Jump to content
 

deepfat

RMweb Gold
 View Profile  See their activity

  • Posts

    178

  • Joined

  • Last visited

 Content Type 

Blog Entries posted by deepfat

  1. deepfat

    My name is Bond ,Ballast Bond

    By deepfat,


    ballast
    All the Youtube videos I have seen on ballasting casually gloss over how to deal with points and specifically that hardest of all points,  the double slip.  My hypothesis is that it isn't the ballasting itself that is an issue, rather it is the pre-wetting of the ballast as that seems to disturb the ballast and that's the very thing we need to avoid around points.
     
    To set the scene -  my peco 75 track is laid on DCC Concepts foam bed, not cork, and I have spent thousands converting the loft buying trains and dcc equipment so I was happy to splash the cash and use Deluxe Products Ballast Bond.  It is pre mixed and dries matt, however even they suggest pre-wetting. So for my points I prewetted the track using an empty bottle of Ballast  Bond filled with a strong solution of washing up liquid and dripped that on to the tracks first, to reduce movement of the ballast.  If there was movement I carefully moved the offending bits before applying the glue and if I didn't like it I just let it dry again and redid the process.
    Here is my first go using  a DCC Concepts club tub ( Steam era, brown & grey) of ballast.

     
     and here is another view now it has dried

     
    I think the result is OK, and although time consuming I feel the time and cost are worth it to properly set off my expensive train collection.
     
    A couple of other observations;
    Ballast Bond doesn't create the hard shell you get with PVA which should make the layout quieter to run.  I can't say it does for sure but it is no louder than it was and I notice the track bed still flexes with ballast on top and when I had to pick out the odd mis placed stone I noticed that the bond was more rubbery than PVA It's theoretically possible to remove the ballast with hot water if needed but I have not tried this It only took overnight to dry off as my application of water was much more targeted and summer is a good time for this and this stuff must dry quicker
  2. deepfat

    Creating a backscene in the loft

    By deepfat,


    backsecene
    My layout is in the attic and was designed when I was new to model railways and BRM. With hindsight and reading more I probably got the track to close to the edge of the layout (120mm in some places). To make matters worse  the basic pyramid shape of my loft means that my backscene is only about 120mm high:

     
    Also  the transition between the backscene and the roof is very obvious, even with both being painted with the same pale grey emulsion.  This looks even worse/odd  in the corners:

     
    To fix that obvious line and extend the scenery up the sloping roof I decided to create a paper mâché fillet between the two to create a smoother transition from one to the other. I then covered that with lining paper, but it was very fiddly to do and despite my best efforts it did bubble and wrinkle in some places. However it did look a lot better, again repainted in grey to highlight the flaws in it:

     
    After a couple more weekends cursing at my lack of wall papering skill I wacked in some lightweight filler to smooth things even more. I then painted the background sky using Dulux Nordic Sky at the top fading into First Dawn at the bottom:

    So looking better still but not great, and like me it looks worse in real life!  Also the wallpaper came away a bit at the bottom in places so that meant more fun with pva and filler.
     
    Now for the clever bit -  add clouds to camouflage the unevenness/ To do this I used white acrylic artist paint dabbed and smeared with cut up bits of sponge from an idea I saw on YouTube.  This is my work so far this time in another corner and this is looking nice enough to impress Mrs Deepfat:

     
    There is still a bit of shadow but with the right lighting on that starts to disappear.
     To finish up a I put a batten painted in metallic gold along the top edge of the sky to frame it like a picture. And around my velux windows.

     
    Now I need to add something below the horizon to create an illusion of depth and set the context for each side of my layout with aim of making the layout appear to be bigger than it is:

     
    If you know Hastings you'll realise this  is not accurate but my intention is to create an homage to where I grew up in an earlier age so it has all elements the I want in it but not necessarily in the right  place!
    In my next post I'll show what the sea looks like and how I did that.

  3. deepfat

    Backscene West

    By deepfat,


    backsecene
    The easiest bit of my backscene is the sea to the west of my layout. It's inspired by the view from West St Leonards with Hastings seafront to the west and Beachy Head to the right.
     
