Sumida Crossing

[KenS]

Hi, I'm Ken, and my particular area of interest is the urban trains of Tokyo, particularly the JR East commuter/rapid trains, and the limited expresses. I'm not entirely immune to the lure of Shinkansen or trams, but that's secondary for me.

I started building my first permanent layout for my Japanese trains a few months ago. So far, Sumida Crossing is just benchwork and some initial pink foam, with the first loop of track laid (using unitrack) but not fully divided into electrical blocks. Although I have a track plan, it is still a bit fluid.

What I'm aiming for is to be able to run a number of trains on separate lines simultaneously, with several double-track continuous loops (so things can run unattended on one track, while I play with a train on another). My current intent has a "subway" loop that's mostly below ground, and two surface/viaduct loops, one representing a local commuter line, and the other which can do double-duty as a rapid or shinkansen line.  Finally, a short freight-only branch featured in my earlier plans, but I ran into some clearance issues with that, which caused me to change my plans. There's also a long-term plan for under-table storage with a helix down, but that's well in the future.

I've attached my latest track plan, which varies a bit from the one presently on my website, as I've been rethinking the freight line as a possible tram line. This was drawn in the previous version of Rail Modeler, so I had to mark elevations manually on the diagram (the "+2" numbers represent height in inches above/below the top level, which is 2.75" above the wooden table-top). One of the things I'm still considering, not shown here, is lowering the "urban station" to be 2" above the subway level, rather, rather than 2" above the top foam, and making the subway tracks "street level" in this scene, underneath the viaduct station. That has potential clearance problems I haven't fully explored yet, but I think it would make it more visible/interesting than just looking at the underground station through a cut-out in the layout fascia.

As you can see, the right side of the diagram (the "unsceniced end") is still very rough, and I haven't worked out all the connections there. I'm likely to make use of flex track to get things to link up.

I should mention that while the "riverside station" draws inspiration from Ochanomizu station on the Chuo line, nothing here is intended to exactly replicate the prototype. In particular, the subway will be connected to the surface line, allowing run-though operation, something the subway under Ochanomizu wouldn't offer, since it's a different gauge and uses third-rail power.

Since my experience with Unitrack and n-scale trains is only 18 months old, I'd certainly appreciate any comments about the operational characteristics of this plan, as well as any other ideas for how to make it better.

[Bernard]

Ken - Welcome to the forum and you have put together a very detailed track plan. My first question is how much clearance do you plan to have between the upper level to the lower level where you have your yard? I ask this because think about the room you will need and working with ease to get to the trains, track, turnouts, etc.

I took a brief look at your website and you are off to a good start. I see you are thinking about DC for this layout, you might want to take a look into DCC, it might be easier in the long run with a layout this size.
Again welcome to the forum.

[CaptOblivious]

I think having a proper underground subway station could be very interesting. Provides lots of opportunities to experiment with lighting, and getting the ambience just right. My original plan was to model Monzen Nakachou station on the Tozai subway line, as I spent a lot of time there when I was in Tokyo (closest station to my apartment). So I am inclined to say, try your original plan. But that's just my own selfish thinking.

[Mudkip Orange]

I think the quadruple track around the left side is awesome and I look forward to seeing construction progress photos.

[bill937ca]

A table 24 inches wide seems very tight for 4 radius of track.  Have you actually laid out all the track?

[Toni Babelony]

Looks like a really interesting project! You can run quite a variety of trains on the line as well. From locals to the 189 Series Express-Through from Shinjuku Sta. to Tokyo Sta. over the Chuo-Commuter-Line.

However, please be careful running Shinkansen mixed with normal trains in mixed traffic; it's quite unrealistic. But in the end it's all your layout and you can do whatever you want with it

[David]

A table 24 inches wide seems very tight for 4 radius of track.  Have you actually laid out all the track?

I couldn't find the table dimensions listed, where did 24" come from? I see curves marked 414/381 along with a double crossover (and inner curves marked 348/315), and I see a station for 11 cars with what works out to 248mm Kato straight sections marked off. Working from those two figures I get a depth estimate of somewhere between 42" for the track and 48" for the table being shown (assuming all the whitespace is table). The smallest radius I see is the 11" loop leading to the tram - everything else seems to be 12" minimum (~oh, and it looks like the subway has some 11" curves too).

[bill937ca]

Just out of curiosity, where do you buy your Kato trains and buildings?

[Mudkip Orange]

I couldn't find the table dimensions listed, where did 24" come from?

The guy's using modules 24" x 48". You can see them donated by black lines on the track plan.

The total table size is 4'x16', which of course has enough room for two R381/414s with a double crossover in between.

