The Reversing Loop That Didn't

Been away from the blog for awhile but had a fairly productive weekend adding power feeders, buss bars under the layout and basically wiring up section after section on the lower portion of the layout.  I had the reversing loop wired for awhile and added in the Digitrax AR1 Reversing Module this weekend.  It seemed like a simple plug and play, until it didn't work at all.  Hmmmm....  It was supposed to be a fairly simple plug and play.  Essentially, this is a compute module that senses the "short" caused when the loco wheels cross into the reversing loop and automatically adjust the power to match the leg where the loco entered.  In DC layouts, there is a manual switch that controls the reverse.  Automatic is much nicer. Well, I turned to my never disappoint me resource of N Scale Net DCC Forum for some help.

I posted my layout schematic (which will be used on my control panel) along with some labels showing where my "A" circuit was located:

I had several suggestions for adjustments to the AR1 and then a User by the name of Heiko made a startling revelation:  "Apart from your AR trouble, which you should solve anyway: The purple section is not your reversing section. The section where you will connect A to B and B to A is the horizontal blue track you reach when you go counter-clockwise through your purple section - or, if you want, you can add the purple track to the part switching it's polarity, but including the T/O between the purple and the horizontal blue track.  The purple section is "just" a siding to the black section it follows. So the AR should not be required at all for that section.

WOW!!!!  He was right.  I never realized when I converted the Atlas DC layout to my own layout, the area Atlas called the Reversing Loop - really was not working that way.  I posted some replies and settled on the following rework of the layout:

I had also read (on the Digitrax site) that it's a bad idea to have the insulated gaps of a reversing loop at the end of a turnout.  Well, I had my insulating gaps at the ends of a turn out on both ends of the reversing loop.  The rework shown above eliminated that issue and also made a true reversing loop.  The change added some length to the reversing loop, which translates into a longer train capacity.  The longest train on your layout is limited to the length of the reversing loop.

The downside of this rework meant tearing up track that was nailed down and roadbed that was glued down.  Had to do what I had to do.  It was a couple nights of work to rework.  Once I relaid the new segment of roadbed and laid the new track in, I had to go back and check all the track feeders.  So far, so good.  The moment of truth came when I wired up the AR1 module.  I powered up the layout and the first test of the reversing loop was a complete success!!!!  I had a few folks ask to see a video of the reversing loop in action so here it is:

You may notice that the loco has a millisecond of hesitation as it crosses the reversing loop insulated gap.  While standing at the layout, you can actually hear the AR1 make a click sound as the loco keeps right on trucking.

Another very valuable, and somewhat painful, lesson learned.......

Turnout Turmoil - Part II

So the Tam Valley Depot products came in yesterday and it seemed a little overwhelming at first.  Lots of electronics.  The first thing I noticed was that the "power supply" I bought looked like a standard laptop charger and the end connection would not fit into the Octopus III Controller.  The Octopus needed 2 DC wires.  Hmmmm.....  As I've mentioned, I'm not really an electronics, or even electrical whiz.  I quickly posted to N Scale Net and started using the vast resource known as Google.  After about an hour of searching, I found out that there is a way to add a connection to the end.  A quick run to Radio Shack and I ended up with this configuration:

Not to blow my own horn, but I dove right in and soldered the wires to the connection.  I ran a quick test using a regular Atlas Turnout Controller and one of my still motorized turnouts.  No sense in frying the $32 Octopus before I knew I was basically configured right.  Success!!  I added the wires to the Octopus and it fired up, ready to go to work.  As a side note, I saw a later post that said I could have just cut off the "charger" connection and had 2 wires to use.  Live and learn...

I had ordered one Tam Valley Switch Wright Servo Bracket, along with one servo and one Bi-Color Fascia Control button.  In looking at the servo bracket, it's a nice product and certainly holds the servo where it needs to be.  On the down side, it's $4.25 each and I have 14 turnouts.  Not to take anything away from Tam Valley, just not sure I want to invest the $55 on brackets.  I will say that the instructions were excellent and included a full size wire bend guide.  One option might be for Tam Valley to sell the bracket pieces for do it yourself assembly, thus reducing the cost.  They do that with their fascia controllers.  Here's the sequence showing the assembly process as I got the servo ready:

This is what comes in the Switch Wright Bracket package

SG-90 Servo in Bracket

Servo Horn with Music Wire (0.032")

So, here's the completed Servo, Horn and Wire mounted into the Servo Bracket:

One note about the servo and Octopus Controller.  I have never worked with servos before and I have to say, it took a little work but I think that was mostly my lack of knowledge and failure to read through the entire process before diving in.  With my order, I bought the Servo Centering Tool (battery operated device that will center the servo - OK, that was obvious).  I also bought the Remote Aligner which will allow the servo to be calibrated while installed in the final location and with the Octopus mounted to it's permanent location.  It comes with a 7' connector cord, which will be handy once everything is in place on the layout, especially since I have a couple turnouts that will be nearly 4' from the Octopus.  I spent about an hour last night learning to get the servo aligned and adjusting the left and right limits.  Servos can move up to 70 degrees each side of center.  With the turnout requiring only about 1/8" of travel, I need to really tighten up the range of motion.  The Tam Valley instructions for the Octopus II were excellent.

