thickness, then modify the cosmetic details. Eventually, there;ll be a database built up so we have scans of a lot of commonly modified parts.

did a search and found many places with software. 

even one that is free "Another browser based program using WebGL, this is free as long as you share your designs under Creative Commons"

I have a 3D printer (a first generation RepRap mendel) and have built some basic things with it for my various hobbies. They were mostly designed using Art of Illusion or downloaded from sites like Thingiverse.  I was able to build a set of replica Mercury capsules for an old Lionel car by downloading the 3D model of the mercury capsule from NASA's website.

I have to agree with GRJ and others, for complex real world objects, the real challenge is the 3D model design & CAD software. I have access to AutoDesk Inventor at work and have been trying to teach myself in my spare time.  The learning curve is very steep.    Every time I step away from it for a month or two, I virtually have to start over again.  The only way to get proficient with it is to get trained in it, then use it regularly.  I'm the Electronics/Project engineer, not the Mechanical Engineer or Drafter - so I haven't been able to justify getting money for the Inventor training.

Still, I think this is the way of the future - the technology is only going to get better, cheaper & easier.

Jim

Here is a 1/10th scale ASF 100 ton roller bearing truck.  The 3d model was drawn in Autocad 2015.  The physical model was printed on a commercial FDM machine (fused deposition modeling) using the same process that most consumer level printers use. FDM is when a plastic filament is melted and forced through a nozzle onto a surface. The model is built up in layers.  In many of these pictures the layers are visible, especially on curved or angled surfaces.

I will be happy to answer any questions about this project or anything I know about 3d printing or modeling software.  However, I'm leaving for Chicago for the March 2-rail show so I may be delayed in answering posts.

By the way, to print this project start to finish would take somewhere in the neighborhood of a week plus, printing 24 hours a day.

Jim

Jim, That came out great - I would love to get that proficient with it (and get the printer working that well).

Jim

To be fair, I've been using Autocad for a long time.  Probably 20 years in the 2d realm. The 3d stuff isn't all that dissimilar.  A line is still a line.  But of course there is a whole new tool set but it's generally pretty intuitive.  Instead of trimming a line with an intersecting line, you would slice through a solid object with a plane.  Things like that.

As far as the printer goes, any FDM printer will have limitations based on layer thickness, filament width, and the simple fact that the nozzle has to take up physical space. Complexity of objects is limited by default.  

This project was printed in my community college lab.  The school currently owns two FDM printers in the cost range of $40,000 to $70,000 a piece.  The price is misleading as they are probably five years old.  Technology has improved while costs have come down.  A modern Makerbot can print at a higher resolution for a lot less.  They are talking about getting an SLA style printer, which fuses resin with a laser and UV light. Even more expensive, but the finish quality and complexity of objects is much higher than FDM. Much thinner layers, and no nozzle to get in the way.  An SLA layer would be in the range of 16 to 25 microns (roughly .0005 to .0009 inches) versus .007 and .010 inches for the FDM printers at school.  So an order of magnitude thinner.  It would be possible to get an SLA print without needing to do finishing work to smooth the steps and layers.

Formlabs makes a good desktop SLA printer for about $3200.  A desktop FDM printer like the Makerbot Replicator will cost less but provide a lower resolution print with layers in the range of .004 inches.

There are definitely a lot variables between printers, software, costs, and goals.  More than I have time to discuss at the moment.  I'm supposed to be packing my suitcase after all!  Maybe after this weekend I can jump back in with some discussion on software.

Jim

I pointed this out mostly for the free part. It won't cost you anything to have it around to tinker with in your spare time.

AutoCad 2015 is not what I'd call affordable for the typical user, let alone the "stick-my-foot-in-the-water-to-see-if-I-have-any-aptitude-for-it" beginner.

I don't have images handy, but in SketchUp, I built an office desk, and was halfway through building an elevated subway station (had the columns, crossbeams and stringers made and laid out along with about a city block's worth of actual el structure) when I had issues getting the platform stringers to snap together properly where they angle toward the main structure at the ends of the platform. Not unsolvable, but I was only playing around during some spare time, and wasn't inclined to fully research the solution by the time I hit that snag.

I did download 123D, and when I get a chance to play around with it, if I actually make something I'll re-visit the thread and post a pic

(assuming the thread hasn't closed from age )

---PCJ

This sort of reminds me of the late 80s when the print publishing industry when digital. I was right out of college and young and hungry and bought a Mac and made good money and started a business because I could run circles around the guys who had hundreds of thousands tied up in what became archaic tools in the span of about 2 years. My first laster printer was over $5K so these 3D printers look pretty cheap to me. At the time everyone was afraid of the software, learning curve, etc. Within a few years anyone who wanted to stick with repro and an exacto was out of a job - it happened really quickly because of demand and cost. But, to some degree this is a young person's game - I've gotten too old for the stay-up-all-night-learning-software-game. I'm not sure I want my hobby to venture back into the realm of computers even though the lure is strong.

As a bit of background, I have over 25 years of experience with progressional drawing and graphics software. I've also worked with some 3D animation software where I manipulated objects other people rigged for me. That said, I was able to get somewhere with Sketchup pretty quickly but it really felt primitive. I dont think I have the mental bandwidth or time to jump into something more advanced right now. 

I showed my wife what I had just made and some of the cool output from these 3D printers and her reply was "if you can print me a new living room and dining room I'd be all for it" (these are the 2 remaining rooms to finish in my 10 year long house rehab) so there's that - she's been more than patient. 

I showed my wife what I had just made and some of the cool output from these 3D printers and her reply was "if you can print me a new living room and dining room I'd be all for it" (these are the 2 remaining rooms to finish in my 10 year long house rehab) so there's that - she's been more than patient.

