With the cork done, the next task is to install switch machines, which requires laying out the track temporarily, so as to accurately position the turnouts and find exactly where to drill holes through the foam base for the switch machines. Ziggy's right paw is close to the blue line marking the location of a turnout control. This is where a hole must be drilled through.

With the locations marked, it's time to drill. The hole is carefully located where the wire from the switch machine will operate the turnout points. It needs only a half a centimeter of throw, so a 3/8-inch (10-mm) drill bit is adequate. It can easily be reamed out a bit if need be, anyway. The foam is very forgiving.

The next task is to figure out how to mount the switch machines. I am using Tortoise machines, since they are enclosed and have mounting flanges. After a great deal of head-scratching, I decided to mount them from below by creating a keyway for the switch machine housing to fit into. This worked nicely, and the top of the Tortoise machine box is flush with the bottom of the top layer of foam, exactly 1" below the top of the foam (bottom of the cork) which is the perfect height for the wires that come with the machine for operating the turnout.

Here's how it works: First using the holes drilled through the foam for a guide, locate and orient the switch machine key slot, and mark it out. The machine can be oriented either way, but must be aligned parallel with the track.

I found the best tool for cutting through the lower two 1" sections of foam cleanly and at once was a lowly pocketknife.

The switch machine simply goes up into the wide part of the slot, and the flanges catch between the top and second layer of foam. Then the switch machine slides back into place, and is held firmly by the snug cut of the foam. It is best to cut a bit small and make it bigger if necessary. Too big and you're sunk. Once the machines are positively located, they can be glued into place with a bit of foamboard cement on the edges.

I did that for all 21 switch machines. Whew! Then, each gets electrical wiring soldered up, and the stiff control wire installed, and the turnout is ready to install on top. Each turnout connection required some fine tuning, of course, but after the first couple of installations, the others went quickly and with only minor tweaking.

Later, after ballasting the track, I had to revisit the turnouts, since in the process of arranging and cementing ballast, and despite my best efforts not to foul the mechanisms, about half of the turnouts no longer worked smoothly. That is, the wire from the switch machine would move and impart a force on the turnout's throwbar, but the points would not move or would not match up snugly as they switched sides. This is unacceptable. I reworked the ties and moving rails within the turnouts to clear things up, but some of the turnouts remain stubborn and have not returned to their previous levels of performance.

I talked about this issue with the nice folks at Circuitron, who manufacture the Tortoise Slow-Motion Switch Machine, and was told that their solution to stubborn turnouts was as I suspected: Use a stiffer wire. They recommended 0.036‑in (0.914‑mm) diameter piano wire, or even 0.039‑in (0.991‑mm). (See Tortoise tuning below.) But I am getting ahead of myself...

At last all the turnouts could be located and installed. The track between the turnouts was simply filled with flex track (and that is pretty simple), and then all of it got wired up (not so simple). Which is what I did, but realized later that I should have done more work at this point, fixing the electrical connections within the turnouts, wiring the frogs to switched track power, and adding supplemental wiring to the points. But all that would come later—for now I was fixed on having some track to roll stock on, even if it was not powered.

Ah—the switch machine operation wires could finally be mated to their respective turnouts, and I could finally see how this was going to look. But I couldn't really run trains around until the switch machines operated. Which meant I had to start on the electrical stuff. That would have to start with a completely new design campaign, from household 110‑V power to transformers to switching on a control panel...

Again, I am going to deviate from the chronological order of things, though. The quick story is that I proceeded to wire up the switch machines, and had a grand time switching unpowered track. Then, of course, I had to power the track, which was another large effort. All that is covered in the electrical section. But I found, after wiring up the track, that there was a problem. I had assumed that these fine Walthers Shinohara code-83 nickel silver top-dollar (and, sadly, no longer available) turnouts would have adequate power, drawing from the powered track to which they were connected. But, no! There were dead spots, and of course the frogs were not powered, and the engines stuttered across the turnouts, often coming to a complete stop.

