Upper Hembury, East Devon - a GWR / SR Branch line

[BWsTrains]

Some of you will already be familiar with this topic from my posts in the “Layout & Track Design” area of the site and to you plus newcomers, a warm welcome.  As my plans have been progressed as far as they reasonably can without doing something material, it’s time for me now to move on to the building. The locale and its connection to the railways are documented as a back story in the first few posts of planning Topic:

https://www.rmweb.co.uk/topic/170320-upper-hembury-designing-a-gwr-branch-line-in-east-devon/

 

For newcomers who wish a short cut, subtle changes long ago led to this sleepy area of East Devon becoming a prosperous and busy centre focussed around Upper Hembury. A market town which lacked only one thing in the late 1870s – connections to the wider world via rail.

 

By 1882 the GWR had arrived and the LSWR (later SR) gained some limited access from the South on not the best of terms. It was and remained primarily GWR territory. The layout will be built in OO Gauge Code 75 bullhead rail using British Finescale turnout kits throughout. It will have a 7-lane stock storage traverser which will be motor driven. At 4200mm * 1400mm (at its widest) it represents the approaches to and the full station precinct.  A spur leads to a small local industry powered by the nearby River Tale, the Upper Hembury Tannery.

 

The setting is based in the mid-1930s after 50 years of operations. Local businesses have supported and prospered along with the railway, and we will find them mostly based around the Goods Yards.

This is the plan, I hope that regular reports from here on will generate a reasonable amount of comment, suggestions, etc. to guide and support me on my way.

 

Edited September 2, 2022 by BWsTrains
keywords update

[longchap]

Looking forward to your further progress Colin, particularly as you now have your cooler seasons. 

 

Bon courage,

 

Bill 

[Graham T]

Looks interesting, I'll be watching this one!

[BWsTrains]

The L Girder framework was built for me by a carpenter friend. As a result it went together accurately and quickly 😎, a job I'm glad that I outsourced. The original temporary front supports were replaced by angled stays when all was in place.

 

 

A bay for the traverser tray

 

Underside view with final supports to give greater extension for the Tannery area.

 

 

 

[BWsTrains]

Building the Traverser Drive Assembly

1)  The drive

I chose to use a traverser design for storage because with my total free length of 4100mm (13.5ft) I needed all the usable space I could get and putting in a rake of sidings would be wasteful in that regard. With 910mm allocated for the storage traverser plate that left me with 3200mm free space.

 

A friend of long standing is a whiz at all matters mechanical, so I presented him the challenge of mechanising a traverser for me. He encouraged me to buy a stepper motor# and his solution was ingenious, many levels above my capabilities##.

 

What I can disclose are the mechanical aspects which cleverly overcome the problem of trying to achieve accurate control via programming “step counting”. Individual stopping points are defined simply by photo sensing devices and when one is triggered the motor stops. The operator chooses a photo interrupter to be “active” and the stepper motor advances until this one is reached. Fine control is thus achieved by the positioning to the detectors. We have achieved +/-0.3mm stopping accuracy but the acid test lies ahead at deployment.

 

There are cut-outs at each end if the drive is directed to travel beyond the final stop and the controls are simple.

·         Five stop positions

·         a direction selector - forward/back

·         an emergency stop; JIC!

 

# My motor cost me about $A80 from a reputable on-line retailer

##He is also in the process of getting the electronics design published so I can say little about that aspect, but I suspect that anyone capable of controlling a stepper motor could design their own version as the principles are well known  

 

[BWsTrains]

Building the Traverser Drive Assembly

 

1)  The Traverser plate design

My operational plans required 7 storage lines (tray width 320mm) and I looked at possible two designs, single- or double-line access.

A single access line to the storage would mean the tray traversing 6 track placements, this in turn requiring a longer more expensive motor, 7 stop detectors and wasted space to house the tray across the entire range of its movement.

Hence, I settled for using two approach lines as shown in the plan from my earlier post. Now the tray travel is reduced by 90mm, and this valuable space is gained at the front where the full 4100mm length is available for operations.

 

2)  Positioning the Tray

The tray was built from quality 20mm high density particle board with epoxy coated aluminium angle bracing supporting front and back edges. The tray was designed to fit into this space created in the framing.

The ends of the tray resting on the polished slides at each side (non-load bearing, simply guides). Two ball bearing drawer slides were fitted to the bottom of the plate, these to slot into their supports ~1/3 way in from the edges, with the motor placement in the centre.

 

 

The motor assembly was installed after a little more cutting, it was wider and deeper than I’d expected. I bet they didn't have this problem when a new part for the first A380 arrived at Toulouse from Broughton UK!

 

 

A support block bolted to the drive provided a means to secure the tray from the top and make fine adjustments to its position

 

Edited August 15, 2022 by BWsTrains

[Dragonboy]

Looking forward to watching this develop

 

Brian

[Ighten]

Me too.. 