[KenS]

I'll try to answer all of those questions:

Bernard, my current plan gives about a foot of clearance between the bottom of the layout deck, and the top of the storage deck. That takes a 5.5 loop helix of double-track viaduct track (or equivalent flex) to reach down to it.

CaptOblivious, I'd still have the subway station in front of the Riverside station, it's only the one under the other scene that I'm thinking of bringing to the surface. One of the things that inspired that is Keikyu Kamata station, where they recently raised the Airport line to avoid a grade crossing, creating a station with tracks above other tracks. The Toden Arakawa crossing under the Yamanote line's Otsuka station also had me thinking about two-level stations.

Mudkip, I'm looking forward to seeing four trains passing on that end curve myself. That'll be a payoff for all the work it takes to get to that point.

Bill937ca, as noted the structure is an oval with two 24" tables back to back, so there's 48" for the end curves.  I started out buying through Newhall Station (and, aside from their costs, I was very satisfied with the service), but once I found out about the alternatives thanks to this forum, I've been using Hobby Search.  However, I've found a few trains (and most buildings) domestically, through MB Klein and dealer's tables at model railroad shows. Lately I've been putting all my money into layout construction, and haven't bought a new train since October (I broke down and bought the Chuo E233 then); I think I'm going into withdrawal. 

Toni, I'll have to move the Shinkansen through the commuter line to get to the storage tracks, at least as currently planned and assuming I have more than two Shinkansen (today I have one, a 500 series). I couldn't work out a way to make a direct link.  But I plan to do that at start/end of operations, rather than in the middle, because I agree that it would look seriously out of place next to one of my E231 trains.  Since my focus is on the commuter/express trains, my thought is that I'll just set up a couple of Shinkansen running in circles for background motion, and mainly play with the commuter and subway loops (and occaisionally stop one of the Shinkansen at the station).

David, yes I tried to keep all of the visible curves to 315mm (12") or larger radius, and the Shinkansen track is 381mm (15") or larger to give the best appearance to long cars. I'd really like to have even broader curves, but even if I was willing to give up Unitrack, I just don't have the room for it. The subway does have 282mm (11") curves, but those are all hidden from view.

[Mudkip Orange]

The subway does have 282mm (11") curves, but those are all hidden from view.

In fairness almost everything you can buy that would actually operate on a subway is capable of operating around R249 or even R216. I haul my E531s around R249 all day and have never had a problem.

[David]

In fairness almost everything you can buy that would actually operate on a subway is capable of operating around R249 or even R216. I haul my E531s around R249 all day and have never had a problem.

I've run the E531 and E231 Jobans down to Kato's R216 (8") and they still work. Getting some samples of Tomix mini-curve (R177, R140, R103) this month to see just how far some of my stock can go. Apart from S-curves the only thing I've found that doesn't work on Kato R216 is my 800 series Shinkansen (on R282 [11"] the cars start to "lean" into curves as the couplers are trying to pull them off the tracks).

[KenS]

I've been mocking up the left-end curve, to see how it looks, and to make some clearance checks on the catenary poles I'm planning to use (Kato's older "narrow" style 23-059 and 23-060 sets, rather than the "wide" style, 23-061, introduced with the superelevated track). I still haven't decided on whether to sink the "urban station" scene down, and put the subway tracks at ground level in it, but this mock-up is what the curve would look like if I did.

The houses might be at the level shown (2.75" above the river), but more likely I'd drop them an inch or so, and have the tracks on a slight embankment above them at this height, partly to focus the eye in front of the tracks, and partly because embankments or cuts seem to be the norm in Tokyo, at least for the major lines.

The other alternative, which I haven't mocked up yet, puts the tracks about 2" higher than shown here on the urban station side (near left in the photos), and drops at least the inner one, and possibly both down to this height at the riverside station scene (far left). That's about a 2% grade if I start and end as far left as possible, but it puts the bridges on a grade, which is one of the reasons I was growing dissatisfied with the original plan.

Right now, "level" track on the curve is looking much better to me, but I still need to mock up the other version for real (I did something a while earlier, but it wasn't this complete).

Note: the lower level tracks on the left are the subway tracks, which would be "underground" here, with a street above them, and a road bridge crossing the river to the houses.

[Mudkip Orange]

Lookin' good.

[Bernard]

Ken - I like what you are doing with testing out the layout before making it permanent. I'm not sure if I understand what you are describing with your alternate plan, I'm guessing you're debating about having a 2% grade that with go beyond the bridge but might prefer to have it level?
I like what you are planning. Do you know what you are going to use to simulate water?
Also you're planning on having a helix to the lower level, do you have a plan for the helix? (or how many loops to the lower level?)

[KenS]

I really need to finish re-doing the track plan in RM4 with real elevation data.

