M&K Junction Railroad

M&K Junction Railroad
Another train of eastbound coal crosses the Cheat River

Thursday, June 30, 2011

Installing a DCC Decoder in a Brass Steamer

Creating the videos below was a long and painful process, but they're finally done! The videos take you step-by step through the process of installing a DCC decoder in a brass steam locomotive - a Sunset/3rd rail B&O Q4b in O Scale to be precise.

There's no reason to be hesitant to disassemble and modify a brass locomotive; and the Sunset locos seem to be well soldered and sturdy. Hopefully these videos will give you the information and confidence that you need to tackle the job yourself. In fact, getting the installation right took me a couple of attempts to find the right combination of miniature connectors and flexible wire to make the installation work - but that is all explained in the film. Be sure to view the videos in order.

There is one omission that I must mention. In hooking up the backup light I forgot to film (or mention) the necessity to include a resistor in the line going to the backup light. This resistor is necessary once the backup light is hooked up directly to the decoder. Use a value of 500 to 1000 ohms and put the resistor in either line of the LED backup light.

In viewing the videos, do not click on the "HD" unless you have high speed internet and a fast computer (dual core processor and good video card); and you still may want to allow the video to buffer before playing.

Part 1

Part 2

Monday, June 20, 2011

Fast Tracks Tips and Techniques

A few months ago I purchased a full set of the Fast Tracks turnout construction fixtures. The point/frog forming tool, the stock rail forming tool and the turnout fixture itself for a #6 switch. The videos on the Fast Tracks website provide a good overview on how to use each of these tools and I watched them before buying.

However, the O Scale fixtures are a little different then the ones for the other scales, and are not covered by Fast Track's videos. In addition I developed some techniques for using Fast Tracks that may be helpful. So I made this video to demonstrate all that I learned.

One more tip that came up after I finished the video. I installed a bunch of these turnouts in a yard ladder and as part of the installation I checked for shorts. While I did not find any dead shorts, I found some leakage between the left and right rails. This drove me nuts for a while until I isolated it to melted flux bridging the gaps in the PC board ties. Since the Oatley flux is a tinning flux, it's mildly conductive. So if you use the Oatley tinning flux, be sure to clean it out of all of the gaps (or use less flux than I did). I found that "Goo Gone" is a good cleaner to use for this purpose.

This is the first video that I've posted that is compatible with the iPhone & i Pad. If you have trouble loading or playing this video, please leave a comment outlining your problems so that I can correct them. Like this video? Leave a comment. Don't care for videos? Leave a comment. Like videos compatible with mobile devices? Let me know; if there is enough interest I can go back and change other videos.


Wednesday, June 15, 2011

A Few Simple Solutions

Mounting Tortoise Switch Machines

When I first started building this layout, I had to put in 3 single crossovers, requiring 6 switch machines. I spent a considerable amount of time under the layout affixing switch machines - and bumping my head. Now that I'm working on the yard ladder into hidden staging and I've come up with a simple and obvious way to attach switch machines without all of the hassles.

The problem with attaching the Tortoise machines is that the built-in mounting ears are too close to the machine to reach easily with a screwdriver. Plus the machine is small and it's hard to position it critically and drive the screws without pushing the machine out of place. To alleviate these problems I've attached each Tortoise machine to a scrap of plywood. The size of the plywood is not important so long as it extends well past the machine proper. The plywood is, in turn, mounted to the underside of the layout. The plywood forms a 'handle' that makes it easy to position and screw the machine in place without movement.

These photos should be self-explanatory.

You'll notice in the pictures above that the actuation mechanism of the Tortoise machine simply extends past one edge of the plywood mounting plate. In this way it is not necessary to drill a hole for the actuation wire through the mounting plate. Also note that I have pre-drilled and countersunk holes for the mounting screws. Four holes are not really necessary, but it provides some flexibility in where the mounting screws will be driven. The combined thickness of the mounting plate, subroadbed (plywood) and roadbed (cork and foam) is well over 1". This would be too thick if I were using the music wire actuator that comes with the Tortoise; but I am already replacing the actuator with heavier wire, so I simply cut this wire longer.