Hey Tim, while you're at it print out that new 4-door Mini I keep getting mail from my dealer about

I recall seeing a flat screen computer monitor for the first time in the XOs stateroom on the USS NORMANDY.  I think he said they paid $1,300 apiece for them, no more than a 12"-15" screen.  Mass production to supply the demand does make things cheaper...and freed up a lot of room on his desk!

I was investigating a rapid prototype supplier today. Asked them some questions.

And yes there is a size limitation. These rough acrylic detail parts are probably best suited for my needs. Problem is there is a size limitation on there machines.

Roughly  284 × 184 × 203 mm was a common machine available, the larger ones cost more monies.

So with the size above I could make a 40 ft box car or smaller parts. Any thing larger like a 21 inch car would cost incrementally higher.

Plus then there is volume... If I made on or two items, then wanted 1000 or more parts well then its much cheaper to soft tool it...

3D printing is great for prototypes and very low production. We use it in early stage of plastic kit production. But the amount of time it takes to print a part true production is not viable. RTV silicone molds, epoxy molds, aluminum molds can all be used in lower production numbers.  

While not train related, a friend of mine and I have been making small parts for vintage keyboards/synthesizers that are no longer available. Special knob, battery covers, and so forth.  These are small parts. Much smaller than printing boxcars, etc..

Just like any new to the masses technology, there will be early adapters.   Prices will eventually come down and more user friendly commercial level software will appear.

I have what is probably the lowest cost 3D printer out there.

A Solidoodle Press, Retail $600 and comes complete, ready to power on, with a starter spoor of ABS material.

The build area is an 8" cube and it has 0.1mm resolution which is phenomenal for the price.  Not that I have any successful prints at max res. I do have good prints at 0.3mm resolution.

This was a startup purchase (I paid early & less to get a first run machine) So Documentation is limited and upgraded docs are slow coming, they are having lots of issues with it.

I did get a response to my latest inquiry with a suggestion I have not followed up on yet. Hopefully this will get it working better.

I have had limited success using it to date (A bit over 2 weeks now).

I made an O scale Picnic table and a pair of Pallets no trouble at all.

I made a 3.5" long BTTF Delorean (Just a hair over scale size), with only minor issues (glue req'd to mount some parts). But hey, the wheels roll  as printed !!!

Larger Items consistently fail to print.

Items requiring support do not print well on this machine (so far).

It is a single custom head machine and I suspect it tends to overheat the print head on longer prints.

Running it with the lid partially open seems to help but more work in this area will get more data. (along with the suggestion from tech support on an adjustment)

I haven't done anything with it this week, Family life and my 26th wedding anniversary over rode everything else.

Now, on the comment about larger items such as a complete boxcar.

I will have to print such an item in chunks. Up to 7.75" length wall sections with a lip to glue together is doable. I would print them inside down, outer surface up for detail. OR I could do steel walls face down for absolute flat surfaces (glass face on the print bed) and make the rivet strips separate.

To quote a catch phrase, The possibilities are endless !!!

We NEED photos of that BTTF Delorien!!!! I've wanted a 1/48 version for years!!

If anyone is mechanically inclined and just wants a very small printer to play with and learn on...

http://www.thingiverse.com/thing:671209

(and that's only one of the do-it-yourself kits; there are about a dozen.) Be aware that English is not the first language for a lot of those posting, so you do have to decipher some of the directions.

I was impressed by the very small one (something like 3x6" maximum print size) designed so students could afford to get started.

Also, just today someone posted directions for a do-it-yourself metal printer. I have no idea how it works yet.

Interesting that BTTF came into the conversation.  I originally built my 3D printer with  the idea in mind of making the BTTF Time Train.  I've been using pieces of that as training exercises in Inventor. I've printed a few test pieces but got away from it for a while and haven't put anything together.

I don't know about 1:48 but there is a 1:43 BTTF Diecast Delorean around. I think it's from Vitesse.  I bought one off the website of a DeLorean owners group a few years back.  I think I've seen them pop up on Ebay occasionally.

Jim

Dave:

The BTTF Delorean is all natural plastic color and only partially assembled at this time.

It needs some work to be worthy of a picture or go on my layout.

In the Meantime Check Here for commercial 1:43 and other scale versions.

Shapeways has a link to people who will generate  CAD files from your drawings.

Of course, if you master AutoCad, you're on your way to a second career!

3D ModelTech uses Solidworks 2014 software, not sure what type of printer they use.

I was a draftsman back in the late 60s, early 70s, I used pencil and paper most of the time, even had a battery-powered eraser (got used a lot!).  Then we had to make the drawing, let the "Checker" look it over (hated the guy , but he was good at his job) then send it down to the machine shop for fabrication.  Sometimes it took weeks to see what I had drawn end up as a hard piece of steel or aluminum.

Now, you can draw it on your computer and make it within hours.  Of course it might not hold up under the stresses that NASA put on our pieces, but that time's coming.  I think I read somewhere that NASA will have, or has, these 3D printers in orbit on the space station to make parts/tools they need instead of having to fly them up on a rocket.

Interesting discussion, but a caveat to be aware of.  I ask rhetorically:  Are there any legal restrictions on copying, producing, and/or selling a part that mimics MTH, Lionel, or other manufacturer's parts, especially parts that are in production?

I would believe if you do not sell the part and only use it for yourself you have nothing to worry about.

Earlier this year they emailed a wrench up and printed it. I think R and D Magazine had the photos in their daily newsletter.

Thanks BT&GM, that's what it was   I knew I wasn't dreaming