Building smaller code rail?

[daniel_leavitt2000]

This is sort of a hypothetical question right now as I have no layout and will not be looking at track for a few years. I was wondering what it would take to shape a smaller code rail than commercially available. I know Dave took a look a few years ago and ended up using squarish wire. I am curious as to how one would shape this wire to a more prototypical shape.

Idea 1: Mill hardened steel with a hole the shape and size of the rail in question. Thread a tapered wire into the hole and pull. The resulting wire should be shaped to the size and form desired. I wonder if the act of pulling would cause irregularity in the rail size though. Would formed parts be streached out while trying to pull the rest of the wire though? Would there be away around marring the formed rail with tool marks from the pulling device?

Idea 2: Mill two hardened steel rollers with half the rail shape. One roller would have a shape like "]" along the center while the other would have a shape like "[". When the two rollers are brought together the hole btween them would form the rail. All one would need to do is feed a wire about the same size and the rollers would form the wire to the desired shape. This would eliminate the pulling/tool-mark issue of Idea one but may result in flash and an uneaven rail head.

Comments, Ideas?

[peteski]

Um, what you describe in Idea 2 not an idea - that is pretty much how model rails are already being produced. If you examine certain brands of N-scale (or other scale) rail you can see that the rail was formed by rollers.  At least that is how it looks to me.

As far as Idea 1 goes, I also think that this is already being used for making metal wires and rods (and wooden dowels too). I'm not sure how well a rail-shaped opening would work but I suppose it is possible. But it would probably not be a single operation (like your Idea 2).  The die would have to start close to the shape of the original wire stock you are using for the rails, then the next dies would be closer and closer to the final shape of the rail.

[avel]

RLW sells code 30 rectangular rail.

[DKS]

I've found that below Code 25, it's hard to tell what the shape of the "rail" is to begin with. Even under macro photography, it's hard to tell. It would probably only be worth the trouble on foreground track. This is Code 25 rail in Z scale:




BTW, as Petesky notes, both of your ideas are industry standard practices. He is also correct that pulling wire to rail shape could take anywhere from two to four or more hardened steel dies, and at a cost of thousands of dollars per die, it's a very expensive proposition. Not to mention difficult: you need quite a lot of force to pull the wire through. In wire shops, big hydraulic machines do the work.

Rolling is a lot easier, something one could potentially do in one's garage with a hand-powered roller. The problem you will face, however, is wire twist: the wire will want to curl in one direction or another, requiring straightening after rolling, a process is quite similar to pulling wire, using lots of tension. You can usually tell when rail is rolled as it will sometimes have what looks like a seam running down the center of the base.

I have been pursuing the purchase of my own professionally-drawn/rolled rail for a new product line, and I can tell you first-hand that, one, it's really expensive, and two, it's actually very difficult to find a shop that will make small quantities of it. One shop I contacted had a minimum order of six tons of material. That's the equivalent of hundreds of 1:1 miles of rail! I'd need to rent a warehouse to store it.

[havingfuntoo]

You cold pull it through a drawing bar but to keep it from twisting could be difficult and you would need some way of keeping a constant pull during the drawing process, bees wax can be used as a lubricant. Depending on your stock size you may have to have several goes at it through step down drawing bars ....... a lot of work ......... hope you don't have a big project in mind!!

[nkalanaga]

Actually, six tons of 1:1 rail won't go very far.  Assuming 120 lb rail, that's only 100 yards...

[DKS]

Actually, six tons of 1:1 rail won't go very far.  Assuming 120 lb rail, that's only 100 yards...

To clarify: six tons of raw material makes many 1:1 (real) miles of model (scale) rail.