Yes, if I put the viaduct station in the "urban station" scene so track level is roughly 2" above the lowest level of pink foam in the pictures (which puts the "subway" tracks at "ground" level below the station), then the tracks exiting the viaduct station platform go straight onto the bridges in the foreground of these photos, and I can keep them level around the curve. Ground at the "riverside station" remains atop the second layer of foam, but it is a ground-level station, so track on that side is atop the second layer of foam (first layer is 0.75 inch, second layer is 2 inch).

If, instead, I keep the subway a subway, put "ground" atop the second layer of foam at the "urban station" also, then the viaduct platforms are about two inches above the foam, and I need to bring the inner (commuter) loop down to foam level by the time they reach the other station (the "riverside station"). Using the full curve (from viaduct "urban station" to "riverside station") will let me do that on a 2% grade, but it means the bridge also tilts down at 2%, which is really unprototypical (I don't know of any truss bridge built on a grade; I'm not even sure it would be structurally sound to do that for real).

Using less than the full curve gets me closer to 4%, and wouldn't look very good I'm thinking. I'm trying to keep in maximum visible grade to 2%, and the maximum hidden grade under 3%.

Right now, the plan for water is what you see (green paint on wood).  I may do something more sophisticated (woodland scenics or some other resin product) eventually, but it's not a priority.  And having the table seam in the middle of the river would make that tricky to do, and leave me with a seam in the water anyway. 

There are rough sketches of my helix plan on my website (under the Construction page), but I haven't really thought much beyond the "how big, how steep" aspect. I drew it as a simple circle of 381/414 radius viaduct track (mostly because it was easy to draw that in RM and it fit the space; I wasn't planning to actually use viaduct). It takes 5.5 loops to get down to the planned level of the storage tracks. That gives me a grade on the outer (ascending) track of about 2.5%, so it would be workable.

My original thought was to use the "overlapping 1/8 inch glued plywood" method used by many helixes (helices? heli?), and then I stumbled across plans for a Unitrack Viaduct Helix, which looked pretty cool. The only problem (aside from cost) is the lack of old-style flat curved viaduct track, now that it's all been replaced with superelevated curves. Trains on a helix really shouldn't be leaning into the curve, it just makes toppling much more likely.

The clever thing about the viaduct helix design is that it uses the catenary mounting holes in the viaduct for the support poles, and simple rectangular plastic tubes slid over those to hold the viaduct levels apart. Very simple to build and modify. The negative aspect is that I'm likely to have clearance problems with those supports if I use the superelevated viaduct.

So I'm leaving the decision on that aside to percolate a bit.  I wasn't really planning to build the helix anytime soon anyway. I want to get the three track loops all operational, and some basic scenery down first at a minimum. I can always pack/unpack trains to the storage boxes they came in for a while (and should as long as I'm fooling around with dust-making scenery work anyway).

[Bernard]

That is an interesting and simple design for a helix. I originally thought was that you were going to make it out of  plywood cutting the parts yourself. Great find!

[qwertyaardvark]

Very much liking the mockup and river scene~ 

Right now, the plan for water is what you see (green paint on wood).  I may do something more sophisticated (woodland scenics or some other resin product) eventually, but it's not a priority.  And having the table seam in the middle of the river would make that tricky to do, and leave me with a seam in the water anyway. 

If i may suggest, instead of a resin-based water, how would you feel about a contoured sheet of plastic lightly stained blue, or whatever color you wish, in the shape of your river? It would (1) hide the seam in your rivier and (2) be easily removable, either by not attaching to either module or attaching it to only one of the modules

[KenS]

That is an interesting and simple design for a helix. I originally thought was that you were going to make it out of  plywood cutting the parts yourself. Great find!

I was actually (and might still), but that Unitrack helix looks much simpler.

[KenS]

If i may suggest, instead of a resin-based water, how would you feel about a contoured sheet of plastic lightly stained blue, or whatever color you wish, in the shape of your river? It would (1) hide the seam in your rivier and (2) be easily removable, either by not attaching to either module or attaching it to only one of the modules

I'd heard of people using plastic fluorescent light diffusers for water, and had been thinking I might look into that.  I hadn't realized there was a commercial product for it though.  The problem is that it appears to be about $17 for a single sheet that's less than one foot by two feet, which is prohibitively expensive for the amount of water I have. Definitely something to think about. I need to order some stuff from HS one of these days, maybe I'll get a sheet of that and see what it looks like. 

Thanks for the pointer.

[qwertyaardvark]

I will admittedly say that that size is rather expensive, and for most part the prices seem to hover around that $/area range. Also to think about is that what i posted was the "official hobby" version with typical higher prices, so you might really be better off with a trip to your local DIY store (Home Depot, Lowes, etc) and comparing costs with the price of considerably larger sheets of light diffusers, made usually of plastic, and in some rare cases glass. One problem generally with diffusers is their texture: they often dont model water well (looking more like real water placed on top of some loud speakers), let alone N scale sized calm waves on a river passing through Tokyo.

