Making a template

The reason I don't use plans for hand laying my turnouts is that I rarely want a standard turnout.  One of the advantages of hand laying is that you can build anything, working in situations where no commercial turnout or standard geometry will work.  With that in mind, I make my own paper template for each turnout.

In this particular case, I chose to copy an Atlas #6 because I had one handy, and since this is just a practice run I didn't have any specific geometry which I needed.  The technique and result are the same, except that I omitted a couple steps by copying an existing turnout.  Here's what I would do to get a custom template.

First, I lay a piece of flex track in place for one of the two routes.  I pin the point end as securely as possible, and make sure the shape is exactly what I want.  I then lay a piece of paper over the track, pinning it to the layout on one side only, and trace the locations of the rails.  A crayon works quite well for the purpose, though as you can see, I used a pencil in this case.  I then flip the paper out of the way, move the flex track into the position of the other route, and trace that as well.

The result is a plan with two pairs of lines converging into a single pair, and everything should be the exact geometry needed for the situation.  The next step is to figure out where the points will end, and mark that on the template.  This should be exactly the location where the two paths begin to diverge, but it is sometimes wise to err a little on the short side, moving the points a hair closer to the frog.

Ken, Did you read the

Ken,

Did you read the February " Up the Creek" column?

Charlie

Stock rails

I like to start by layout out my stock rails, and tweaking the rest of the parts to fit.  First, though, I find it helpful to have some pieces of brass to help hold things together.  I cut scraps of .005" brass and taped them to the template beneath the frog and about half way between the frog and the points.  This is one of the things I'd change - next time I'll use larger pieces which can span all the way from one stock rail to the other, making it much easier to keep everything in gauge and aligned.

The straight stock rail (if there is one) is the easiest, since it's just a plain piece of rail.  The diverging stock rail is only a little more difficult.  I mark the location of the points, and put a slight kink in the rail there.  This will allow the point to fit neatly against the stock rail later on.  I then carefully bend the rail so that it matches the template as closely as possible.

On a curved turnout, only the inside stock rail would need a kink.  On a wye, both stock rails need it.

With the rails ready, I solder them to the brass plates on the plan.  The double stick tape I used to hold the brass in place softens with heat, which allows the rails to move around later on.  That's why next time I'll use a single piece and trim it after the fact.

Shame on me!

I confess that I did not.  I've only skimmed the Feb. Issue, but now that you mentioned it I went back and read the column.  I note a couple differences, especially that you're laying on PC ties and that you have notches in your stock rails.  The method of making the plan is identical, though.  I also note that you're building in place, whereas I'm building on a workbench.  the combination of working on the bench and not using PC ties means that I need another way of holding things in proper alignment and gauge until the track is laid.

The frog

With the stock rails in place, I start on the frog.  I use a large flat file to shape all of the rails, working either against my fingers or a wooden block, depending on what's easier at the time.  I file each rail down to a sharp point, as close to 1/2 the frog angle as possible.  For a curved turnout, I bend the rails to fit the plan, but for a standard turnout I leave both rails straight.  When they're properly shaped, I solder them to the brass plate.  If there is a slight gap, I fill it with solder as well.

At this point, I solder some scraps of brass to the tops of the rails to hold them in proper gauge.

It's helpful, if there is enough room, to put a second piece an inch or two away from the first, which will help keep the rails from warping.  I ended up adding these later on, after the gauge in the middle kept moving around on me.

The points

Here comes the trickiest part of the whole thing.  Since I don't notch my stock rails, the points have to be made very carefully in order to fit up against them.  It's not difficult, but it can be a little time consuming.

I start by making the point, and then figuring out the length last.  The first step is to put a slight bend in the rail, away from the stock rail.  Then, file away the head of the rail so that it is flush with the web, tapering back to the point where you bent the rail.  If you've done it just right, the bend and the tapered head should cancel each other out, leaving a perfectly straight edge for the wheel to run on.

The next step is to file away the back of the point in a very long straight taper.  Next, taper the inside bottom edge, so that it will clear the base of the stock rail.  When it's finished, the point should be quite sharp, and will fit neatly against the head of the stock rail.  I typically taper the bottom a little extra, which means that the top of the point is a little longer than the base, but it works well and looks fine in my opinion.

Once the point is properly shaped, I figure out where it will sit and bend the frog end to make the flangeways.  I solder it to the brass plates, making sure that there is a smooth, straight path from the point rail, through the frog and onto the rail beyond, and that the gauge is consistent through the turnout.

The second point is done exactly the same way.  It is a little easier to figure out where to bend it for the frog, since you can just make it match the first one.

