Jump to content

JR-Chiisai: Enoden T-Trak


Martijn Meerts

Recommended Posts

Martijn Meerts

So, because time is limited and because I currently don't know what to do with the large layout, I decided to try and pick up the T-Trak idea again.

The idea is to model the Enoden line, starting with a couple of the typical scenes and slowly expanding it to eventually include all stations. Scenery wise, I will be modelling the 4 seasons, to add lots of variation.

The modules will be made in such a way that I can use them in a pre-defined order for the Enoden line, but also that some modules could be used to mix with other T-Trak modules. Considering I really want DCC, there will also need to be a system to switch between DC and DCC.

I already had some module bases lasercut and glued together a long time ago. I kinda got held up trying to find a nice colour stain to paint the bases with, so apart from adding the track, nothing's actually happened with the modules.

The first thing to decide now, is which scene(s) to start modelling. I'll definitely want to model 1 of the single track stations, although I might want to get a double length module for that so the entire station fits on 1 module. I'm also considering doing scenery only modules to add some depth to the whole thing.

So, apart from staining the bases and giving the track itself a coat of paint, the only thing I can do now is research into what to start modelling. Any suggestions and good pictures/videos are more than welcome :)

Link to comment

First, i think the Enoden is a great theme and it looks like you will be doing mostly single track ttrak modules. This makes it easy as you will have to deal with a single 'loop'' only. Since you are aiming to allow DC and DCC operation, you may have to focus on two things. One is handling power everywhere in DCC mode, the second is handling DC block control. You might also want to use block occupancy detection for DCC for automation purposes. (but for station stops, i would suggest threadle type sensors used by Tomix and some of the analog Kato automatic signals)

Since the line is single track, you must aim for bidirectional traffic. In DC, this means isolated blocks with something like an A, Z or Ü circuit. This is great as this requirement makes it easy to install block detection in DCC. Generally you have to isolate both rails of all station tracks. You can leave the turnouts unmodded as the isolators will go onto their ends anyway, so having tiny powerless sections near the frogs on non selected directions is not a problem. In digital, you just have to connect all mainline and station blocks into the DCC bus by plugging them into a distribution strip.

In analog, each mainline block can have a dedicated controller or you can have cab buses. you have to connect a control box in place of the distribution strip that selects between the up or down controller or no power at all (generally a double pole, double throw, center off switch or a 3 position, 2 circuit rotary switch). This allows each block to be powered from one controller or another. If you want run the whole layout with a single controller, then you will only use one input and connect all mainlines together.

I'll draw some diagrams...

 

Link to comment

The first diagram is how to wire up the modules to be used in any configuration, with separate, isolated blocks for each station track and dedicated feeders for each block on each module. (this is overkill, but it's easy and cheap to do it) The 2nd diagram is how to wire up for DCC, just plug everyting into the main DCC bus. The 3rd is the same with block detection added and with double isolation, you can use any type of detector. The 4th is the DC wiring for one throttle for each mainline block. This is the easiest and the suggested analog block control setup by Kato. The 5th is the same but using only a single controller for the whole 'loop' and not using a traction power bus. The only differences here that the throttle selector has only one input connected and there is a bypass wire going around the station. If you want it to be simple to set up, you can add a bypass switch to the control box, which just shorts together the up and down inputs. This is shown on the last diagram.

ttrak_block_control-20170831-1.thumb.png.ec8f88b6fe3ed506be430f8290f05101.png

Generally speaking, you can create this box with the two controller selector switches (up-stop-down), station bypass switch (on-off) for DC operation and replace it with a simple DCC power distributor strip for DCC operation. Turnout control is fully independent, but you may want combine the turnout control switches into the same box for analog operation (needs a 12V DC input) and the DCC stationary decoders next to the distribution strip and any optional block detectors for DCC operation. This would allow a full plug and play operation in both modes, just plug in the feeders and the turnouts into the DC or the DCC box.

ps: What i wrote above works for more station tracks, double track mainlines and most complex combinations of tracks. Fremo analog operation generally adds one or two local shunting throttles for the larger stations, so you can select from up to 4 throttles (and off) for each station block, but imho this is not really needed for an Enoden layout.