[KenS]

One problem generally with diffusers is their texture: they often dont model water well (looking more like real water placed on top of some loud speakers), let alone N scale sized calm waves on a river passing through Tokyo.

Yeah, that was what I thought the first time I looked at a plastic diffuser.  Even for the smallest, the "wave" size seems more appropriate to a large bay or a relatively calm ocean than to a river. On the other hand, too smooth doesn't look good either.

Once of the things I like about the current paint is that applying it with a "trim" roller (fine foam roller about 7 inches long) left a rippled texture that looks "right" for a river. The only problem is that it's a very uniform color, and rivers just don't look like that. Even if they're deep enough not to show depth variations (and most have lighter edges where they are shallower), they'll reflect things to the side. Perhaps just a sense of lighter coloring from reflected buildings, but where you look will change the coloring.

On the wide river, I may try doing something to add some variation between the edges and side, even if I don't go with some other surface treatment. Perhaps airbrushing a lighter green or gray-green near the embankments, although my airbrush skills amount to "big swaths of uniform color". But I think I'll leave that to later, and try to get the dirt/pavement/greenery on the land to look like something more real than pink foam first.

And before any of that, I really need to finish sculpting the land, and putting up track.

[KenS]

The best laid plans...are usually wrong.

I mocked up the "urban station", well one end of it, to check clearances and appearance. In the process I discovered what should have been obvious from the track plan: there was only about 6 inches between the station and the scenic backdrop (maybe nine, if I'm generous).  And there was a large gap between the shinkansen tracks and the commuter tracks inside the station (where the "Elevated Shinkansen Station (16 car)" title is on the track plan).  This was because I had put a 124 mm straight in the middle of the curve leading into the station from the bridge to avoid an S-curve.

After a bit of thought, I removed the 124 mm straight, and built the station as three sections deep rather than four (see photos 1 and 2). This means I now have a 315/348mm s-curve between the bridge and the station, but the commuter cars seem to navigate it okay (see photo 4).

That's not my only problem. As I suspected, I have a clearance problem where the commuter line bridge footing goes above the subway track.  Which means I'm probably going to have to raise the track (and thus the whole viaduct station) at least a quarter-inch.

Even with that problem, I'm still liking the idea of having the subway tracks at ground level under the station.  Photo 4 shows a side view. This is on a temporary set of viaduct piers that are sitting directly on the subway track; I'll need to built some kind of support structure that spans the subway tracks, but that's a minor problem.

So, one question: does anyone think the 315/348 s-curve is going to be a problem?  It has straights at both ends, to avoid problems with the bridge or the switches in the station, so I think it's okay, but I'm a little bit nervous about it.  I was really hoping to avoid any s-curves on the two above-ground loops.

[NozomiFan]

Bill937ca, as noted the structure is an oval with two 24" tables back to back, so there's 48" for the end curves.  I started out buying through Newhall Station (and, aside from their costs, I was very satisfied with the service), but once I found out about the alternatives thanks to this forum, I've been using Hobby Search.  However, I've found a few trains (and most buildings) domestically, through MB Klein and dealer's tables at model railroad shows. Lately I've been putting all my money into layout construction, and haven't bought a new train since October (I broke down and bought the Chuo E233 then); I think I'm going into withdrawal. 

Heh... Heh... I had my experience with Newhall Station....
Good luck on the Layout! 

[Bernard]

Ken - You've really done a lot and it's looks good. I don't have an S curve on my layout but to be sure test it out if you can by running trains. Since you have a straight piece before the switch I don't think you'll have problems with derailments. If the long Shinkansens can make it you should be fine. You say there is a clearance issue is that for the track supports under the platform? If so you can cut the bar in the track pier that is blocking the train. (that is if I reading you problem correctly)
In all it looks great!

[disturbman]

Ken... I'm missing something here.

You metro line should (be lets say) the level 0 of your layout. Then your elevated station should be at +2 or here, you showed us a metro line at 0 and a elevated station at +1. Where is the ground? Because withtout it you can't have above and underground rail lines. Or I missed something on your layout plan.

[KenS]

Ken... I'm missing something here.

You metro line should (be lets say) the level 0 of your layout. Then your elevated station should be at +2 or here, you showed us a metro line at 0 and a elevated station at +1. Where is the ground? Because withtout it you can't have above and underground rail lines. Or I missed something on your layout plan.

