Custom machined train cars

Those are cool!

KENLY, nice work. You may be able to get into mass production if you have more then one head on your CNC. What type of wood are you using? 

I'd like more info too. Material, design software and type of CNC, thanks

I could use one of those 68' flat cars.

Do you plan to sell them as kits or completed rail cars?

Thanks guys! So I'm a mechanical engineering student at BU but I had to take a medical leave (long story) so I have some time before I go back next fall... I was actually considering selling some cars once I get to a point when my programs are repeatable. If I sold anything they would be complete and assembled and likely painted as well. All my models operate on O-45 track.

Right now I'm still in the planning stage. I almost always draw out my basic idea first on paper, then I model the parts in Creo. After I finish the design there I import the CAD files into my CAM software to produce the tool paths.

**Update**  for legal reasons I've switched to Autodesk fusion 360. I no longer use CREO.

We have a Tormach 1100 CNC 3 axis milling machine. The spindle has a single phase 1.5hp 240v motor and the table rapid is 110ipm. Because of that I can cut really fast and really aggressively on the roughing passes. The depressed flatcar was only about 2hrs of run time on the CNC in total. It required an entire day to get ready to even turn on the CNC though...

I want to mill these cars out of aluminum eventually but it's more expensive, less forgiving and takes about 4x longer to mill then wood. I've been using clear white maple. Little pricey as well but definitely worth it.

Pic of our shop

I've been experimenting with glossy oil based paints

few renders of the schnabel design

Very interesting and looks like a lot of fun. It must be very rewarding when the finished product is held in your hands. Maybe you can create a casting and pour the aluminum. We used to do that in metal shop back in high school. 

Those are so cool, I'd prefer a kit. I feel kit building is something that's been lost in this hobby. 

Thanks for the info!!! I'd like a flat someday!!

Thanks Dave Z! Yeah it definitely is! I've actually considered that... We don't have a furnace though so I'm not sure how I'd melt the metal efficiently enough for production.

Will- that's interesting you'd prefer it in kit form... the reason I was thinking I'd assemble them is I assumed nobody wanted to do so. You might be disappointed though as the depressed flatcar is only 9 pieces not counting the trucks or hardware. I spend a lot of time during the design stage making sure I've come up with the simplest way to do things.

AMCDave- The 68' flatcar design is ready, it still needs testing on the layout though. I'm just worried how it might behave at the front of a heavy consist. Couple hours of testing should answer that

Thanks....I'd prefer kits too. A heavy load would make it run fine in any consist. 

Yes lets us know if you are planning on selling them.

I'd like one of the 68' flat cars just the way it is. I can paint it myself.

That's great to know you guys like kits! I hate to disappoint though but the 68' flatcar is one solid single piece of maple. Nothing to assemble. Ya just gotta screw on the trucks. I bet if you guys are skilled at weathering you could make these cars look beautiful. I'm terrible at weathering and decorating models lol

Just out of curiosity what length train do you guys normally run? I've had a loaded 20 car train behind both cars for about 2 hours with no problem. I was just a little concerned the screws didn't have enough bite due to how thin the deck of the flatcar is

KENLY, can you drill a larger hole and press a small shallow "T" nut in or screw in a wood insert nut with some glue applied. One thing nice about the wood insert nut, they can be easily cut off with a Dremmel tool cutting disc if your hole is too shallow for it.  Are you using Sugar (Hard Maple) or Red (Soft Maple)? There is a big difference in weight and density of the two.

kenly posted:

That's great to know you guys like kits! I hate to disappoint though but the 68' flatcar is one solid single piece of maple. Nothing to assemble. Ya just gotta screw on the trucks.

That's perfect!! A bare single piece body is still a 'kit'. I'd add sturip steps, some brake detail....maybe deck detail....plenty to do!!

I usually run 15-20 cars

Count me in if you decide to offer these for sale.

Yes the brass screw inserts are nice because the wood is less likely to strip out over the years.

Your work looks great but if you go into commercial production you must recognize that there are thousands of cars designs already out there at low prices. There is always the niche market for something new but since Lionel made a four truck 16 wheel flat car with transformer and  bridge beam loads, its hard to tell where it will be.  Good luck on the venture.