[BWsTrains]

As I'm sure you'll realise, the progress with the build which I've reported so far didn't happen overnight. Other important considerations meant I could take a steady approach but I'm now almost up to date with these two photos of the back scene boards.

 

My particular challenge was the solid feature brick wall on the long side and my solution was designed to avoid the need for intrusive / damaging mountings. As you'll see I managed to mount the primed MDF panels completely flush with the wall. What's more they're easily removable for cleaning behind or for changes.

 

 

The mounting mechanism is very simple. A piece of Neodymium magnet was mounted on a "U" shaped steel strip, the flat bottom of the "U" acting as a field intensifier. This fits neatly into the rebated mortar course so it can sit flush with the wall, I recall I used about 7 magnets in total, ones at the boards' interior edges placed to hold both boards.

 

A thin steel strip was glued to each corresponding position on the board, and this holds it firmly to the wall.

 

 

Edited August 17, 2022 by BWsTrains
magnet details

[BWsTrains]

I've returned to the Traverser now with various key pieces of track to hand, and I set about installing the 7 lines.

 

Originally I'd designed the tray to be 914mm long for three reasons:

• It suited the maximum typical train compositions I intended to run,
• was no more than 1/3 the available running space (which I read somewhere is a good ratio not to exceed)
• Equals a length of flex! (no track cutting)

In the end, point 3 went out the window.

 

Firstly, to improve access for work I removed the tray, the design is an easy lift off after removing the locking screws. It gives a view inside of the final structure with the mounting block

 

 

Finding myself with a substantial supply of set-track, very handy when you want dead straight parallel lines, I assembled the 7 lines easily in no time. Then to wiring up, requiring just one dropper pair which I'll connect to DCC via a flexible lead.

 

 

 

Set track for the tray has another advantage aside from being dead straight. This comes from the design which is made to stay together unattached to a base, viz. the track connectors have never failed me regarding connectivity. This is clearly one design aspect the manufacture got spot on. Anyone who has impaled themself working with the track joiners will know what I mean; they're made to go on and stay on!

 

With this job all but done, I'll be moving on to the serious track laying tasks on the layout.

 

 

[Physicsman]

Colin, thanks for the heads-up on your layout thread. Really good to see you on here and, as you know, I do like a bit of baseboard construction and you're doing an excellent job with all of this.

 

I rarely stray outside of the S&C in my random visits to the Forum, but I'll keep an eye on this. What you are building oozes quality. If the base isn't properly built, heaven help the rest!!

 

Jeff

[BWsTrains]

With the storage all set and ready to go there was no excuse for delays to laying track.

 

Starting at the storage entrance I have two "S" curves leading the onto the exits from a B7 "Y" turnout. I used cork underlay here to get the levels matching across the gap however to keep things simple I worked on without laying it very far, at this stage.

 

Thanks to the features of the turnout kits (see post to follow) I realised I could make a break in the "S" section at around the midpoint, simplifying the track laying. In the first photo the view from the traverser shows the exit roads as they join onto the turnout The track is only held down at a few anchor points here so is not fully smoothed curves. TBH I'm feeling my way through this as working with very light, delicate and motile track and flexible turnout kits is all new territory. Hence I'm doing a dry run of installation fixing in place with easily adjustable clamp plates, no track joiners, droppers etc..

 

I think I'll get about to the mid-point before having enough "feel" that I'm not making any major errors. So far, so good!

 

 

The "Y" turnout has the exit lanes extended by ~90mm, the joins being marked here with arrows

 

 

A first wagon has run all across the full length of laid track and the crossing from the storage tray was satisfyingly undetectable.

 

 

 

Edited August 25, 2022 by BWsTrains

[St Enodoc]

2 hours ago, BWsTrains said:

With the storage all set and ready to go there was no excuse for delays to laying track.

 

Starting at the storage entrance I have two "S" curves leading the onto the exits from a B7 "Y" turnout. I used cork underlay here to get the levels matching across the gap however to keep things simple I worked on without laying it very far, at this stage.

 

Thanks to the features of the turnout kits (see post to follow) I realised I could make a break in the "S" section at around the midpoint, simplifying the track laying. In the first photo the view from the traverser shows the exit roads as they join onto the turnout The track is only held down at a few anchor points here so is not fully smoothed curves. TBH I'm feeling my way through this as working with very light, delicate and motile track and flexible turnout kits is all new territory. Hence I'm doing a dry run of installation fixing in place with easily adjustable clamp plates, no track joiners, droppers etc..

 

I think I'll get about to the mid-point before having enough "feel" that I'm not making any major errors. So far, so good!

 

 

The "Y" turnout has the exit lanes extended by ~90mm, the joins being marked here with arrows

 

 

A first wagon has run all across the full length of laid track and the crossing from the storage tray was satisfyingly undetectable.

 

 

 

Looks good Colin. Try to make sure you have a short length of straight track (a couple of inches will do) between the halves of the reverse curves. That will improve running considerably.