    In my last post I showed how I set up my backscene and created the sky.  For the sea behind my layout I cut a 100mm x 3000mm strip of lining paper and airbrushed that with a slightly varying dark teal colour and stuck that so that the top was completely level to simulate the horizon.  
     
    For the NW corner of my layout I then added to more bits of lining paper one for the shingle beach and one for Beachy Head.  The beach gradually curves from upright to horizontal to create the illusion of depth and I sprinkled sharp sand and scatter scenery to simulate the shingle beach

     
     
    I used yet another bit of lining paper to simulate Bexhill in the distance  and then when that was done I could add grass, some scratch built beach huts and a tea room. There's more to do here like add figure benches etc. but you get the idea:
     

    The South West corner is inspired by Hastings Seafront fading to Fairlight in the distance.  Again I used collage and painted the scene on lining paper and created a curving up beach to meet the sea and this scene. This looks a bit bent because it's the corner of the layout but it is square  and mrs Deepfat likes it and that is good enough for me. I've added some more scratch bult beach hits and some foreshortened fishing boats like there are at Bulverhythe if you have ever been there
     

     I am no Canaletto but I this is good enough for me and now I have done the backscene, I am going to paint & ballast the track in front of it to test out how to do it and to get the look I want
     
  4. deepfat

    3 - Baseboard construction

    By deepfat,


    Baseboard construction
    Note this post has been updated following the loss of photos on RMWeb
     
    There are some beautifully engineered baseboard on RMWeb, and even if I was working in an empty room my woodworking skills are not that good.  My layout is built on existing loft ties about 1.5m off the floor as above those ties it's an empty void.  I have had additional 100x50mm struts on which the baseboard can rest but this is an old house and the tops of these timbers are not level and have also twisted a bit.  
     
    I want to avoid having a flat layout as Hastings is my inspiration and it's hilly. Add to this a desire for rakes of 6 coaches being pulled by a 4-4-0 Schools class and I need long shallow inclines. To do this I set sea level at the middle of the north section of my layout and made the south of the layout 50mm higher than that. This gave me a gentle incline at the two edges For the centre section I could climb from 50mm to 100mm over the centre section at a 1:50 incline to give height enough to cross the lower track and also have another track going from 50mm to 0 to complete the figure of 8..
     

     
    To correct the uneven levels of the existing roof supports I used a 2.4m level and a laser level to determine the heights I wanted and then fiddled with shims of ply to get everything at those heights.  In the photo below you can see I am following the plan above:

     
    To try and deaden sound resonating from the track all the ply is laid on floor underlay already stuck to the supporting bearers - you can see a this in the bottom left corner, however this hasn't really worked but. The ply was then held in place with  a rubberized floor adhesive and a few screws driven just far enough to level up the edges of the ply with each other. 
     
    Inclines
    The steeper inclines will be the 2% offerings from Woodland Scenics, and actually the shallowest they do. Here you can see I am starting to lay them:
     

     
    You can sort of see SCARM ( the software I designed the layout in) running on my tablet to check measurements.  The raised area in the foreground is sheet of 25mm expanded polystyrene which I cut to shape with a hot wire.  You can also see I am using Proses variable track gauges, Woodland Scenics Foam-Tack,  DCC Concepts 3mm foam track-bed and of course coffee!
     
    The only problem with the ready-made inclines is that you have this sudden change in gradient which needs to be avoided by adding and removing some sort of fillet to graduate the increase / decrease in the incline:

    You also want to avoid changes in incline near points and actually the SCARM software doesn't allow points to be on an incline so my design took this sort of thing into account.
     
    You are entering a PVA and Cork free zone
    Cork beds don't bend round corners and have to be cut & chamfered .
    PVA is cheap but:
    It forms a hard shell that reflects sound really well and has no give. It is too strong especially when used to ballast so any repair work or modifications result in destruction of the track. So I have started using Woodlands Scenics Foam-Tack not just to fix the inclines but to fix the underlay and fix to tack the track down on the underlay instead of pins.  This is great as mistakes can be undone by applying warm water and the main hold will come from the ballasting.  On that note  I'll be researching PVA free  ballasting probably starting with Deluxe Materials Ballast Bond (more on that if I ever get round to ballasting).
     