The track plan up above shows the original plan, which had the urban station "ground" 2 inches above the subway tracks (level +1 in your terminology), and the viaduct tracks two inches above that (level +2). I haven't updated the track plan, but what I'm presently thinking about (mentioned briefly in an earlier posting) is changing this scene to bring the subway "above ground" by lowering the ground to the level of the subway tracks (level 0). So the viaduct tracks would be roughly 2 inches above the subway tracks, rather than four inches.

With that change the "subway" would be surface level here, and would enter a tunnel at the end near the bridges (likely just before it begins to curve to avoid the river). From there it would be an underground line past the river and through the "riverside station" scene on the other side of the divider.

These photos show the mock up of how that would look. I really need to finish updating the track plan.

[CaptOblivious]

The S-Curve will pose no problem. It's very broad, by Japanese standards, and the expanding close-couplers on the commuter trains will actually help, rather than hinder progress across them. So you're fine.

[disturbman]

The track plan up above shows the original plan, which had the urban station "ground" 2 inches above the subway tracks (level +1 in your terminology), and the viaduct tracks two inches above that (level +2). I haven't updated the track plan, but what I'm presently thinking about (mentioned briefly in an earlier posting) is changing this scene to bring the subway "above ground" by lowering the ground to the level of the subway tracks (level 0). So the viaduct tracks would be roughly 2 inches above the subway tracks, rather than four inches.

With that change the "subway" would be surface level here, and would enter a tunnel at the end near the bridges (likely just before it begins to curve to avoid the river). From there it would be an underground line past the river and through the "riverside station" scene on the other side of the divider.

So indeed, I missed something here. I read the all thread but with all the projects going along I just kept your plan in mind. The idea is interesting and I think you can find other piers or one that could help you accomodate the metro line under the elevated ones. I saw such things a lot in japanese model railroad videos a lot. Maybe with pillar instead of those four legs things. You could also manufacture very easily your own supports pillars or structures. Think about metro stations a bit heavy with concrete but that could the trick for you.

I'll be curious to see what you'll come up with. :)

[Mudkip Orange]

does anyone think the 315/348 s-curve is going to be a problem?

JDM trains are a LOT more tolerant of S-curves than US market stuff, even within manufacturers. All things considered, a Kato EF58 with Rapidos is going to give you less trouble then a Kato GG-1 with Micro-Trains.

Especially considering that the S-curve is on a *dedicated commuter* line (i.e. no Shinkansen), I don't forsee any problems whatsoever.

[David]

JDM trains are a LOT more tolerant of S-curves than US market stuff, even within manufacturers. All things considered, a Kato EF58 with Rapidos is going to give you less trouble then a Kato GG-1 with Micro-Trains.

Agreed. Body mounted micro-trains derail on anything resembling an s-curve. Anything. Given all the difficulties with automatic coupling not being automatic oncurves (and often even on straights because the couplers are out of alignment from a previous uncoupling), to say little of "automatic" uncoupling and the assorted problems with micro-trains compatible couplers being wildly out of gauge, it's been a breath of fresh air going back to rapidos couplers. They're square, but they actually work out of the box.

[KenS]

Just FYI, my series 500 Shinkansen can negotiate the s-curve okay at low speeds, but at the 3/4 setting on the Kato power pack (what I'd use for a high-speed train running through the station) it derails the car behind the motor car. I'm guessing the heavy motor car follows the track in a sudden shift and yanks the following car sideways.

As it is a line that I don't plan to run Shinkansen on except for moves to/from the staging tracks, that probably means I'm okay, since the E233 doesn't have a similar problem at speed. Hopefully I won't find a Limited Express train that has this problem. At worst I'll have a real-life "speed restriction" on the entrance to the station. 

The Shinkansen does handle the "almost an s-curve" (two R481 curves with a 45mm straight between them) of the Shinkansen line station okay, even at speed.

This gives me 11 inches between the station and the backdrop, which is just enough for two rows of buildings facing each other across a moderate-size street, or a street adjacent to the back of the station and a two-deep set of buildings at the rear. So I think I'm going to go with that track arrangement.

I'm still debating if I want a 16-car platform for the Shinkansen, as shown on the track plan. I'd really like one, but it may be too long to look good given the space I have.  I actually don't have enough viaduct station expansion sets to mock it up without taking apart some of the commuter platform, as my original plans had called for an 8-car platform.  It seems wrong to have the Shinkansen platform shorter than the 11-car commuter platform. Maybe I'll compromise at 12. 

[Mudkip Orange]

which is just enough for two rows of buildings facing each other across a moderate-size street

No matter what, if you're Anglo, you're going to make your street too wide on the first try. Japanese streets are NARROW. For instance, check out this commercial street right near Seibu-Shinjuku. It's only 15 feet wide. That's 30mm, or 1.2 inches - less than a single Kato track spacing.