In O Scale a #6 switch is approx 18 inches long and track center-to-center spacing is typically about 4"; so switches cannot be densely packed and there is plenty of room under the table for these oversize plywood mounting plates. In the smaller scales, this approach may not be possible.

Mounting Servos as Switch Machines

I'm taking the opportunity of motorizing all of the switches in hidden staging to experiment with servos as switch machines. While I will eventually experiment with the servo actuators from Tam Valley Depot and Duncan Mcree, I currently have an ESU stationary DCC decoder that drives servos. To use the decoder that I have I had to buy a servo at the local hobby shop (ouch!) rather than pay less and wait for it to arrive via mail-order. How to mount the servo was a problem. After thinking about it for a while, I decided on the following technique. I simply glued the servo to a right angle bracket sold in home centers for building decks.

The bracket is Simpson 'Strong Tie' A21Z, note: this is the right angle bracket with one arm longer than the other, that's the arm that the servo is glued to. I used Gorilla Glue to attach the servo. I roughed-up the bracket and servo with sandpaper before I dampened the bracket with water and applied a THIN coat of the glue to the servo, then clamped the two together. Since I used a thin coat, this time the glue did not foam up all over the place; I've finally found a use for Gorilla Glue.

Here's the servo switch machine installed:

For fitting the actuator arm, made of music wire, to the servo crank I came up with what I think is a novel solution which you can see here:

I bent a length of wire into the shape above. The length of the straight section to the right should be enough to reach up through the subroadbed, roadbed and through the throw bar on your turnout. If in doubt, make it longer, any excess can be cut off after installation. The form of the wire on the other end is shown in the detail close up here:

The distance between the bends should match the hole spacing on two opposite arms of the servo crank so that it fits like this:

In case it's not obvious how to get the bent wire into the servo crank, here's a short video on how it's done:

The self-retention feature that you see in the video works because I am using relatively stiff 0.039 wire as a actuation arm. That allows the right angle bend at the end to exert enough outward pressure to hold the arm in place. If you use significantly thinner wire this might not be the case.

Wednesday, June 1, 2011

This 'n That

It been a while since my last post; 3 of the 4 weekends in May were taken up with activities away from home that kept model railroad progress to a minimum. That is not to say that I got nothing accomplished. Here are photos of a scratch built 62"/48" curved turnout that I started in May and, technically, finished in June. The paper template for the turnout is laid upon the back of a 2' x 4' ceiling tile. Why a ceiling tile? My last scratch turnout was laid down on a piece of plywood and each spike hole had to be pre-drilled - a real pain. The spikes (actually "Perfect" brand track nails) push into the ceiling tile easily and hold well enough to solder the turnout; when the turnout is finished the spikes can be removed and reused. The ceiling tile is, in turn, supported by plywood as it would bend and break otherwise.

Here's the frog laid down:

And a close-up of the frog:

Stock rails in place:

One Closure rail in place:

The Completed turnout:

Close-up of the frog-wing-guard rails:

This latter picture shows the frog before the excess solder was cleaned up by filing and sanding.

The Fast Tracks frog/point form and stock-aid tools were used to aid construction. I have to say that I enjoy hand laying turnouts; I just wish that I was more adept at it as it still takes me far too long. However getting a smooth running turnout that is an exact fit for the location is nice. For instance, the 62" outer radius exactly fits the curve where this turnout will be located and the turnout was made with the code 156 rail found in the old (1970's) Atlas flex track as the hidden trackage is the old Atlas flex.

Another May project: I did some mold making. A couple of months back, I saw some "Ultimate, Make-A-Mold Master Castings" rock castings by Ultimate Scenery Center (www.ultimatescenery.com) on a vendor's table at a train show. It was love at first sight with these castings; unfortunately, I had expended all of the money that I wanted to spend that day. I recently found some of these castings at Mainline Hobbies in Blue Ridge Summit, PA (BTW an excellent train store) and snapped up a couple.

These are highly detailed, urethane foam castings that the manufacturer actually intends to be used as masters for the purchaser to use to make his own molds. I like this concept, no longer do you have to be in a gray area of legality or morality by making a mold from someone else's casting.