    A look ahead
     
    This gives an idea of where I am going finishing off the edges of the layout in ply to match the level of the terrain and use of expanded polystryene to create hills and embankments much like any other layout, to keep the weight down. You can also see I have left  a square hole so I can get into the tunnel:
     

     
    Next up Laying track
  5. deepfat

    2 - Layout design with SCARM

    By deepfat,


    Layout Design
    Note this post has been updated following the loss of photos on RMWeb
    Converting a loft for a model rail layout
    The only internal timbers in my loft are three ties at roughly chest height you can just see all three in this shot of the loft as it was insulated:

     It seemed logical to use this to support my layout.   Timber in 1946 was clearly of better quality as these are just 50x100mm in section and yet they stop the roof sagging by being connected to purlins around the edge of the roof.    I couldn't easily move the tie in the middle of the picture (the one with the red pressure tank next to it) and I would have to duck under it hence my need to automate my layout. However I could make a feature of it by making some sort of figure of eight maybe to change direction. As I mentioned in my last post I also got my builder to add in in some more 100 x 50 beams at right angles to the existing ties them to further support my intended layout. You can just see three of them right over the back coming out for the foil to meet the furthest tie.
     
    Given the physical constraints of the loft I would really need to think about a suitable track plan. I know a lot of people rely on a few sketches or marking stuff out on the boards once they are in place  but I am a software engineer and I designed our house in CAD so how hard could a layout be?  Well actually I found it a very challenging and interesting exercise in its own right.  I used a freemium SCARM track planning solution which I bought once it was released.  It takes a bit of getting used to but allows for accurate positioning of flexi-track, both to limit the tightest radius (750mm in my case) and to set inclines precisely as well.  I did buy the trains simulator too and while there are no UK trains on it, it allowed me to judge how the real thing would work:
     

     
    The layout
    Before I get to the layout design a word on my design goals -
    All my layout should be in view with no hidden fiddle yard as I knew I'd be operating it from  different positions with laptops  & phones over wifi and inviting a few friends to have a go as well. I wanted to loosely model Hastings where I grew up but back in the 1930's when steam engineering was at its height. There should be lots of options for routes mainline, branch-lines with and some freight working too My layout is pretty big  and those beams are in the way so automating the points would be essential.  I had no legacy equipment so I could standardise on the latest thinking. For example I was impressed with  the great point motors and systems from DCC Concepts. and their excellent advice on wiring for DCC I wanted freedom to stop and start trains wherever and have them moving at prototypical speeds so DCC would also be used for train control, and I wanted to explore software control from a touch screen like Hornby Railmaster possibly with Hornby Elite as well (which I have now replaced with a Rocco Z21 and iTrain as detailed n this post).   For the most part I'd be using 50mm track centres as there are no issues with overhangs on rolling stock at 750mm radii and this is what Peco 75 points are engineered for.  
    After many iterations my layout design looks like  this:

    The blue outline is the baseboard: The centre section is sitting on one of that exiting ties I can't move so you have to duck under it anyway.  The access hole top right is need to reach the edge of the layout in the top left (NW) corner. The loft hatch is not shown but comes up in the cut out are in the SW corner (so lower left).   
    SCARM this in 3D complete with the supporting timbers where brown indicates existing structural ties/purlins and the orange the new timbers added by my builder, so you can get a better sense of your design

    However the tools for creating buildings in SCARM are quite weak so I have just made a rough copy of Hastings station as was in the 1930's and designed the station to hold a  pacific loco with six coaches .  
    I have to admit the track plan does look a bit like a toy train layout, however as I said already, the radii are much larger to enable fitting detailing packs to locos, and to permit close coupling of tenders and coaches.  One other thing I did in SCARM that sort of worked was to try and plan around the supporting timbers

     
    In the 2d view layers can be hidden to aid clarity.  I have layers for the various design elements on the plan:

     
    Notes on the Design
    Essentially this is an two line oval track with a branch lines which cross over to change form one track to another.
    The three  branch line sections shown in grey are those likely to need auto reverse units as will the turntable. There is a turntable as I'll be running express steam trains and I have added to long sidings (in yellow) to feed that which will double as storage for  rakes of coaches or wagons.
    The two into one siding at the bottom of the layout will server some sort of industry - possibly a brewery!
     