Or check this residential area just up the JR Freight line from Hama-Kawasaki station. Many of the houses front little alleys that aren't even wide enough to drive a Kei car through, let alone an SUV. You can compare the scale to the freeway next door.

[bill937ca]

which is just enough for two rows of buildings facing each other across a moderate-size street

No matter what, if you're Anglo, you're going to make your street too wide on the first try. Japanese streets are NARROW. For instance, check out this commercial street right near Seibu-Shinjuku. It's only 15 feet wide. That's 30mm, or 1.2 inches - less than a single Kato track spacing.

Japanese streets are generally the width of a wide North American sidewalk. Many have no lane markings at all and most streets do not have sidewalks.

http://mikkagashi.cocolog-nifty.com/blog/2009/06/post-0a73.html

http://mikkagashi.cocolog-nifty.com/blog/2009/11/post.html

http://mikkagashi.cocolog-nifty.com/blog/2009/05/post-bb1f.html

http://mikkagashi.cocolog-nifty.com/blog/2009/04/post-5d8f.html

http://mikkagashi.cocolog-nifty.com/blog/2009/03/post-abc8.html

[CaptOblivious]

Ken, about the S-curve:

I wonder, even if it's not a problem, if you could rework the station throat to get rid of the s-curve, by eliminating the symmetry. Have a look at the image. I know these aren't the same track parts you're using, but the idea is the same.

[disturbman]

It looks good Don, a tad more flowing and natural.

[KenS]

I wonder, even if it's not a problem, if you could rework the station throat to get rid of the s-curve, by eliminating the symmetry. 

Nice thought, and I like the flowing curves.  But I can't come up with a version that (a), has proper spacing for the platforms, (b), will match up with the curves at both ends (one of which is constrained by the presence of the double-slip between the inner and outer loops), and (c), will fit on a three-wide Viaduct platform (307mm).  I'll probably try a few more variations before I admit defeat, but I'm thinking I'll need to learn to love the s-curve.

And I really want to use the three-wide platform rather than a four-wide.  Not only does it give me space for a larger street, but with only about 8 inches of space to the backdrop the wider platform makes it hard to even see street level. 

And while many Tokyo streets are microscopic (and I plan to model those in the residential area), I'm presently aiming to make this section look like some of the larger multilane commercial avenues you find in front of stations (e.g., Keiyo road in front of Kinshicho station), although likely two lanes each way rather than three or four.

Once I settle the track plan, I'll place a few buildings and take a photo or two to show what I'm trying to do.

[KenS]

Well, I did get the "flowing curves" to work, although I had to cheat a little.

What I did (see diagram) is put R481 curves (mostly) with #6 switches, with one R718.  This caused the separation on the front commuter tracks to be 60mm instead of 66mm. The platform pushes the two apart to the correct separation, and the Unitrack has enough flex that this works fairly well. Similarly, the back commuter tracks are 72mm apart, but again there's sufficient flex that the 6mm error isn't really noticeable. This wouldn't fit on the overhead platform without a fourth segment, except that pulling the rear track forward 6mm to meet the platform is just enough to make it work.

The rear track does have a tendency to want to slide back against the wall, so I may have to glue it in place or something for reliability purposes. But it looks like I have a solution that looks better than the s-curve, and my Series 500 navigates the curvy rear track at quite a high speed without problems.

The end of the platforms away from the bridges is a bit of a mess. I really don't like extending the tracks about two feet before bringing them back together.  So I may try to come up with a slightly different track arrangement at that end. And, as can be seen in the diagram, there are a couple of gaps I need to deal with, but that's a minor problem.

This will force me to go with the 12-car Shinkansen platform rather than a full 16-car platform; there's just no room for both the long platform and gentle curves for the rear tracks. But I was leaning in that direction anyway, and the only Shinkansen I own today is an 8-car set (so now I only need one expansion, not two; money saved!  )

Thanks for the suggestion, Capt. O., I like this a lot more than the s-curve.

[CaptOblivious]

Great, I'm glad you found a way to make that work out! And it looks good too. I bet you could still fit a full-length shinkansen platform in there too…

[KenS]

I think I've solved my "how to get to the subway/helix" track layout problem.  As you may know, I'd tried (on paper) several arrangements to get down from the commuter line at what I call the "riverside station", and none of them worked without looking like I'd need to get creative with flex track, something I'd really prefer to avoid. And, I couldn't get both of the descending tracks to connect to both subway and helix, so I was going to have to run trains the "wrong way" on the visible station tracks on some movements, which was really getting on my nerves.

So the other day I started playing with unitrack full-size on the layout table, trying to see what would work in practice, as opposed to what Rail Modeler thought would work.  There's enough give in how Kato track connects that a long run can diverge pretty far from a straight line.