Dave- The white maple I'm using is considered hard I believe. Great idea! I like those brass inserts. I'll order a few from mcmaster and try them out. I've been meaning to place an order there anyway. I had no problem with the car last night but I wonder about longevity. To be honest I didn't put much thought into the flatcar yet. The big 32axle schnabel is my main project, these other ones were just stepping stones along the way.

Dennis- You're right, I'd never even try to go into mass production! The market for non-scale models like these is way too small. I was just hoping to maybe make like 10 schnabels this summer before I go back to school. I didn't expect this much interest in the other ones lol The interest is great though!

Little info on the current car designs:

the 68' flatcar is in 1:48 scale so the deck is 17" long by 2.5" wide. It's big, makes the scale lionel 60' boxcars in the video look small. Capacity due to the sprung trucks is 2.5lbs but the body is begins to flex at 50lbs!

the non-scale depressed flatcar is 24" end to end and the usable deck is 10"x2.5" (I believe the lionel one is only 15.75" long?) Capy due to the trucks is 5lbs. I haven't tested the body yet

32 axle schnabel without a load is 2.5" wide 4.13" tall and 41" long. It has a working 0.5" lift mechanism to clear gradients in the track. Loaded I plan on it being 3" wide by 60" long. Usable capy due to the trucks is 20lbs

kenly posted:

...Dennis- You're right, I'd never even try to go into mass production! The market for non-scale models like these is way too small. I was just hoping to maybe make like 10 schnabels this summer before I go back to school. I didn't expect this much interest in the other ones lol The interest is great though!

Little info on the current car designs:

...32 axle schnabel without a load is 2.5" wide 4.13" tall and 41" long. It has a working 0.5" lift mechanism to clear gradients in the track. Loaded I plan on it being 3" wide by 60" long. Usable capy due to the trucks is 20lbs 

 

 

Any chance of the Schnabel being boosted to 36 axles like the prototype WECX 800/801?

The innermost span bolster is a split design, the outer end riding on a single truck and the inner end on one more span bolster similar to the ones under the rest of the carbody (the above video shows a lot of details)

I already have the 18 trucks from an existing initiative to scratchbuild this car in styrene. My project is currently on hold since (among other things) I was puzzling out how to articulate the load arms to handle uneven track since the prototype ones ride on massive ball joints--which probably explains how the 801's arms managed to flop right off the car when they bottomed out on a grade crossing en-route to its first job....

I can still build the various platforms, control cabs, lights, hydraulic lines and railings on my own if they're outside the scope of your project. You just seem to be a lot closer to having actual span-bolsters and load arms built.

---PCJ

haha you read my mind! I plan on doing a WECX 801 after I complete the 32 axle one. I want to start with the 32 axle one because there's more symmetry which allows for easier and faster part production. I modeled an O scale 36 axle schnabel years ago before I had any concept of manufacturing. It needs very significant refinement before I could even begin to make parts. The way I intend to deal with articulation is exactly how the real one does it! just a ball and socket milled right into the wood. I'm a little concerned how that's going to perform in scale. I ruminated about that decision a lot. Only 1 way to find out if it's going to work or not! You certainly wouldn't be able to pull a lot of cars after it... but they don't in real life! so that's ok with me

Heh, I do have 3 of the MTH Schnabels, and only run them in a dedicated train with idler flats. My intent with the 800/801 was to simply run them unloaded, as if headed to a job, as in 'Meet the Schnabels'. The last idea I had with the load arm articulation involved using a bar magnet inside the load arm joint suspended over another magnet placed just out of its reach inside the carbody below. That way the arms could rock side to side relative to the carbody, but always return to center where the greatest magnetic attraction would be.

There are also magnetized steel balls available from suppliers like Amazing Magnets:

Neodymium spherical magnets, though these may be a tad strong in the most practical (i.e. scale) size...

They also carry Chrome steel, nickel-plated steel, and stainless steel balls that might work better with a disc magnet below, or even in conjunction with small ball bearings inserted in the socket portion.

---PCJ

Will,

Nice CAD work but could you tell me more about the “PTC” letters on the lower right corner of some of your images?

Are you using a student version of CREO & is CREO creating the “PTC (Parametric Technology Corporation) marks on the images you are saving?

I have used CREO (Pro-E) since 1999 & have only seen such marks on student versions. If, you are using a student license, I would advise you against offering the parts made of these designs for sale to anyone, even on this forum, as they could violate the terms of use.