    Inclines
    Hastings is hilly so I wanted at least a couple of inclines with bridges and so on.  The trick to setting up these is to create enough room to get one track over another at the  top centre (north side) given that I want to use 2% inclines. To achieve this, the southern end of the layout (the bottom edge of the plan above) at about 30mm above the lowest level (0mm) of the Northern centre section end.  In this view of the middle of the layout you the right far side is 30mm height so far double track climbs 60mm to give 90mm clearance over the track off camera on the left. The track in the foreground drops 30mm as it goes into a tunnel to connect with the bottom level of track.

     
    This gives a rough idea of how the track passes over itself and in this view I have put in a road bridge scenic break to show what this will look like:

     
    One snag with this use of inclines is the sidings (in yellow on the diagram above). They have to be nearly level so unattached rakes of wagons and coaches don't roll anywhere and the same applies to my station. SCARM helps with this but I also tested what the maximum incline I could have through the station without coaches moving and then built that into the design. 
     
    So now I could get busy with actually making the baseboard and that's for next time
  6. deepfat

    1 - Introduction

    By deepfat,


    Layout Design
    Note this post has been updated following the loss of photos on RMWeb
     
    Like many kids back in the day I had a train set. In my case my dad got my brother & me a Hymek Freightliner set from Hornby. It was setup on an 8' x 4' board and stowed under my bed when not in use.  We got bored of it pretty quickly and eventually it died a horrible death like some of the figures in the original Toy Story! Fifty odd years later and along bought of illness & enforced rest and I suddenly got the urge to look at the train again, like the ones my dad used to ride on. He went to St Olave's Grammar School and there's a loco named after it which by coincidence was designed by Richard Maunsell to cope with the hills and turns on the line to Hastings where I grew up.  I saw that Hornby had made one of these, Chartehouse for the era I wanted (Dad's Army meets Foyles War) and that these could be electronically controlled and adapted to make noise as well - I was hooked.
     
    The question then was where to put a largish 00 layout that would do justice to the advances in technology and build quality in model railways. The only two options being shed, or loft.  The loft won because we needed to properly insulate and board it anyway for storage. as you can see

     
    This meant the on cost of a layout would really be the those extra modifications needed - Velux windows, eaves insulation and electrical work to put certified wiring for lights and mains. 
     
    I had to wait a while for our builder to carry out the modifications as mucking about with lofts like this is best left to experts.  He fitted loft insulation and clad that in 5mm ply to a basic standard rather than leaving us with the mylar space ship look - it is a loft after all! I also asked him to add more structural timber (100x50mm) to the stretchers which would support my layout and avoid it affecting the roof. 

     
    While I was waiting I had plenty of time for research and planning and also for retail!  So as well as lots of good ideas and draft layouts I have a lot of the rolling stock I wanted:
     
     I am well kitted out with SR stock as it was 1935 -1945, I have a couple of Schools, an Adams radial, Wainwright, Terriers , Greyhound and T9s and a brass kit of an air-smoothed Merchant Navy by PDK  to make from scratch.  I have also the necessary Maunsell/Bullied coaches and wagons to suit I also decided to collect the key streamlined trains of the period - my layout my rules! I have acquired the  Duchess of Hamilton (R3339 as preserved) and Silver Fox (R3309)both with six of the relevant but rather badly modelled coaches (the Hornby jubilee coaches from LMS and LNER).  We honeymooned on the Orient Express so a few Pullman coaches as a reminder of our best train journey ever Mrs Deepfat loved the SECR Rails D class when she took me to the NRM in York so I have that and SECR birdcage coaches to suit  
    Given this rolling stock I'd need a sizable layout with lazy curves but at least I had the space now!
     
     
  7. deepfat
    Having deleted RailMaster, put my Hornby Elite in the electronics recycling bin and swapped in a Roco Z21 controller with iTrain I wanted to see how they worked with my layout before getting into the detection side of things. 
     
    Here's my layout in Hornby RailMaster Pro:

     
    I have enabled displaying the point's ID numbers as I want to refer to them when setting up the same points in iTrain. The old Railmaster  diagram does resemble my layout in SCARM

     
     
    It's roughly the same but RailMaster is confusing for me - Even at 50% there aren't really enough squares to make it look like it is and you have to fudge 3 way points and double slips by using two points for each of these.
     