So here's the solution. As you can see, RM doesn't think things quite line up, but in practice they do.  The two blue lines descend, while the yellow (commuter) line climbs as you move left to right (all grades are 2%).  At the bridge, a concrete tunnel entrance for the blue track, and a painted viewblock below the bridge will ultimately hide the mess of track behind the bridge from sight. The crossing (a 15-degree "Left") and some 64mm straights connect the up (front blue) line to the outer (climbing) arc of the helix, while the switch in front of the bridge connects it to the subway.  Behind the bridge, the down (rear blue) line connects to the inner (descending) arc of the helix and to the subway.  All switches are #6, all curves other than the helix are R718.

On the subway itself (not shown since it's under the green "Shinkansen/Rapid" loop) both blue lines connect to the inner arc of the subway, since there's no way to cross to the outer in that limited space.  The subway has a crossover at the right side of the curve (below the green/yellow crossover), so wrong-way running will be limited, and will occur on the hidden portion of the line.

If you look closely at the track plan you'll also notice I'm doing something a bit odd with the platform tracks.  The rear track uses the usual 64mm spacer at the switch, to bring it out to fit a 41mm standard platform size.  But the front track actually uses a 124mm track, making a larger gap between the front blue track and the center tracks.  Since these are on different levels, I'll need to scratch-build or kitbash the platform anyway, and this gives me more room to work with where the platform is visible (I'll cheat on hidden portion of the rear platform to make the top part look good).  It also was needed to make the tracks under the bridge come together.  One last note, the rear blue track actually starts curving forward a bit before the curved track and will be shorter than a 10-car train as far as a platform goes, but that's hidden behind the raised center track, so I don't care.

Track plan: the bold +/- numbers are elevation above "river" level in inches (+0.75 is subway level, +2.75 is the nominal "ground" level in this scene). The fainter +/- numbers along some track are Rail Modeler's elevation in mm, which isn't quite correct (I'm still struggling with grades in RM; one of it's problems is that is can't/won't apply a grade to a switch, crossover, or Kato double-track; I can see the logic on the first two, although I will have a switch or two on a grade, but the double-track limitation is just annoying).

Photo 1: the view of the front of the bridge, with the low-level bridge across the visible portion of the river into the subway. You can see how both lines tie into the inner subway loop in this photo.

Photo 2: what's hiding behind the bridge. Note that they pink foam below this track is temporary for testing purposes.  The eventual structure there will be plywood.

Photo 3: an overview of the Riverside Station, with a 10-car Sobu train for scale.

Photo 4: a side view of same, showing the Shinkansen line above the subway line (eventually a real "ground" will be there, and a fascia with cutouts for viewing the station platform will be in front of the lower track).

[SubwayHypes]

yo awesome track plan and i cant wait to see pictures once you finally get this down.

i also want a nice double decker layout with subway metro and commuters too,  i used the WW risers as well.  check out my layout and see what i did.   im rebuilding it from the ground up and i still want to implement my ideas.  i want a nice underground train storage yard and hopefully a little lighted underground station would be nice too.

[qwertyaardvark]

Very interesting layout indeed... I think it only just occurred to me exactly how ur layout track plan was gonna pan out! Definitely interested how the station platforms on this side will turn out. A station on top of another station on a grade plus a waterside view~ And all those risers! Also can't wait to see shinkansen running along that long straight stretch there~ 

As far as RM goes....

In preparing for battle I have always found that plans are useless, but planning is indispensable.

[Mudkip Orange]

Does the Kato doubletrack truss have some kind of snap-in piece that lets you build a double length (496mm) bridge, or did you kitbash it?

[KenS]

i also want a nice double decker layout with subway metro and commuters too,  i used the WW risers as well.  check out my layout and see what i did.   

Actually, your thread last summer (back when I was lurking on this board) was one of the things that inspired me to design my layout to incorporate a subway. And although I'd already been familiar with Ochanomizu Station, the photo posted to your thread helped convince me that I wanted to pattern one of my scenes after it. I'm looking forward to seeing what you do with the new version.

One of my big regrets with this layout was not being able to include a topside yard for the passenger trains. The lower-level storage I'm going with seems like it will be useful, and it should help keep the dust off the trains, but it won't have the same visual appeal.

Does the Kato doubletrack truss have some kind of snap-in piece that lets you build a double length (496mm) bridge, or did you kitbash it?

Each bridge comes with a snap-in top section (about 2 inches square) that makes it look like a continuous bridge. It's not limited to just double-length, you could keep extending it.  At one time I thought about making a bridge three or four segments long, but it took up too much space I wanted to use for other features.