These are just my opinion,

Thanks,

Naveen

Is the application capable of exporting models in a format that other CAD programs without a student-version issue can read?

(in much the same way that the free version of SketchUp cannot output files readable by 'normal' CAD applications, but there is a workaround in which you can export a .kml (Google Earth) model that can be read by certain versions of Blender, which can then re-export to an 'industry-standard' file format (or so I am told)

---PCJ

PCJ- The magnets are an interesting idea! If the normal wood ball and socket don't work due to the weight of the car, a metal ball as the pivot with a magnet below the socket would definitely hold it down well enough. There isn't much horizontal force on the joints... as long as the load support frames weigh enough I think we'll be ok. I'd like to make some parts and test it out but I've been fighting with software a lot lately.

Naveen- You must work for PTC if you know something like that? That's very subtle, and something I didn't even know. That being said that watermark isn't permanent. I can change it to be anything I want. Either way the only reason I've been using CREO is I'm stuck with it at my university. I started using Pro-E about 10 years ago so I know it very well. I think CREO is a poorly written outdated program. A single super bare seat of CREO parametric is $6600 (if you're lucky)! And the best part is even if I did buy a seat I wouldn't be able to open any of my old work... gotta love PTC.

I'm totally aware I can't sell any of my old work made in CREO student, and for that reason about a month ago I began looking into different CAD packages I might actually buy. I'm actually on a trial of IronCAD right now. It's pretty good. No matter what I have to buy a new program and redo the schnabels so if you have any suggestions for a good CAD software let me know!

Will,

I faced the same problem with the PTC watermark when I was saving the images from CREO. During my last semester in 2004, before I graduated with my Master’s program in Engineering, I was saving & printing images of my CAD work, to make a portfolio, to show to potential employers. CATIA didn’t give me any such issue but the student version of CREO (then Pro-E) started saving similar watermarks, without me asking, on my images & even worse watermarks on my 2-D drawings. I had to redo my images, by resizing & cropping them to remove these watermarks.

Then when I got my first job & one of my colleagues wanted to review one of my Pro-E models from my student days, the commercial version of Pro-E wouldn’t open any files created in the educational version & we also verified that any model created in the commercial version couldn’t be opened in the educational version of Pro-E, even when both the educational & commercial versions were of the same release level. Apparently, PTC didn’t want people using the heavily subsidized educational versions for commercial work.

Also, the $6,600 you mentioned for CREO is just for a single license of CREO, JUST FOR 1 YEAR.  All CAD software these days are not sold & nobody owns the license after paying these big bucks. One only purchases a license, valid for a year, with all the fine prints. If one wants to keep using the software, one has to also pay for a licensing agreement, costing 10 – 30% of the initial purchase price. A small company or an individual with a single license might be charged a 30% annual renewal fee but a multinational company with 100’s of licenses might only pay 10% annual renewal fee, per license. Other workbenches or applications that support CAD programs (Product Lifecycle Management, Computer Aided Manufacturing, Analysis), online training & support would also add to the price of the software making a $6,600 CREO license go up to like $24,000, for the first year. I am speculating that that might be the reason why the largest O-Scale / O-Gauge manufacturer, Lionel, could only afford (need) SolidWorks, not CREO, CATIA or Siemens NX.

Since you asked me for advice on CAD software my advice is to keep using CREO. If you don’t like CREO, & once you go back to regaining your student status, try CATIA or NX. I once knew SolidWorks & AutoCAD but during my job search I found out that companies that needed skills in CREO,  CATIA & NX paid a lot more than companies that could only afford SolidWorks, SolidEdge, Inventor, etc. I would much rather have a well-paying job to afford the trains I want than use the same time to train & be proficient in cheaper CAD software, like the work smart, not hard, American saying.

These are just my opinion,

Thanks,

Naveen

Interesting... Well the watermark is removable within the program for me... I think on certain rendering resolutions it just defaults to that yellow PTC.  All the companies my dad has consulted for use CREO so I know I'm stuck with it professionally. But for home use I have to figure something out if I want to share any of my designs with anyone this summer. Otherwise I won't be able to do anything aside from post pictures. I know how expensive CREO can be! That's why I wouldn't be able to justify it just to sell a couple of model train cars. I still have it and can use it for my educational purposes though which is nice... i guess. I'm not sure what to do about this yet honestly

Hey Will have you tried Autodesk Fusion 360 for your personal work? I’m not familiar with Creo so I wouldn’t know how it compares feature wise but it’s free for students and start ups making less than $100,000/year. 