    To be fair my first go at iTrains was daunting. Some things are familiar but there lots of detailed settings and of course the jargon used is different. I started by slowly by just creating the switchboard to reflect my layout:

     
     
    To my eye this looks neater and more like the actual layout:
    There are specific symbols for 3 way points and double slips rather than the Railmaster bodge of using two points to represent them. I can see my station and the turntable is resizable although I have yet to try and see how to get an ADM turntable to work with it The size of the grid doesn't matter so much and the curved corners are more elegant  
    Another key feature for me is that iTrain supports modern operating systems from Windows 10/11 (I am running it on windows 11 myself)  to Apple (OSX and IOS) and linux distros
     
    The obvious downside of iTrain Plus edition is at least double the price the price of Railmaster Pro. However unlike RailMaster which is strictly per device, iTrain can be installed on many devices so I can design layouts in my office, use my all in one Touchscreen PC as my main layout controller, and have it on a laptop right next to where I am working if I am troubleshooting or configuring new stuff.  I can use any of these to drive my layout. Moreover iTrain is designed around detection and automation and because I’ll be using those features, it is better value for money for me. 
     
    I also think some railway modellers can  be put off by the complexity and richness of iTrain.  That's partly why I am writing this.  I could have spent hours getting confused on YouTube watching others try to explain how to use it and use it and all the theory, but I chose a different path. Start simple, learn the basics and apply the principle of RTFM  (Read the ... Manual) :
    Get the switchboard looking like the layout you have Configure every point so that they throw correctly to match what you can see on the switchboard Add a test loco and run it! All the advanced stuff for routes, detection and  speeds can wait until you are ready to explore more.  It's like learning to play guitar , learn three chords (may be enough if you are a Quo fan) , learn strumming, learn timing and rock 'n roll you are making music.  Sure there are more chords, licks and playing with your teeth, but that's for later.
     
    That's what I have now done and a couple of days I feel at home and I really like using it - For example the fact that all of the settings for every accessory, loco and can be seen in a list that can be sorted and there is a diagnosis screen to show you what needs fixing later -For example I have not done any measuring yet (more on that later)
     
     
     
    It's also really really stable.
     
    So now I have found my way around iTrain my next task is to fix the wiring  - Both to move to a track/accessory bus setup and to wire in the Digikeijs DR5088RC detectors.
     
    Finally a polite request  - if there is anything wrong in here please let me know, as  I do see a lot of conflicting advice out there, and my mission is to keep things simple but accurate.
     
     
     


  8. deepfat

    Euston we have a problem

    By deepfat,


    wiring
    Apollo 13 suffered from a main B Bus undervolt resulting from an explosion. my layout used to just have a single bus but now I do have two one for Track power and on for accessories. However I did get a couple of shorts after some redesign work and that took a long time to find so here’s the sorry saga of how not to wire up your layout and what I did to fix it, which I hope you'll find useful
     
    As I say, I now have a dual bus setup and if there is a short on the track  it trips the Z21 booster, not the z21 controller which mean I can still through points etc. Also unlike the tired old RailMaster controller the Z21 resets itself as soon as the trip is fixed so I spend less time crawling around the layout to test fixes.  I should add  just have the one bus as my layout only has 70m of track 40 points, and I'll never run more then 4-5 locos simultaneously as I have no automatic control. However I and others who adopt this approach get a lot of flak, partly because it can be hard to diagnose shorts (you can't make this up can you), etc.
     
    Originally I when I just had the one bus  I used the bus wiring connectors from DCC Concepts to attach dropper wires and wiring to my points to the bus:
     

    Frankly I always found these connectors a pain to use -
    I am rubbish at soldering and I found it hard to get a good joint from these to the bus which occasionally meant something didn't work I need to add a lot of these to the bus and that means breaking into the bus again and again.  
    What really got me thinking about this was that I needed to add my equivalent of a fiddle yard (which isn't really as it's on show)  and the trackwork to go around my awesome new ADM Turntable (the subject of another post). I wondered if there was a better way to wire up the new track points and turntable to my bus which looked tidier and was easy to understand. On the DCC Concepts forum I noticed my good friend Bunkerbarge using a different sort of connector (from CM3 models) to attach lots of droppers to his track.   These gizmos allow up to 9 feeds to be connected via one of these blocks which in turn is connected to the bus and it has screw in terminals which meant less soldering👍  
     
    So I ordered a load of them and proceeded to rewire my whole bus using these new CM3 connectors and here's that some part of the layout again with just the one bus

     
    Having done all of this when I turned on my Z21 I got a short, but how to find it?  With these new connector blocks all I needed to do was disconnect the feed wires from each these connector blocks to the bus in turn until there was no short.  It did not take me too long to realise that I hadn't used insulating joiners for the middle route on a three way point.
     