I started with #123 "Pennsylvania Shale", a large (13" x 9 ") casting exhibiting both stratified and 'blocky' rock in the same casting. It is a good match for the rock formations found along the rights-of-way that I model. Here's a photo:

Here's a detail picture:

Click on either photo to get a full-size view.

I was going to use the technique that I used in my O Scale Trains article to make an open-face mold, but the manufacturer included detailed instructions for making a reinforced latex 'glove' type mold, so I decided to follow their instructions.

Here's a photo just before starting the process. I used the same 'Mold Builder' latex mold compound that I bought at Michael's and used for the OST article.

Mold Builder is not the manufacturer's recommended mold making compound, but I had it on hand. I don't know if the substitution effected the outcome. The surface that I am building this mold upon is a 18" x 18" ceramic floor tile obtained from Home Depot. I chose the smoothest 18" x 18" tile that I could find, although this was a floor tile and not a glazed wall tile like I build smaller molds on. To make the tile's matte surface smoother, I coated it with clear gloss spray from a spray can. A smooth surface will prevent the latex from adhering and facilitate removal of the finished mold.

And here's the first coat of latex applied.

The latex is brushed on. The first coat of latex must be applied thin and without bubbles - unless you want bubbles to be a permanent part of the mold and of each subsequent casting; so watch your brush strokes. I extended the latex beyond the sides of the master casting to form a lip on the finished mold about 3/8" wide.

Here's a pic of the mold after a couple of coats of latex have been applied and dried.

Ultimate Scenery instructs to apply four coats of latex before the next step. Being a 'belt and suspenders' type, I applied five.

Next you are instructed to apply a coat of latex and wait for it to get 'tacky' then layer the entire thing with gauze. The gauge is supposed to be forced into all of the cracks. Even though I cut my gauze into 3" long strips for ease of manipulation, this master casting has so many small cracks that I could not get the gauze down into all of them; it would lift out and/or bridge over the detail. Maybe the latex that I used was not tacky enough.

After the gauze is applied, you're supposed to add another coat of latex before - as I read the instructions - the latex/gauze coat has had time to dry. I did this and it lead to further lifting of the gauze from the master. Here's a picture of the mold after adding this overcoat of latex:

After the latex/gauze/latex coat as dried; four additional coats are to be added before the mold is done. Just to be sure, I added six.

Here's what the finished mold looks like ready to be peeled from the master casting:

And here's a video of removing the mold from the master casting:

Making the mold took a lot of latex, the lion's share of the jar was consumed in the 14 layers of latex that I applied.

So now it was off to make a casting from my new mold. I mixed up a batch of plaster in an empty Cool-Whip container and I thought that I might need two of these to fill the mold. Mixed the second batch and it was just about enough to cover the face of the mold, but not fill the body. I switched to a larger mixing bowl and two additional large batches later the mold was filled. Here's what it looked like:

The chocolate pudding color of the plaster came from a failed attempt to color the plaster with the craft store acrylics that you see peeking out in the upper right of the picture. Actual rock samples and photos taken from the area I'm interested in are a chocolate type brown, but darker than the color I got with the acrylics.

Filling the mold took a LOT of plaster - the majority of a 5 lb. container is in that casting. Next one I may only fill the face and not the entire mold. After 24 hours to let a casting of this size dry and solidify, I took it out of the mold. Here's what it looks like:

The casting is not bad, if you can look past the weird color (which will be fixed by spray paint). The details seems to reproduced well, however there are a few bubbles that will need to be fixed, probably the result of my pouring technique.

A final observation: If you look at the pictures of the mold, you will see that some gauze extended beyond the latex rubber. I cut most of this away, but did not seal the ends of the gauze with additional rubber. During casting the gauze absorbed water and wicked it into the mold between the layers of rubber. This made the mold slimy and it took on a milky color. Now two days after demolding the mold is beginning to dry out and return to it's normal state. It's possible that this is responsible for the chalky white deposits that you see in the close-up. I find it hard to believe that water between the layers of latex made the mold slimy, but it did. Next time I'll try sealing the gauze completely.