Edited by kvp
Link to comment
Martijn Meerts

Enoden isn't exactly a very unique theme, but it just seems to fit well with T-Trak, and it offers lots of great and totally different scenes. There's also plenty reference material of it available, even if you've never ridden the line (I do hope I can do so during my next Japan trip though)

Power everywhere isn't a problem. I currently have 8 module bases (4 corners and 4 straights), and all of them have feeder wires using the standard Kato plug, and I plan to do the same with all modules. Things like signals I haven't even thought about really. Considering I want to go computer control for the DCC bit at least, I can have the software control the signals, and I could make custom ones based on the actual Enoden signals. Of course, in DC mode they won't be operational in that case, but the question is whether that's required anyway. If I do end up using my modules in a larger setup, there's a fair change the larger setup will just be manually running trains. I could just decide to not use any station modules or modules with turnouts on it in that case. There's also the fact that I don't exactly know a lot of people doing T-Trak in general in the Netherlands, let alone Japanese T-Trak.

For the moment I think I just need to get started with a couple of modules before I start overthinking everything again. Of course, any suggestions for controlling things and schematics are very welcome either way :)

Link to comment
5 minutes ago, Martijn Meerts said:

For the moment I think I just need to get started with a couple of modules before I start overthinking everything again. Of course, any suggestions for controlling things and schematics are very welcome either way

For a start, i suggest you to put double isolators behind every turnout at the station side, so your station tracks are all isolated. Set your turnouts to power everywhere if they support it, or jumper wire around them if they don't. This would allow any control scheme to be added later. (isolators are needed both for block detection and analog block control, so adding them at the start is imho a good idea)

For signals, you can wire them up to run on their own DCC bus, so you can keep the turnouts and signals DCC powered/controlled, even with analog trains.

For automation, i suggest you to consider threadle type wheel sensors for station stops as block detection is not pinpoint enough to stop right at the platform marks. Threadle sensors are available from Tomix (TCS sensors), many european track manufacturers also make them and they can be hacked out of Kato analog automatic signals and the old non optical Kato crossing gate sensors. This is important as the reliability of the automation will depend on these sensors. Adding them later to Unitrack actually requires lifting and relaying the track, so imho not a good idea. If you skip automation and choose manual only, then they are not needed at all.

Link to comment
Martijn Meerts

Occupancy detection can be accurate to the millimeter if you set it up right actually. Used it a lot to stop various trains at exactly the same spot consistently. One issue is, does an Enoden always stop at the same position at a platform, or does it depend whether it's a single or double set. I also know there's at least 1 station that's too short for a double setup and the trains actually drive past the platform. So there's definitely some research required for those situations. Also a good reason to start with a simple, non-station module :D

 

Link to comment
Martijn Meerts

Interesting device.. Of course, of all the digital equipment I have, none supports Loconet, apart from some developer kit of a DCC system that one of the forum admins worked on :)

Anyway, I just got finished with the first round of staining of the 8 module bases I have. Probably need at least a second round, it's still fairly light right now. I'll post pictures one the first layer has dried, and there's some better natural light.

Link to comment

I read up on that device. It's a microcontroller that acts like a programmable loconet throttle. It can take input from loconet sensors (block detectors) and also has 4 local digital inputs. It allows the operator to get rid of the computer but still have some automation. (considering a raspberry is both small and cheap, having a computer in a layout is not a big issue nowdays)

Link to comment
Martijn Meerts

I have no problem with adding a computer to a layout. Heck, the ECoS I have pretty much is a computer in itself, runs a custom version of Linux ;)

 

Anyway, did the 2nd layout of staining today. Not sure yet if I need a 3rd layer, but the time it was dry, there was no daylight left so it's hard to see what colour it is now. The image shows the 2 of the module bases with 1 layer of deep mahogany stain.

 

large.t-trak_stained.jpg

Link to comment
Martijn Meerts

This is what they look like after 2 rounds of staining. Not sure they need another round really...