Because I used two small pieces of brass beneath the points, instead of one, I had to add a couple pieces on top to hold the stock and point rails in proper gauge.  Also note that I ended up with points a little longer than I'd planned.  When I made the first point, I bent the frog in the wrong location.  Rather than remaking the point, I opted to adjust everything else to suit.  All I had to do was to re-do the slight kink in the stock rail so that the point fit against it properly.  I could also have re-bent the frog, or reshaped the point, both of which I've done in the past.

The throwbar

This, I'm particularly proud of.  Until today, I've never been able to make a decent throwbar which had the required strength and flexibility, leaving my turnouts with a rather obvious weakness.  I think I've solved the problem, finally!

I started by cutting a small square strip of PC board, smaller than a regular tie.  Tight-grained wood would work just as well, and might be a little easier to cut.  Working in the fiber backing only, I drilled a couple #73 holes spaced the distance that I wanted my points to be.  I bent a couple pieces of brass wire at right angles, and slipped then through the holes, then bent the back sides over to trap the wires on the throwbar.

I then soldered the wires to the bases of the point rails, being careful not to accidentally solder the point and stock rails together!  While it's not obvious in the photo, I beveled the top edge of the throwbar so that the copper would not short the two stock rails together.  As I sit here thinking about it, wood would be a much better choice, since there's no way it can cause a short.

Here's the finished turnout.  To be prototypical, I should not have soldered the throwbar to the ends of the points, but it was a little easier.  I can always change it if I need to.  So, there you have it.  A hand laid turnout without the need for any fancy jigs, tools, or materials.  Once in place, I can cut and insulate the frog, and of course remove the braces from the tops of the rails.

Kudos for undercutting the points properly

It's nice to see working parts modeling done thoughtfully and with potentially long term reliability.

From a working model point of view, are you aware that 99+% of US turnouts laid before the year 2000 or so have hinged points? You could probably adapt your throwbar pivot design to do something similar for those.

Nice Job

Nice job Ken. You just showed how one does not need expensive tooling to have fun.

I'm going to be building 3 turnouts for my HOn30 portion of the layout, but I'm going to use PC ties. I cut my own PC board with a carbide saw blade in the milling machine. Ooops, expensive tooling, Dang.

Bernd

How long to make?

I followed Charlie's technique using PC ties and have now scratch built 4 turnouts (2 curved no. 8's and two regular #7's).  The first two took me a day each.  The next two, one day for the two.  Just out of curiosity, how long did yours take to make?  I'm trying to figure out how to speed up the process.

John C

Responses

I am, indeed.  However, I chose to keep the points continuous for two reasons.  They are simpler to make, since they don't require any hinge or flexible joint.  And they operate a little more smoothly, since the rail forms a smooth continuous curve.  Since I wasn't going for super accurate prototypical track, but just reliable and decent looking track, I feel the compromise is acceptable.

With the exception of the brass plates (which make things a LOT easier!) I developed this method back when I was a teenager.  I had to come up with reliable HOn3 turnouts on a very tight budget, so I sat down and figured out how to do it.  I haven't had the need to hand lay any turnouts in almost a decade, and now I can afford slightly better tools, but this method works well for me.  And they're still cheap!  I made that turnout out of rail salvaged from my layout and scraps of brass in my metal bin.  Total cost: $0.  Pennies, if I had to purchase all the materials new.

Hmmm..  Working off and on, between phone calls and lunch and stopping to take pictures..  maybe 3-4 hours?  If I sat down and worked on it in a more dedicated fashion, I expect I could have a turnout in an hour or two, no problem.  Filing the points is the most time consuming part of the process, and as I finished up the second point I found better ways of holding and supporting the point, which made filing go much faster.  It's definitely something which gets faster as you get more practice.

Since my method is pretty much identical to Charlie's, except for the points, I expect they'd both take about the same time.  I didn't count the time I spend laying ties, since I tend to do that in batches, and the ties for a turnout take me only a few minutes to glue down.  If I had PC ties, I'd probably be using them, but I don't, so I make do with what I have.

what about shorts on the throwbar?

I have some concern about your PC board throwbar.  I understand that you are putting the PC board on edge, so the copper side isn't rubbing on the bottom of the rail.  However, it seems to me that there's a chance of the edge of the copper cladding still making contact on both point rails at the same time, and that would cause a short.  Have you examined this possibility?  From that point of view a wooden throwbar would be safer.

I use PC board throwbars on my turnouts, but I cut through the copper on both the top and bottom between the two polarities to make sure there are no shorts.