So, it was time to figure out what to do next. As usual, the Internet came to the rescue, and I found this advice on wiring the new Walthers DCC-compatible turnouts, by Allan Gartner. (You see, I had got the new DCC-compatible ones, thinking that perhaps some day this layout might be converted to DCC, though I really like the operational challenge of a DC block-wired layout.) Anyway, so I read up on all the things that I have to do in order to get these turnouts to work. No—they do not work right out of the box. Oh. They are lovely, though, as seen in this scan of a #4 turnout:

There were three wiring tasks that needed to happen to each of these, as I learned from Allan Gartner's sage advice:

• Each had to have its cross-wiring checked, and fixed if necessary. This is mentioned in the aforementioned wiring guide's section called Check Your Turnouts for Good Electrical Bonds. OK.
• Each point had to be wired (with a flexible wire) to the rail that feeds it. Oh, please! Although this seemed like overkill I am glad I did it. (This was advised for "long, reliable operation." Well, who doesn't want that?)
• And finally the frogs needed to be properly powered with track power, so that the polarity changes with the position of the turnout.

So I did all that. Here are the details, for those of you who are still with me:

Fixing the Cross-Wiring

A very cool thing about these turnouts is that there are wires connecting the rails electrically, hidden within specially-designed ties. That is, the right-hand rails are wired together across the entire turnout, as are the left. At least, that is the intention. Someone at Shinohara, however, was not getting the soldering done right, and someone else was not checking them sufficiently. In other words, about one third to one half of these connections was improperly made. I checked each one, and resoldered (in my own comparatively clumsy way) the ones that did not conduct electricity in the first place, or came loose with a gentle tug with a dental pick. By the way, dental picks are of great utility in this hobby. Ask your dentist for any dental tools that are no longer worthy of dental work. She'll probably give you a small pile of them. While you are at it, ask her for expired dental impression medium—this is great for making impressions of rocks for scenery.

I apologize for the lousy photo. Here is the double cross-over after having undergone some rather unsightly cross-wiring surgery. Ignore the brown wires. The big blobs of solder are my ham-fisted fixes to the bad crosswire connections.

Wiring the Points

The points of these turnouts are connected to the rails by small rail connectors, barely engaged at the rail end, so that the point may move freely. See detail:

This little connector cannot be relied upon to provide unfailing electrical service to the point for decades to come, so Allan Gartner recommends giving it a bit of help. This comes in the form of attaching wires on the underside of the turnout, like so:

Of course, these wires extend below the base of the turnout, so the roadbed has to be cut with a little trench or divot to accommodate the wires, so that they can move freely. After all, the points are still being moved only by the switch machine's control wire. Again, the trusty pocketknife is the tool of choice, and the divots are cut after carefully marking their position through the ties:

This led to the next quandary. These divots will easily fill with ballast later, which might cause them to stop working as intended (that is, move freely). So, I installed a small sheet of wax paper between the track and the wires, which would help keep ballast out of the divot. I hope it works!

Update following ballasting, years later: Enough time had passed between this turnout work and track ballasting that I had forgotten why I put the wax paper in place, and for some reason did not think to look at my notes. So I foolishly removed it. I fear that some ballast (or at least glue) got into these divots, causing issues with the switch machines (addressed through heavier wire as mentioned above.)

So, I did that for all 21 turnouts, too. But wait! Before I could install them, I had to do the frogs, too.

Wiring the Frogs

Wiring the frogs, and in a few cases adding additional track power wires to parts of the turnouts, meant that I had to run wires through the foam slab, meaning that I had to locate the holes with care. In the end, I found it easiest to lay the turnout on the roadbed, carefully positioned, then drill a hole, with the turnout in place, where it was needed. I used a super-long 1/8‑inch drill bit, which allowed me to drill the holes at odd angles, which was sometimes necessary in order to avoid a switch machine or piece of benchwork below.

I fashioned some heavy gauge solid copper wire for making the leads, with an offset double bend at the end to fit neatly into the outside profile of the rail. Again, it's not a good photo, sorry.

While it was a huge pain in the neck to follow Allan Gartner's advice, I believe it to be sound, and I expect these turnouts to be wonderful and trouble-free forever. Or something like that. But wait! There's more to be said about the turnouts.