IMG_0928.JPG

  • Like 2
Link to comment
Martijn Meerts

Stain I believe, although it's a bit less water-y than I'm used to from stains. I'm probably just going to start building 1 module now, and see what it looks like with some scenery. Just need to figure out which scene to model :)

 

Link to comment

Kato - I wrote the guy, and this is his response:

"I'm pleased it's still being read but no, sorry, I ended up retiring the boards. I've built a couple of other layouts since then - if you search for my name on Flickr you should be able to find them."

Link to comment
Martijn Meerts

Shame, looks like a fun project, and lots of thought and planning already went into it.

 

As for my version, I'm probably going to start with stitching together a lot of google maps screenshots so I have a map of the entire line including scenery. From there I can convert the track plan to something that'd work in T-Trak and decide how to divide up the parts into the 4 season. I'm guessing the bit along the beach should really be summer, and I want bits with lots of trees to be autumn, to really add lots of colours there.

 

Other than that, I also need to check the blocks. Judging from the video's I've seen, there's not a lot of traffic, so turning each module into a block isn't necessary at all. Would be more than enough to have blocks for each of the stations, and then 1 block in between the stations. Will have to experiment with it a bit I guess.

 

Link to comment

I've found the route map on the linked site and after removing one non existent track at the Enoshima yard, i colored in the blocks.

Enoden-ttrak-20170905-1.jpg.f3fd2f742a6b8e7c973a5819372ee136.jpg

Each color represents a line block. Each leaf green is a station, with one block per track. Gray tracks are yards, with only an on-off switch required.

This way, each line block can have one speed controller (throttle) and the station tracks could be switched over between the up or down mainline, depending on the arrival direction. Turnouts should be set to power everywhere and double rail isolators and feeders added to each station track. (this is also needed for block detection)

 

Stats:

-speed controllers: 6 (for analog operation)

-mainline turnouts: 11 (should be motorized)

-service turnouts: 9 (could be manual)

-analog blocks: 6 mainline stretches with stops (need throttles) and 12 station tracks (switchable to up or down, aka. left or right)

-detectable blocks: 15+22=37 (for digital operation, one for each mainline section and one for each stop/station track, without yards)

 

The 3 stops near Fujisawa could be one block in analog as there is no way to have more than one train there. In digital it has to be split to mainlines and stops for automatic control. The same is true for Kishigoe and Kamakura, since only one train could be there, but digital operation requires 3+2 blocks for the stops.

 

Generally speaking, you should only isolate station and stop tracks (both rails on both ends) and leave the rest as normal. (including station turnouts electrically in the same block as the mainline next to them) This would allow you to wire the modules up in any configuration later.

Edited by kvp
  • Like 1
Link to comment
Das Steinkopf
4 hours ago, katoftw said:

I wish the author of this blog updated it.  Would have been awesome to see the final product.

 

1 hour ago, velotrain said:

Kato - I wrote the guy, and this is his response:

"I'm pleased it's still being read but no, sorry, I ended up retiring the boards. I've built a couple of other layouts since then - if you search for my name on Flickr you should be able to find them."

 

 

 Will Vale is one of my Flickr contacts from many years ago as we were both avid LEGO builders, I was chatting with him recently about the Enoden project and if he was still working on it, sadly he said he gave it up as he came to the realization of how many buildings he would have to make for the layout and was somewhat terrified of the prospect of doing it. 

  • Like 1
Link to comment

Yeah - he doesn't seem highly energized  "Feel free to let me know if you find the new locations of the links and I'll see if I can update the content."

 

 

Link to comment
Martijn Meerts

Well, been quiet again from my end the past weeks. Work got busy once again, so haven't been able to get anything done after work really. Current project is almost done however, and the new one that I have to start afterwards doesn't have a crazy deadline, so should hopefully have time again soon to work on things.

 

I have been cleaning up my workspace a bit, and added some modules to my modular workbench system. Once everything's cleaned up I'll post a picture of that. Also gave my airbrush a good cleaning, and will likely start painting the first of my Gundam models sometime soon.I have 1 kit that's been partially prepared and ready for a base coat, just never got around to it :)

 

Other than that I've been slowly ordering/buying some needed materials for building scenery. Still a lot of stuff missing, but I'll get that when I need it. The idea is to sit down next weekend and trace the Enoden line and then determine which bits should be in which season, and then which one(s) I want to start modelling. I'm also planning another trip to Japan for February, and riding the Enoden is 1 of the things we're likely going to be doing.