At the end of Turnout Turmoil Part I - I had installed a small extension on the turnout activator bar and planned on using the side wire mount method.  I reread Duncan's (owner of Tam Valley) article in the July 09 Model Railroad Hobbyist (great FREE online magazine - with every issue online).  I saw a section about having the servo wire in the middle of the activator slide bar.  It occurred to me - that's the way to go.  Those plastic extension were going to be nothing but a potential problem down the road.  I moved out with my test turnout and drilled out a small hole in the center of the activator bar.  Serious lesson learned - drill from the top not the bottom.  As the drill bit (hand drilled with a pin vise) came through the activator bar, it pushed off the top cap and completely trashed the turnout mechanism.  Great.  One turnout sacrificed.  Second attempt worked great and it was on to the test bed.  In order to mark the hole to be drilled, I cut the head off a push pin and ran it through the small hole.  This way, when I pulled the turnout off the layout, I still had something to show me where to drill.  Then, I just marked a circle around it with a Sharpie:

I pretty much knew better than to take the power drill to the cork and foam bed.  I used a 1/4" bit and just spun it in my fingers to get through those two layers.  Once I had the hole through the cork and foam, I used the drill to go through the 1/2" plywood base.  I used the included double stick tape on the bracket and lined everything up.  Once that was done, I still had about 5 - 10 minutes of fine tuning to get the throw distances correct.  Overall.  I think this is going to be an excellent system and it will look much more realistic than those damn chunky surface motors.  The Bi-Color LED lights on the fascia controller are really cool!  Red shows the "route" closed and green shows the route open.  Here's a not so great quality video but it still gives a good idea on how everything works:

Thanks to all for the great encouragement I've been getting.  I recently discovered the N Scale Model Trains group over on Facebook and have had some great suggestions.  I just added a Follow Me by Email link to the Blog.  If you like what you see, and want to see future updates, go ahead and add your email address.

One last detail I just remembered - if I ever decide to get really advanced, the Octopus III is adaptable with a DCC decoder.  Not sure I'm up to running DCC turnout control - but the option is there.

Turnout Turmoil

So having a background as an architect, combined with a lot of years as an architectural model builder, has left me with a rather critical eye for detail and accuracy.  As the layout started coming together, I couldn't help but keep looking at the Model Power turnout motors and thinking there had to be a better way.  As you can see, here's a 40' box car in comparison to the turnout motor:

Where would you ever see a 40' long x 5' wide mass along side a railroad track? No where!

I started doing a lot of research and found there were numerous ways to control the turnouts.  Most of them were expensive.  Multiple expensive (about $30 per) times 14 turnouts = too much $$$.  Relying on the ever useful N Scale Net site, I cam across the process of using model airplane mini servos to control the turnouts. I looked at Tam Valley Depot and saw they looked to have perfected the process.  They had a great article in the July 2009 Model Railroad Hobbyist Magazine that really showed the entire process.  I was also really impressed that Duncan (Tam Valley owner) pointed out, on his site, that you could get servos off Ebay in bulk for lower costs.  He even described what to get off Ebay.  Amazing!

Fortunately, I have a few extra turnouts and can do a little research without sacrificing the layout.  I started by pulling one of them apart to see how they were built.

Underside of the actuator mechanism

The first thing I noticed was that the Model Power turnouts were built a little different than the Atlas version.  Even though there was a slide activator for the tracks, it was a really short piece.  I found some great info on the servo method from Jeff over on N Scale Net who shared some pics as well as his diagram:

Server under layout and brass guide tube

Notice the longer actuator bar on the Atlas turnout

Jeff's actuator mechanism

Looking at all this, it looked like a fairly easy solution to controlling the turnouts.  I went ahead an ordered an Octopus Controller (can handle 8 servos), one servo and a fascia mounted actuator button with dual LED lights.  I wanted to be sure that the system will actually work, although Duncan from Tam Valley swapped a couple emails with me and gave me some pointers that let me feel like everything would be good.

Since the slide activator bar was so short, I considered having the wire run under the turnout and come up to the bottom of the slide activator:

Slot and hole added to the bottom of the actuator bar

Slot test with wire test

Although the slot system worked, adding the wire quickly revealed that this wouldn't work.  There's no way the wire would twist on axis and still move the turnout.  That led to what will, in all likelihood, be the final solution - a .040 (same thickness as activator bar) x .25 styrene extension:

The only critical issue with this solution is getting a really good bond between the styrene extension and the activator bar.  There's only a little less than 1/16" of activator bar exposed when the turnout is thrown, so no real way to lap joint.  I use SC-125 acrylic adhesive which is sort of like super glue on steroids.  It's the same consistency as water and dries in about 5 seconds.  So far, 50 - 50 on confident bond.  There won't be a lot of torque on the extension but it needs to be stable enough so as not to have a continual problem one the turnouts are in place on the layout.

Next phase of this will be the actual op test with the Tam Valley products, which are due in this week!