[Shdwdrgn]

It occurred to me last night that I should be able to create my own jigs for hand-laid turnouts. The plastic won't hold up to a lot of sloppy soldering but it should hold up well enough to get the rails in place and tack down some key points. Even if a jig only gets used a few times, they would be so cheap to print that you could just throw them away.

Looking over Thingiverse, I did find one item which printed the ties to go with a Fasttracks jig, and that seems like a good idea to include. The real trick is being able to write up something that can generate the jigs to whatever size you need. I could lay out something on Tinkercad but that seems like a waste of time. Guess it's time to learn some other CAD modeling program.

 

[Lemonhawk]

Or you could use Central Valley Model Works ties, which sounds like what your trying to 3d print. Much easier and less expensive than Fasttracks jigs. They will also conform to being fit into a curve.

 

[Shdwdrgn]

Printing the ties is just a secondary item, I'm trying to print an actual jig.

Finally took some time to start learning one of the CAD programs today. I'm kinda stuck for the 'proper' way to generate a curved line but I made do with a big circle to get out a test print. I have the code roughly generating the tie slots then adding a pair of straight and curved rail slots on top (similar to Atlas snap switches). Once I figure out how to make curved sections bending to a specific angle then I should be able to also generate a proper turnout of a given angle.

Had to dust off my 3D printer, guess it's been awhile since I used it. Got the first test jig printing now, will be interesting to see if the rails or ties fit in the slots.

I'm trying to build this as flexible as possible. All of the dimensions are based on prototype sizes, then you enter the curve radius and angle, the scale to print the jig at, and everything is generated from there. I'll also have the option to print a single side or both right and left jigs together. If I can pull this off then it should really help me start whipping out all the custom turnouts I need for my planned layout.

 

[wvgca]

Got the first test jig printing now, will be interesting to see if the rails or ties fit in the slots.

so, how's it turning out ?? a jig is a little big in HO scale, i think .. dunno ??

 

[Shdwdrgn]

Unfortunately the tube popped loose on my printer and filament went everywhere while I was eating dinner. Looks like there's a problem with the hardware so I'll probably have to swap in a new tube or fitting before I can try again.

Oh, does anyone know the standard spacing for older (8x6) ties? For the 9x7 ties I found a spacing (center-to-center) of 19.5" but enough of my first print finished that I can see this is a bit too wide for my older style track.

 

[wvgca]

Unfortunately the tube popped loose on my printer and filament went everywhere while I was eating dinner.

you're talking about the bowden tube ?? they come in various lengths, mine is only two inches long, just inside the assembly only , nothing extra ..

 

[Shdwdrgn]

Yeah. Mine is about 10" long. I trimmed a little off the end, still wouldn't stay in place. Turned out the fitting was shot so I swapped it out and it's printing again. Will see what happens this time.

 

[Lemonhawk]

With the CV ties you don't need the jig!

 

[Shdwdrgn]

Finished the first print... and my math is off. The slots are too small for both the rails and ties, and the rails are too close together by about 1/2 rail width. Well I had 5.375" for the base of 100# rail (0.0617" in scale), and my code track actually measures 0.0685". Also I put 56.5" for the measurement from center-to-center of the rails, but if this is actually supposed to be the distance between the rail heads then it would explain the error. Ah well, that's enough for tonight.

@Lemonhawk - How many different radii does CV carry? My current jig is for a 20.75" radius curve but I have different turnouts from 18.5" to 72" I'll be making. And then there's all the curved turnouts with various dimensions. And then we get into all the possible dual-gauge turnouts. I don't NEED a jig for any of these, I can just build the one-offs directly from my printout, but I do have a handful that I use the same dimensions for and it would be nice to have a jig so I can make them consistently.

 

[Lemonhawk]

http://www.cvmw.com/ #5,6,7,8 and 9. Since the tie strip bends, you can make custom curved turnouts.