[SquirrelHollow]

The biggest reason I haven't been around here for a few years, is that I started my own company - selling specialized reloading and bullet casting tools for hunters/shooters.  To get many of my ideas off the ground, I had to learn many more details about manufacturing processes, than I ever imagined.  ...Because the design process for every part includes the classic question: "Build or Buy?"
Understanding the actual manufacturing processes, tooling costs, labor cost, and "machine time" costs makes a tremendous difference when doing cost analysis for that question.
Based on what I've added to my grease monkey/blue collar knowledge base over the last few years...


I would say, without a doubt, that using rollers would be the best "hobby" method.
And, it would be more cost effective to make multiple passes while using a cheaper material for the rollers, than to try to make a single pass with harder (expensive) rollers. (but that requires the ability to adjust the rig)

I have a few ideas of how to build a rig to do it.  But, it would be expensive to machine the rollers.  I'll bounce the ideas around in my head for a day or two, and see if I can come up with something simpler.



DKS, I completely understand.
I don't do any machining myself.  I just do minor fitting and finish work on prototypes.  I don't even own a "mini-lathe" or "mini-mill".  All of my machine work is done in local machine shops or 'job shops'.
Usually, the only difference between the cost of a prototype and the cost of a production run, is materials.  Setup and tooling costs are killers for small production runs....

(To explain, quickly:  For me, "Build or Buy" is a question of finding an existing part or piece of stock material to build my design around, or if it's better to have it machined or cast to my specifications.)

[ednadolski]

Just a $0.02, don't forget that rail is only part of the trackwork modeling picture.  There are also ties, tieplates, spikes, joint bars, etc.  To my eye these are much more noticeable than rail profile or rail height.

BTW I'm presuming that you have N or Z scale in mind, but if you really want detailed track you may want to consider something larger.   The best trackwork that I've seen is O-scale P:48, and there are a number of very good components that already exist.

HTH,
Ed

[nkalanaga]

"To clarify: six tons of raw material makes many 1:1 (real) miles of model (scale) rail."

True, and in the sizes being discussed here would probably supply the N scale old-time and narrow gauge folks, and the HO mine and tram folks, for at least a generation!

[robert3985]

Just a $0.02, don't forget that rail is only part of the trackwork modeling picture.  There are also ties, tieplates, spikes, joint bars, etc.  To my eye these are much more noticeable than rail profile or rail height.

BTW I'm presuming that you have N or Z scale in mind, but if you really want detailed track you may want to consider something larger.   The best trackwork that I've seen is O-scale P:48, and there are a number of very good components that already exist.

HTH,
Ed

I agree 100% with Ed.  I've been hand-laying turnouts for nearly three decades now in N-scale in code 70, 55 and 40, and my biggest gripe about my hand-laid turnouts has been the lack of all of the aforementioned "track furniture" on my turnouts, and to a lesser extent, on commercially manufactured turnouts. 

Luckily, Andy Reichert at Proto87 Stores has N-scale track and turnout superdetail frets of code 40 tieplates and switch tie plates, point hinges, throwbars and precision milled frogs and closure points.  I have several of these frets, and I am both impressed and distressed at how incredibly small these details are.  I'm applying them to my latest seven code 55 turnouts to determine if it's worth all that extra effort to apply them.  Stay tuned for photos and comments in the next week or so.

My biggest gripe about commercial rail is the universal overly large size of the the "spike head", even on new Micro Engineering flex, the smallest of the batch.

The best flex was the Railcraft flex, with its micro-sized spikeheads and uneven ties. WOW!...it was great stuff until the molds started giving out.  Good thing I bought several dozen bundles of the old stuff which I use for my foreground scenes in code 55 and code 40.

I'm posting a photo of a scene I created back in the 90's using Railcraft code 70 flex because it was a set of two Ntrak modules and I wanted the rail size to be close to the then-Ntrak-standard of requiring the use of Atlas code 80 flex.  Unless you're really squinting and looking for it, you don't notice the tall rail height...at all.  However, you DO notice the correct tie size and spacing, and the lack of huge "clampers" that supposedly pass for spikeheads on Atlas code 80 flex. The contrast between my modules with this track on it and those with Atlas 80 flex was commented on my nearly every person who was walking by.  It had nothing to do with rail height.