 

Link to comment
Martijn Meerts

So, been a while since I posted anything related to any of the projects I've been working on. Other than having very little time, life hasn't exactly been kind the past years. There have been times where I've considering getting rid of most, if not all of my model trains. Even now, I'm still wondering if I'll ever continue the big layout for which I've built the hidden yard and helix.

 

I recently came across the little research I had done about the Enoden line, and thought maybe I could pick that up again during the moments I actually have some motivation. I've not done any track planning or really figured out what I want to do though, apart from deciding which parts of the line should be in which season. I've come up with the following:

 

Winter: Fujisawa, Ishigami, Yanagikoji

Spring: Kugenama, Shonankaigankoen, Enoshima, Koshigoe

Summer: Kamakurakoko-mae, Shichirighama, Inamuragasaki, Gokurakuji

Autumn: Hase, Yuigahama, Wadazuka, Kamakura

 

There should also be a transition from 1 season to the next, so it's not a different season all of a sudden going from 1 module to another. Of course, doing all the stations is quite a project, but at least I can do it in small sections, and work on scenery as well, instead of just track planning and laying like on the big layout.

 

No idea where to start, but optimally I want to get a set of double track stations. I do have several corner module bases and single module bases, but I'll likely need to get a couple of double length bases laser cut. On the other hand, once I know where I want to start, I might also start researching buildings first, and start drawing those to get them laser cut and/or 3d printed first...

 

  • Like 3
Link to comment
Martijn Meerts

Been looking a bit more at the Enoden the past days. I've come across a few interesting areas to use for the first module(s).

 

The first is a little restaurant just outside Shonankaigankoen, heading towards Enoshima. It's located trackside, and there's an entrance that requires crossing the Enoden track. The restaurant also has a small Japanese garden. All in all it's an interesting scene, but there are a few difficulties. The restaurant building itself is pretty strange, and I can't find a lot of reference material. Most pictures are from the gate that opens up to the track crossing, as well as the large quarter circular windows area looking out to the garden. Another issue is that there's a height difference, so can't really use one of the existing base modules for it.

 

Another area is between Wadazuka and Kamakura. There's an elementary school and a kindergarten on that stretch, both of which are interesting to model. The elementary school could be split in 2 single length modules due to the way it's situated, whereas the kindergarten would need a double length module to fit the whole building.

 

Obviously, a section of street running between Enoshima and Koshigoe is interesting as well, and could easily be made on a single module base.

 

All of these ideas do have some issues though. All my module bases are very simple, with the track having to be mounted on top of the base (standard T-Trak setup). However, for street running you obviously need to increase the height of the entire module apart from the track by about 7mm. This is easy enough using some wood or foam or anything else really, but you end up getting some rather ugly edges around the module. Same goes for the restaurant idea, where you have to add height to the back of the module. 

 

Another thing I noticed, is that for most of the Enoden, the track is actually not really raised compared to the surroundings, which means that using Unitrack means you either have to compensate, or be content with the track sticking up quite a bit more than the prototype. Obviously, using the T-Trak standard you can never really be prototypical, so the question is how much, if at all, I should care about the track height. On the other hand, if it already more or less annoys me now, I guess it'll continue annoying me later on as well.

 

I'm sort of leaning towards redesigning the module bases so I can cut the sides of the modules in the contours of the required landscape, and for things like street running. However, that would mean the bases I currently have would not be used, which would be a shame since I have 4 straights and 4 corners already.

 

Link to comment
55 minutes ago, Martijn Meerts said:

The first is a little restaurant just outside Shonankaigankoen, heading towards Enoshima. It's located trackside, and there's an entrance that requires crossing the Enoden track. The restaurant also has a small Japanese garden.

 

Would that be this one? Looks quite nice, indeed, but google maps isn't much of a help in terms of modelling either...

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...