     

     
      The second was much harder. Something weird was going on around a particular point in another siding and whenever I connected track to it.  My senior moment was that I forgot to properly reconfigure this Peco Electrofrog point - I hadn't removed the bridging wires connecting the frog to the point blades under the point though I had electrically connected  the point blades to the outer rails. 
     
    This all took me longer than it should have because I did not do what I would do at work when making changes to designs.  I mention this because I work in software engineering at Microsoft and despite what you may think 😃, we spend much more time testing our work than we do writing new code. So if I had done proper testing when re-wiring I would have gotten my answers much more quickly. Specifically
     
    Regression Test  - is everything OK before a change is made? Not doing this means that you have no idea if a change you make is the problem or if it was there already.
     
    Unit Test - Test what you have changed -  Does the change you just made work? For example is that siding you just added working OK?
     
    Integration Test - Test the whole system again. It might be that you have introduced a problem somewhere else and not realised.
     
    However you wire up your layout, add decoders to trains, put in signalling and lighting these principles should help you particular if you are collaborating on a large layout.
     
    Footnote: since writing this I have moved on to loco detection (and here's how my layout is now wired up) and now my track is isolated into 5 zones each powered by a separate Digikiejs DC5088RC  detection module and so I can quickly establish where a fault is and then dig into the specific section 
  9. deepfat

    ADM turntable

    By deepfat,


    turntable, design
    A quick search on the internet for turntables will likely get you a list of products for spinning your vinyl collection
    rather than for an accessory for your model rail layout. I mention this because looking underneath an ADM turntable you'd think it was a record player - It has a toothed belt drive and a pretty sophisticated circuit board amongst the other wizadry.  Land Rover may have had a hand in its design as it looks very robust and is reassuringly heavy and it's so well built you could probably add a stylus and listen to music at say 1/2 rpm!
     
    These things are from cheap and I have to confess what your looking at here is the fully loaded version with detailing, weathering and an adapter kit which I'll explain later. However I don't have the skills to make something as good as this and the adaptation kits for motorising turntables seem flimsy and suffer from indexing and drift.
     
     
    Of course it is a thing of beauty as well and here's the one I bought earlier in lockdown..

    The queen of turntables from ADM.
     
    I'd always wanted a turntable and had planned to put one on my layout.  I thought it might be good to review the design before cutting holes in my baseboard. My loft layout is resting on beams that spread the roof in my loft and the left hole is for the loft hatch and the right hole will have a removable section of scenery for access which presents interesting challenges for my layout. My initial design for including a turntable looked like this..
    with three routes on to the layout and a small engine shed.  However these routes weren't long enough to store rakes of coaches and this is important to me as there is no fiddle yard on this layout as it's all in view.  However I do  I want somewhere to form up trains to change what is running  and to make the turntable a real feature as part of this. I also like the idea of a roundhouse  and tried to squeeze a small one in.
     
    After a lot of versions this is what I came up with using the SCARM software to visualise how it might work. Essentially I stole some space by  shortening the branch line siding to the left of the layout and using that space to run a set of sidings between the turntable and my mainline:
     

     
    With this design  a  loco would drive into either route A or B with six coaches and uncouple to leave the rake behind and continue to drive into the siding C.  It could then reverse out to the main line on route D or reverse on to the turntable turn around and leave on route E or enter the round house. A new loco coming from the roundhouse or the mainline could cross the turntable to pull out a rake of coaches and return to the mainline with them. This looked like a good compromise so  onto the installation.
     