Preventing shorts

I considered that, and to prevent it both edges are beveled so that the copper cladding never comes closer than .030" or so from the rail.  If I find that it's a problem, I can always cut gaps in the copper later.  In retrospect, wood would make the most sense.  My original plan had been to solder the point attachment wires to either side of the board, which is why I used it in the first place.  I think I'll use wood (possibly bamboo) on the next one.

With a lot of PC board

I know on my Throwbars I make 5 isolation cuts

Remember with DCC ready turnouts, the stock rails with points are all going to be the same polarity with the only isolated section of the turnout being the frog area.  If you are really worried about isolating a throwbar, then make it out of plastic or a good hardwood and thread a 00-90 flat topped screw and solder to point to the the screw.  I have done that back in the 70s with a very good level of success.

Throwbars

Very clever solution to strengthening the throwbar joint, Ken. Wish I had thought of this years ago. It appears that you could eliminate the piece of wire protruding on the bottom of the throwbar by simply making the piece of wire in the shape of an L and then inserting into the hole. The next time one of my switch points comes loose, I'm going to fix it with this solution. Thanks for the tip.

Tom Patterson

Why 5 cuts?

I'm curious, why would you have to cut 5 gaps in a throwbar? 

I've never had good luck soldering the points to a throwbar without some sort of mechanical reinforcement.  Remember that the point is VERY thin at the end when it's shaped prototypically, and you're soldering it on edge.  That's why I added the wire - prior to that, every point I soldered broke away from the throwbar sooner or later.

That would be easier to add from the top, but I like the fact that the points cannot rise up with the wire bent over on the bottom.  I considered using some small brass pins, but I didn't have any handy.

Ken A Response

Ken,

Thanks for the question, 'tis overkill, I know, but since I use Jeff Stone's Bitter Creek B-4001 Manual Ground Throws where I run my two electrical feeds up and form them into the locks for the throw rod and because Jeff's ground throws are all metal and can conduct current from the two feeds into the body of the ground throw and I solder my frog feed to the hold-downs of the throw I want to make sure that the mechanical rod to the points throwbar PC board is electrically isolated and I have found over time that I each of the two sections of throwbar under the stock rail and associated point electrically isolated from the other stock rail and points and since I managed to have a derailment during my design testing and found it was best to ensure the area of the PC throwbar between the two point be isolated from any current flow, I end up with lots of cuts so I don't fry anything near the throwbar. 

as I said overkill, but no more problems and I find I really like DCC compatible turnouts and it saves a bunch on extra contacts, frog juicers, etc.  Everything works very nicely and the small switch engines (MEW CM Class 91 0-6-0) and my super pizza cutter wheeled 4-4-0 standing in for UPD&G 331 work smoothly on my code 55 rail.

Just noticed that

there was (yet another)  turnout building article in the latest MR (that my colleague brought by).  AFAICS, Still, after all the times it has been done there, absolutely no mention of the important point pre-bending and proper filing that is so neatly shown and summarized in a tenth of the space by Ken here.

Nice work Ken. I like the way

Nice work Ken. I like the way the points and throw bar work. I've been struggling with this area myself and hadn't thought of threading wire through the edge of a PCB tie. How are these turnouts holding up?

So far, so good, but...

The points are holding up perfectly, Greg.  almost like they've never been used.  Which would make sense, because..  well..  they haven't!  I built the turnout and haven't had a chance to install it in anything yet.  However, I will say that this throwbar design looks and feels much more durable than my last one, which was to have the points soldered directly to a PC tie's surface.  Even on the workbench, that one seemed to have issued of the solder joint breaking, and pulling the point out of the rail joiner used as a hinge.

Bending the points into an

Bending the points into an "L" gives more surface area with Code 55:

Harold

"Now he tells me!"

By the way, there is a very good article that details your method in the June 1983 MR "Building Turnouts at the Workbench" by Lawrence Ballou.  

In any case,thanks for sharing your work.  Your results are excellent and your descriptions are clear.  You make it seem like good results are within reach of even someone (like me) with average skills.  And since I plan on attempting a few #3's for my el/subway layout, you've boosted my confidence.

Cheers!

Hand made points

I'm afraid I use Fasttracks' jigs.  But, if I were making completely handmade points I would still use the Fasttracks forming tools, for crossings, blades and stockrails, they just give such a good result, very quickly.  I use a belt linisher for the crossing rails and pointblades.  I think that the Proto87 point blade hinges and tiebars are worth using for tracks on display, but not for hidden trackwork.  I also cut my own custom made lasercut blocks of ties, as I'm modelling Australian railways, our tie spacing and size are quite different to US practice.  Well done all the same. 

Encouragement for when I need to build some points with odd geometry!

Stephen 

Sydney