 

[gregc]

And then there's all the curved turnouts with various dimensions.

after struggling with an ill fitted turnout, i replaced it with a curved turnout. Unlike others that i built on the layout, i created a template and built in on the bench. The struggle was properly locating the frog. It's easy after that and with the outer stock rail are in place.

i left space for the PC ties when i glued the ties down on the layout. but after spiking the turnout in place, i replaced the PC ties with wood ties except for the drawbar and one other.

for the curve radii i needed, the frog is quite a distance from the points. I couldn't understand how some commercial turnouts claimed to #6.5 frogs. it was only later that I figured out that if centers of the curves were offset, that the curves start at different points, the frog will be much closer to the points and the turnout becomes shorter

 

[Shdwdrgn]

Holy cow, that is one long turnout! I'm surprised there was enough clearance around the points for the wheels. I had some really long ones in some of my initial layout designs, but as I played around with it more I got a better understanding of the geometry and have been able to use much shorter curves. I still have one piece where the dual-gauge splits into separate routes through a long curve, hopefully I can still get away with not using any points for that one.

Been cleaning up my formulas today, fixing errors where things weren't properly centered. Think I have that all worked out now. I added an oversize variable to provide a little more clearance for printing errors. Just finished another test print and it looks like my rails gaps are still a little too tight, and too far apart now by a full rail width. Getting closer though...

 

[Lemonhawk]

Shdwdrgn, didn't you build your own dual gauge turnouts a while ago? Interesting, I see no reason for points when the scales diverge to go their separate ways, you'll just end up with a frog on the outer rail. I'm thinking of building a Gauntlet Bridge and the 2 converging routes would not have any points. Luckily for me, CVMW now has Gauntlet Bridge tie strips!

 

[Shdwdrgn]

Yeah I've built four dual-guage turnouts so far, but those had both gauges going in both directions. When the two gauges split apart there *should* be no reason to need points, but I've run across a fair number of builds where they did in fact use at least one point to ensure the train took the proper direction. Even on the track where the common rail switches sides I have seen builds using points, but my own builds so far have just used guard rails to keep each gauge on its proper route.

A gauntlet has the advantage that the two sets of rails never actually come together, it's more like a shallow crossing of a single rail. Similarly, when I built my weight scale I had points to select the route, but no frogs because the rails never crossed each other. This is one of the reasons why I wanted to learn how to hand-lay turnouts -- because there's just so many cool and unique features you can model that don't have commercial counterparts.

* Update on the jig model: I have the gaps for the rails and ties just right, and I think my tie spacing is matching the flex track I have. Just working on dialing in the spacing between the rails now.

 

[Shdwdrgn]

Got the rail spacing dialed in and decided to print a full jig to see where I'm at. The section of rail I had laid down smoothly into the curve, which is more than I can say for the rails I've bent by hand. I forgot that the rails in a curve will be slightly tighter than straight rails, though, so I do need to open that up slightly. Guess that may explain why all of my flex track seems to be wider than the NMRA gauge. Otherwise it's definitely getting there, although I should probably try a test with soldering some rail to a tie so I can see how badly the heat affects the plastic.

So the next thing I need to do is write up a routine to lay down an arc instead of a full circle, then I can finish the diverging path either in a straight line (such as for turnouts with a given frog number), or I can continue the arc (such as with Atlas snap switches). Then maybe I can determine where the guard rails should be placed and add indicators for where the rails should be cut around the frog.

 

[wvgca]

is there anything on the jig to show where to solder the pcb ties in ??


oh, a ps .. are you using regular PLA or something more heat tolerant ??

 

[Shdwdrgn]

Not yet, but the plan is that once I can do the math on the arcs I should also be able to determine the key points where PCB ties are needed, and then only provide slots at those locations. Or perhaps make that an option, and simply mark the locations on the jig where you would normally drop in PCB ties.

 

[Shdwdrgn]

Well I'm up way too late at night working on this, but I was successful in finishing the formulas. So I can now generate a template based on the diverging path curving at a given radius for a given number of degrees, then continuing on a straight path. And from there I was able to calculate the correct tie lengths along the length. Add in a simple mirror and I have a complete jig.

Shouldn't take much more to mark the PCB tie locations and figure out where to place the guard rails.

 

[wvgca]

Shouldn't take much more to mark the PCB tie locations and figure out where to place the guard rails.

As far as marking the PCB locations, a simple 'P' on the end should suffice ?? I wonder if there is enough room on mine to do one side ?? it's a copy of a prusa i3, think the print area is 200mm on each side ??

 

[Lemonhawk]

You might luck out on the soldering. With some liquid flux dapped on it might solder quickly and the PC ties are sort of insulated from the guide. worst case you might have to raise the spots where PC ties go to insure contact.