    I mentioned my  ADM turntable came with an adapter ring. This is  it can be removed from the layout without disturbing it.  The turntable sits on this ring of mdf which fits in turn into  a square mdf mounting plate.
    All I need to do was cut the right size square in the right place in my layout and secure the base plate to fit flush with the top of the baseboard

     
    The Wiring is easy - you get a DC transformer a cable to connect the controller to the turntable and there are two connectors to get power to the turntable  and this auto reverses the power as needed. Note the DCC power goes through the circuit board so the turntable must be powered on to get DCC power to the turntable bridge.

    And as you can see here I have followed the deign I had in mind..
     

    Except that I reckon I can get a 6 bay roundhouse by using peco lifts to simulate how that would work.  I can sort the details out when i have the roundhouse kit as the turntable has amazing fine grain control to get the rails to line up so I can have the entrances anywhere I want.  Just as importantly when you turn it off and on again it goes through a start-up process to check it's properly aligned and I love how it simulates inertia and slows down as it comes to rest. 
    It's very easy top operate and I have replaced the cat5 cable with a longer one so I can have the controller next to my touch screen. My problem now is to try and get this baby working from iTrain with detection and all, but that might be a post in itself.
     
     
    Finally thanks to Alistair at ADM for great aftersales service (note to self  - rtfm)
     
  10. deepfat

    Detection and automation basics

    By deepfat,


    dcc train automation
    While it's possible to run trains automatically on a layout by relying on accurate measurement and knowing exactly how fast your trains are moving, in practice you need feedback about where your trains are to make automation reliable, otherwise positions will drift  due to all sort of variations and inaccuracies.  There are several ways to detect a train on a track including optical detectors, switches, magnets and by measuring a voltage drop on a section of track.  However in most of these scenarios there is no information on exactly what has been detected, a bit like your neighbourhood fox tripping you PIR activated lights around the house where what you really want is to know when a person is there. As I wanted to know which loco was where on my layout, I went for a Railcom based system because I think this is the only way to do that right now. This allows a Railcom enabled decoder to send its ID and direction of travel back to a feedback detector module and then to the controller and finally to the control software.
     
    The best bang for your buck in Railcom feedback detection is probably the Digikiejs DC5088RC which has 16 detectors on it for about £100-ish and that's what I went for they are easy for a geek like me to setup - you just plug them into your laptop taking care to only c9onnect them via USB isolator if you want to test them while the DCC system they are connected to is on.
     
    Here you can see my APT-E (ID 152) showing up in the Digikeijs software:

     
    The next question is how many feedback detectors are needed and where do they need to be? Here's my plan (with thanks to Iain Morrison for getting me started on this):
     

    You can see I have 5 feedback detectors giving me detection across a maximum of 80 sections of track.  You can also see I have two reverse loop modules and I'll discuss those in another post, when they are installed and working.
     
    I am using iTrain and this works by using a combination of detection, accurate speeds for trains and recording  the lengths of all the sections of track and points that make up a layout.  So when a train is detected iTrain can predict when it will enter the next detection section and use this to set points and stop trains in a station and so on.  However it is not precise -  Say there is a 1% margin of error on a 2m length of track. By the time a train comes to the end of that section you could be 2cm out, so having more detectors where trains are likely to brake and stop reduces this error.  For example in my 3m long station I'll assign 3 detectors , two short ones at either end of the platform for stopping depending on direction and a longer section to connect them.  Precise positioning is also important for shunting and uncoupling and that's another thing I plan to automate later on.
     
    iTrain has a concept of a block that mirrors how real railways work where a block can only be occupied by one train (where a train is made up of one or mor locos plus other rolling stock) at a time.  Each block can have many feedback detectors but does not include points, so just one way in to the block and one way out or none if it's a siding.   As an  example in my main station  I have a block for every platform each with 3 detectors in.
     