Turnout Tuning

There is much to be said (don't worry—not too much) about the mechanical operations of the turnouts. First off, connecting the control wires to the throw bar that the points are connected to is straightforward. In fact, having mounted the switch machines exactly 1 inch below the roadbed (making the top of the switch machine 1.25 inches below the base of the turnout and track), the control wires that came with the Tortoise switch machines were the perfect length. Well, actually, they are just a bit too long, catching on the occasional low-slung rolling stock and on cleaning cloths. You can see the control wire protruding through the throw bar at the lower right in the picture:

The white dust that I added is powdered Teflon, to serve as a lubricant. A locksmith turned me on to the stuff, and I picked up a small bottle (a lifetime's supply) on eBay for a few bucks. Just don't breathe the stuff—it's quite toxic.

There is more to say on turnout tuning, however, following the tuning and ballasting of the track. Ballasting involved the oh-so-tedious careful arrangement of each bit of ballast, and it seemed that despite my best efforts, some would get between the points and underpinnings of the turnouts. On top of that, there was a good bit of sprayed and poured glue going on, and that got on the track itself, as well as between the points and underlying ties.

Cleaning the glue off of the track was straightforward. I scrubbed it well with INOX, and had to resort to attacking it with some wooden popsicle sticks and tongue depressors to remove the more stubborn bits. The real test came when running locomotives around at a very slow speed, which would identify any bit of track that was not conducting properly.

Then came the turnouts. Several of them were behaving badly, including many that had worked just fine before the ballasting. How frustrating! I cleaned them as best I could, but they were simply not throwing consistently. Was there a way, I wondered, to get the Tortoise switch machines to apply more force to the throwbars? As I hinted at earlier, I turned to the fine folks at Circuitron (maker of the Tortoises) and learned that they had a solution for just this problem: Use stiffer wire. I ordered some 1-mm diameter piano wire and some crimp-on electrical fittings (loop end 18-22 SWG wire terminations) and made up 24 replacement wires. Details below.

But first, let's look at the original Tortoise wire setup. Here we see a standard Tortoise wire from under the layout. The turnout throwbar is above the hole in the pink foam at the top, and the wire extends down through the fulcrum hole. The bent end feeds into a small hole in the black plastic moving part that sticks out from the machine:

This wire is held in place with a screw:

I removed the screws and extracted the control wires:

It turned out that the holes in the turnout throwbars were just a bit too small to accept the 1-mm wires, so I drilled them out with a #56 bit on a small finger drill. Here's the bit:

and here is the drill, working from underneath:

Now on to making up the replacement wires. Here is one of the original (thinner) Tortoise wires (note the bent end) next to the new 1-mm wire, which is longer:

Cutting the piano wire was no joke. You can forget using your standard dikes; they will only get chewed up. I followed the advice by my new friend at Circuitron and got these cutters: A 7-1/2-inch WORKPRO Cable Cutter for hard wire ropes, steel wires, and aircraft cables. This did the job quite well! Making up the replacement wires involved crimping on the loop termination. This is best done with a high quality crimping tool. Again, getting the right tool is essential to avoiding frustration and poor quality results.

Now the new wires are installed on each switch machine, securing the loop end under the screw and threading the top through the fulcrum hole and into the throwbar above:

I had deliberately cut the wires to be about 1 cm too long, so that they would protrude above the track. This would be best for testing purposes.

Once I had verified that all the new turnouts worked well with their new wires, I could trim them to fit. But how? The WORKPRO Cable Cutter could cut the wires, but its short and thick cutting shear was too bulky to get close to the track. There was no way that I could remove the wire and cut it to exactly the correct length, especially since each turnout was a bit different. I experimented with my Dremel cutoff wheel and some extra plastic ties. Indeed, if I was not careful, the wire would heat up and melt the plastic. The key to avoiding this is to cut quickly. I found if I just did it fast, the plastic would not melt. So, I cut off all the protruding wires with the Dremel cutoff wheel:

The wires are cut nearly flush with the top of the throwbars, with less than 1 mm protruding. There is one exception where the throwbar in the 3-way Shinohara turnout (shown in the photo with the protruding wires) has to actually move under the rail. That one had to get cut exactly flush, and I was really worried about melting the throwbar. It would be a very difficult job to remove that turnout for repair! Fortunately all went well. Even better, ALL THE TURNOUTS NOW WORK FLAWLESSLY and consistently. Success!