@qwertyaardvark: I'm looking forward to seeing my series 500 blasting down that track too.  Unfortunately it's the only Shinkansen I own, so I guess I need to acquire a second bullet train one of these days...so many trains, so little money. 

[Mudkip Orange]

I guess I need to acquire a second bullet train one of these days...

get the kyushu train. put pikachu decals on it.

[KenS]

My first subway train, a Greenmax "completed model" Tokyo Metro Series 10000, arrived today from Hobby Search via EMS; just eight days after I ordered it (I also bought the 6-car expansion set, to give me a 10-car train).  As I type, it's running in on the "subway" track, something I do for each new train to break it in and make sure there aren't any early failures.

It's a very nice model, despite the truck-mounted Rapido couplers. The detail seems on a par with my Kato's, except for a lack of interior detail (the Kato's generally have some molded seats).  The motor is a bit noisy, but that could be because it hasn't worn in the gears yet.

Here's a photo of it posed below a Sobu 205 train.

[disturbman]

Nice (I'm getting tired of saying that  ). Maybe you could shoot a video of those trains running on a fast bare layout. 

[KenS]

At the moment the track is inoperable, while I get serious about carving/gluing the upper layer of foam down. But once I have something I think is worth looking at (more complete than bare pink foam, but nowhere near final) a video or two is a distinct possibility.

How hard can it be?   

[KenS]

Well, less was accomplished this weekend than I'd hoped (what else is new?), but I feel like I'm making progress.  The goal for the weekend was to finish the topography of the "River Crossing" scene where a small residential/commercial neighborhood I'm calling "the village" nestles inside the four-track curve.  That meant cutting and carving foam, gluing it down, applying plaster-cloth and cork roadbed, and throwing down a base coat of paint to hide the pink.

I never got to the plaster-cloth, cork, or paint, although I'm nearly ready to start plastering (maybe tomorrow night, if I don't have other distractions).

The photos show what I did manage to do:

photo 1: This shows the foam cut to allow the village to sit at a lower level inside the curve. After I glued down the foam, I realized I wanted the road from the bridge (the big commercial avenue from the Urban Station scene) to continue through this area and under the tracks.  Which meant a bridge for the tracks above the road. Which meant a gap in the foam.  The foam I'd glued down the night before.

Fortunately I'd used Woodland Scenics Foam Tack glue, and after cutting on either side of the gap with a large handsaw, a putty knife levered up the now-superfluous chunk of foam without a hitch. A new 0.75-inch slab, to match the "village" height, was glued in its place. The photo shows my first placement test using 186mm bridges.  Those were too wide, so I switched to 124mm single track viaduct sections as a suitable stand-in for a ballasted-deck bridge.

photo 2: a different perspective on the placement test. This one makes the trenches cut into the top layer of foam under the tracks more obvious.  This is where the electrical feeders will go, leading to the edge of the table then down through the holes drilled when the table was built.

photo 3: Once I had a gap, I needed to have bridge abutments.  These are just sections of 1x4 pine left over from the construction of the support legs, cut to an appropriate length.  This photo shows them without paint, with the 124mm viaduct sections. They'll be glued in place prior to the plaster-cloth being applied, so that the terrain can wrap around the front edge (I also made a set for the four rail bridges across the river).

photo 4: and this is after I'd painted them with two coats of latex primer (the same stuff I used to paint the tables).  This is just the base coat to seal the wood before I glue them in place; they'll eventually get a coat of some whiter "concrete" color. The gap between the abutments is just slightly wider than the planned roadway, so they're a good match for it.

photo 5: while I was at it, I decided the corners were too flat, and added a couple of chunks of 2-inch foam to provide a hillside behind the track.  I also added a set of 2% grades to bring the tracks up to the height I'll need for clearance of the Urban Station over the Subway tracks.

Another thing you'll see in this photo is some black window-screen, cut to fit the hillsides.  This is going to be applied to the plaster (or maybe something atop the plaster) to give the texture of a stone retaining wall. I'm following (roughly) the idea described by Capt O. in Pouring Concrete on Shōgatsu. Rather than paint, I'm planning to color the wall with a brown ink wash.  We'll see how that turns out in a week or so.

That's it. Not as much as I'd hoped to accomplish, but it's getting close.  This scene is going to be my practice round for the construction and scenery techniques I'll use in the two other scenes, as this one is the least likely to be closely viewed, so any errors will be easier to live with. And it's also relatively simple in terms of track and wiring, so I can always tear it out and rebuild it later, if I really don't like something.

And with it finished, it will be both an anchor for the track placement on the other two scenes, and an incentive to get them to a similar level of completion so I can actually run trains.  Realistically, the latter is still a month or two away at the rate I'm going.

[Mudkip Orange]

Awesome.