    It's pretty simple if a little tedious to describe blocks and feedbacks in iTrain but then I need to make that real by making changes to my layout:
    Wire up the feedback detectors to my track, Connect them via Loconet (Lnet in Roco speak) back to the DCC controller (Z21) Enter the details of the feedback detectors in iTrain. Test, Test, Test  
    That and splitting my single bus setup into a track plus accessory bus configuration required a lot of rewiring. Part of that was to add track power to each outside rail of all my points as I planned to fully insulate them from the adjacent tracks as I may want to add detection over points at a later date. This means the only things I have directly connected to the track bus are the detectors themselves - all power to detected sections,  point rails and frogs flows through one of these units..

     
     so I sort of have 7 buses on my layout! One for each of my 5 feedback detectors , the track bus and the accessory bus.  I am aware this might be overkill but it works.  It's also worth noting that both rails must be gapped between one detector and the next so that the common rail  (pink in my diagram) gets power from the same detector that's doing the detecting  
     
    Of course drawing diagrams is easy so here's the spaghetti under my layout..

    I guess I could have invested in a more Technicolor wiring scheme but the sticky labels I have attached to the wires are good enough for me.
    I know I do now have a real mess of wires like my DC friends and some of the runs of dropper wires are longer than usual as the black wires all need to go to one of the five detectors.  However, the difference is that I now know exactly where my trains are as we'll see next time.
     
    Finally thanks to my new friends Rich at YouChoos and James at DCC Train Automation who have been really helpful
     
  11. deepfat
    I have several posts on RailMaster(pro edition) on this blog as I got more and more into it. It is simple to use and suited my needs as someone new to DCC. However I have been more and more frustrated with its reliability:
    It is not designed for Windows 10 It has an arcane licensing setup It has a lot of lag on throwing points and sending functions to locos as it is working like an old dial up modem over a com port and this also means it takes ages to read and write CVs to locos It now has an annoying feature (aka bug) where newly added locos only go at full speed when you set any throttle at all. I was also hoping that detection would be added one day..

     
     
    However what I have seen is Hornby reverting to DC control and no innovation of either its ageing Elite controller or the RailMaster software,  so I finally decided to swap out the Horny components, but for what?
     
    My goal has always been to run two or more locos at once  without worrying that they'll collide, derail or cause a short  as they run across points that are set against them.  I have sort of prevented this up until now by having safety track where the power is only applied if the point is set correctly, but this is crude as the loco just stops as do any sounds or lights it has.  
    Now thanks to a great afternoon with Iain Morrison I have seen how detection can bring a layout to life by having multiple trains running without needing to watch over them constantly. I have pretty much decided to replicate what Iain uses, having read a lot of forums and watched a lot of YouTube.
    I budgeted for £2K for this project  which might sound a lot but most of my fellow enthusiasts have spent over 10K on trains alone. I was more worried about how much rewiring I had to do which I'll cover in an a separate post. 
    This is what I ended which is also what Iain has:
    iTrain Plus software for automation Roco Z21 and Z21 single booster for DCC control as I now plan to have a separate track and accessory bus 5 x Digikiejs DR5088RC detection units and a loco net hub However I did do my own research:
     
    iTrain Plus
    My decision was based on quality, reliability and innovation, coupled with ease of use. I could have used the free jmri but the interface is awful I couldn't get it to install and the manual is a train wreck. Train Controller is way more expensive and doesn't seem to be as innovative or offer anything I need over iTrain.
     
    Roco Z21
    Not the cheapest, but supports loads of standards and it just worked when I plugged it into my home network - I did not even need to use the supplied router as I am whizz on computer networks.  I really liked how it resets a short and the booster has the option not to pass the short to the Z21 itself ( the Z21 drives the accessory bus and the booster is used for the track bus).  Plus it's got great diagnostics and update tools and just works with iTrain.  I was also impressed with its programming on main and how fast it reads cvs on locos.
     
    Digikiejs DR5088RC detectors
    They do what they say on the tin. I was impressed with the diagnostics software and if you get an isolating usb dongle to connect one of these to a laptop and have it powered to your track you can see it detecting:

     
     So at £90 for 16 detection points the cost seems reasonable to me
     
    Putting that combination together I was able to quickly test that  iTrain picked up that detection information to display it on the switchboard of my layout.

     
    City of London 3439 in siding C
     
    There's a lot to setting up automation, but I have already seen the benefits of this new setup:
    I can control my layout right near where I am testing as my laptop running iTrain is on the house  network along with the Roco Z21, so if I have made a mistake somewhere I can test much more quickly without navigating all the beams in my loft.
     
    In the following posts I want to share my journey to make all this work, as I have noticed there is good content on each part of the system by not how they work together
     
×
×
  • Create New...