Does the rail height jump out at you here?


Unlike what everybody suggests with Atlas 80, I didn't "hide" the Railcraft code 70 flex by over-ballasting it.  I did paint it and weather it as close as I could to what the real rail and ballast appeared to look like in my reference photos of this spot on the U.P. at "The Monument" in Wilhemina pass just East of Devils Slide in Weber Canyon.

However, when I look at prototype photos of trackage before welded rail, what is noticeably missing in my photo are the joint bars.  Since Proto87 Stores now produces N-scale joint bars (as well as other "track furniture"), I am thinking seriously about applying them!  I just have to work up the courage to devote all that time to do it!

What would be extremely cool would be rail that is rolled specfically for hand-laying.  When manufacturing the rollers, etching in joint bars at appropriate distances would not be that difficult to do.  That would be AWESOME!!  Only problem would be that there would then be rail specifically made for each scale, rather than just code-whatever rail that works for whatever scale is appropriate. Probably not cost-effective, but it would be very cool in my estimation!

My train buddy Gregg Cudworth, who displays his standard-gauge modules bolted up to mine at three shows each year, is filling up his basement with an Nn3 version of the Rio Grande Southern...the whole railroad...double decked!  All of his track is code 30 flat wire, hand laid to PCB ties (every fifth tie).  Frankly, I don't notice the lack of a "profile" either in my photos or when I'm squinting at his latest hand-laid stub switch.  However, Gregg does notice the lack, and has a scheme in mind to roll a black paint "shadow" on the sides of his weathered rail to "suggest" a profile.  I am assuming that nobody would notice the shadow is painted rather than an actual recess in the sides of the rails.  We'll see when he gets around to doing that.

Cheers!
Bob Gilmore

[DKS]

Further to the handlaid lack of detail problem, I have found that it's effective to use 100% PCB ties when using small flat wire, and to remove all of the copper except for small "pads" under the rail to simulate tie plates. The drawback is that it's a lot of work, not something you'd want to do for smiles of track and dozens of turnouts. Here's an N scale turnout laid using Code 25 flatwire rail using this technique:



The turnout is not yet finished--still to be done is to fill in the openings in the frog flageways.

[bbussey]

... Idea 2: Mill two hardened steel rollers with half the rail shape. One roller would have a shape like "]" along the center while the other would have a shape like "[". When the two rollers are brought together the hole btween them would form the rail. All one would need to do is feed a wire about the same size and the rollers would form the wire to the desired shape. This would eliminate the pulling/tool-mark issue of Idea one but may result in flash and an uneaven rail head.

Um, what you describe in Idea 2 not an idea - that is pretty much how model rails are already being produced. If you examine certain brands of N-scale (or other scale) rail you can see that the rail was formed by rollers.  At least that is how it looks to me.

That's definitely the method Atlas utilized when the track products were manufactured in Hillside.  It definitely was impressive to watch their home-built massive rail-forming machines, which ran 24 hours if I remember correctly, in operation.  And they worked with huge spools of wire.

[Bsklarski]

Just a $0.02, don't forget that rail is only part of the trackwork modeling picture.  There are also ties, tieplates, spikes, joint bars, etc.  To my eye these are much more noticeable than rail profile or rail height.

BTW I'm presuming that you have N or Z scale in mind, but if you really want detailed track you may want to consider something larger.   The best trackwork that I've seen is O-scale P:48, and there are a number of very good components that already exist.

HTH,
Ed

I think that hits it. You can take the focus off the rail by all the other details around it. If everything else is done well, not many people will sit there and notice the rail

[Mark5]

Good points here. I'm frankly surprised that Code 30 is not available today, given that nearly 40 years ago I was hand laying Code 40 rail. Code 30 would be of use to some HO narrow gaugers as well.