I’m a Rhino3d/AutoCAD user so I’m just getting used to the interface now but I think it’ll be worth it to learn more.

Check out the user agreement here:

https://www.autodesk.com/compa...ervices-entitlements

Would love to have a 32 axle Schnabel and the 8 truck car. Both would be great.  Just hope I don't miss when you decide to sell these. I like the completed car idea. I do not do many kits anymore. 

As long as we're discussing software, here's the "El Cheapo" way I am employing for my project:

Working from a combination of stills from "Meet the Schnabels", along with every photo I could get my hands on, and a manual for an HO resin kit of (then CEBX) 800, I started building up various views in Inkscape, an free/open-source 2-D vector drawing application. These span bolsters are not exact copies of any diagrams--they're hand-drawn .SVG vectors with modifications to accommodate the raised bolsters of typical O-Gauge trucks. I used and drew MTH trucks since I was able to get hold of 18 of those--but I left off the couplers for this application. (Yes, the coupler tabs are offset from the centerline of the truck, in case you were wondering).

Since I don't have access to any sort of milling or 3D printing apparatus, my plan was to print out the elevations at actual size, laminate them to sheets of styrene and cut on the solid line. Much like a laser or waterjet cutter, but manual, nowhere near as precise, and dead-slow (but dirt-cheap compared to sending the diagrams to some place like E-Machineshop to have them water-jet cut or 3D printed--although to their credit you can design parts in 3D with their proprietary free CAD application )

---PCJ

ATANZ- WOW! That's great info! Thanks for letting me know! I'll definitely give that a look.

DOC- I'd probably offer both complete and kit form. I'm not sure what level of detail I'm going to be able to achieve yet though... I haven't even been able to make a prototype yet because so much of my time has been going into this CAD problem.

PCJ- Very nice sketches! Those look just like the real thing! You were going to make that out of styrene? I don't have experience with that type of plastic but isn't that really soft? My bolster shapes aren't spot on prototypical like yours but I did that deliberately for strength and simplicity of machining. Wood is pretty inconsistent I think a piece of wood that thin would be really fragile... although I take the strength of my models a little too seriously lol

Depressed flatcar took 50lbs with minor deflection

Great work Will!  Let's say you wanted to commercialize your flatcar, but you don't want to sell, package, and ship completed models.  Could you put the CNC machine code on a website, and charge $1 to download it?  If so, could I then take it to any machine shop, and have them make an exact duplicate?

Probably dumb questions, but thinking that this might be a good way to create niche models (or small runs of needed repair parts.)  Thanks again for sharing, very interesting!!

Thanks! But it doesn't work that way for a couple of reasons. Long story short the CAM process, which is the step in between CAD and the G-code is specific to our exact machine... because of tool numbers, offsets and the specific format of the G-code. Every CNC manufacturer has subtle differences in their code so the posting template is different. I actually had to write my own post-processor template to get our machine to behave properly. If you loaded it into any other CNC is wouldn't run correctly, probably damaging stuff. Basically the guy who does the CAM has to run it as well. The going rate to have something made on a CNC is $75 an hour and up... in my area at least. and that doesn't count the CAM which they would have to do as well, or materials. So I'm going to be the cheapest option by a lot!

I have an update for you guys!

Many thanks to ATANZ for telling me about Fusion 360, I spoke in depth with Autodesk and it's free indefinitely as long as I make less then $100k a year! It's nearly identical to Creo as well, at least for my purposes.

I'm still experimenting with load tie downs and other stuff but once I'm happy with the design my plan is to make a limited run of 24 flatcars. 4 are reserved so I'd offer 20 for sale when I do this. At the moment my plan is to offer them undecorated (partial-kit form) with tie down chains/springs... see last picture. Then I'll move onto the depressed flatcars and do something similar

Will,
I would like one of your Flat cars. My email is in my profile.

nice! glad you got up and running with Fusion 360. 

Thanks Big Jim! I'll add you to the list!

Here's a close up of an end pivot mechanism for a depressed flatcar. All stainless steel!

I