[BWsTrains]

A look at assembly of a British Finescale 'Finetrax' Turnout Kit.

 

First, I should clarify that the manufacturer has changed the design of these turnouts since I bought the majority of my needs for Upper Hembury. Mine are the original design for all my LH and 3 of my RH turnouts and these notes relate to assembling one of these including adjusting the curvature.

 

The old designs have switch blades already soldered to the their tie bar pins and more important a single piece cast frog assembly, now replaced by an all rail product. Hence not all my comments can be applied when assembling the all rail design. Very comprehensive instructions are available to download and the topic on RMWeb has lots of useful help.

 

The kit comes as a (very) flexible base comprising all the sleepers and chairs plus the various rail lengths and the frog. As I wanted to introduce a significant reverse curve in my B7 Right to create a "Y" I deviated from the regular assembly sequence (marked below by an #). Note I had plenty of experience by this time, it would pay to complete at least one unmodified turnout before advancing!

 

Using the Templot trackplan which I'd developed and printed out, it made most sense for me to do the assembly in situ so I could mirror the desired curvature from the plan. This photo shows the template with the base as supplied and the two stock rails (#) inserted at an early stage, the purpose being to hold the base together when many cuts would be made to the webbing. The base was held in place as required using the plastic cleats over the sleepers shown in my earlier post.

 

Two things are worth noting:

1) The plan shows the degree of reverse curve I needed to achieve in the turnout to generate a "Y" structure where both radii are the same in absolute terms (~ +/- 2375mm). 

2) The ability to readily extend the exit lanes beyond the formal dimensions of the kit (more on this later)

 

 

Details of the required cuts are on the kit RMWeb topic and when the appropriate webbing is cut the base template readily bends into the required shape. At this point I glued back the cuts in the webbing to retain the new shape. From here the subsequent steps follow the instructions up to fitting the tie bar assembly and switch rails. Then the stock rails are retracted to allow the tie bar to fit in its slot.

 

As can be seen from the plan, the extra sleepers for the exit extensions need to be interlaced /shortened to fit. Something much easier done one sleeper at a time onto bare rail than trying to attach two free lengths of track. Overall the finished "Y" with extensions comes in at 380mm end to end and can be seen in my earlier post of Aug 25th.

 

Regarding the cast frog, while it might not look as quite as good on close inspection, it looks perfectly fine in situ and has two benefits which suited me. A single wire to the entire frog assembly and the ends of this can be easily insulated, removing the need to use insulating rail joiners.

Edited September 2, 2022 by BWsTrains

[BWsTrains]

Jumping around subjects a bit, I've been looking for all the various supplies I need to order in from O/S and one item was Yard Dirt.

 

I was very taken by @checkrail's goods yard design on Stoke Courtney and figured the texture achieved in the yard space would be ideal.

https://content-eu.invisioncic.com/y320084/monthly_2022_05/y1.jpg.831733e2cde6ae49d20bb0dc4494377e.jpg

 

Unfortunately it seems the materials of choice for this job are not available locally or cost a bundle. As I'd rather spend my AUDs on something higher value than "Dirt" I did my own research.

 

I've found an initial solution and am interested in viewers' reactions. The shade as seen is the natural material but can be adjusted if required by application of a dilute acrylic wash, something which can be decided in situ.

 

Particle size is 0- 0.5mm, here attached to some spare board by first applying a very light cream shade of a thick acrylic exterior paint. Hence not biased by the earlier rough paint (mess).

Woodland Scenics Medium light grey ballast for reference.

 

As to the colour I'm happy with and fine adjustments will be possible. As to particle size, 30mm stones in the full scale yard seem too coarse but Its more about look than real people walking on real rocks and I suspect being somewhat over-scale might give a better realism at 1:76 scale. Your thoughts will be appreciated.

 

 

 

 

 

Edited August 3, 2023 by BWsTrains
reorganised layout

[St Enodoc]

1 hour ago, BWsTrains said:

Jumping around subjects a bit, I've been looking for all the various supplies I need to order in from O/S and one item was Yard Dirt.

 

I was very taken by @checkrail's goods yard design on Stoke Courtney and figured the texture achieved in the yard space would be ideal.

https://content-eu.invisioncic.com/y320084/monthly_2022_05/y1.jpg.831733e2cde6ae49d20bb0dc4494377e.jpg

 

Unfortunately it seems the materials of choice for this job are not available locally or cost a bundle. As I'd rather spend my AUDs on something higher value than "Dirt" I did my own research.

 

I've found an initial solution and am interested in viewers' reactions. The shade as seen is the natural material but can be adjusted if required by application of a dilute acrylic wash, something which can be decided in situ.

 

 

Particle size is 0- 0.5mm, here attached to some spare board by first applying a very light cream shade of a thick acrylic exterior paint. Hence not biased by the earlier rough paint (mess).

Woodland Scenics Medium light grey ballast for reference.

 

As to the colour I'm happy with and fine adjustments will be possible. As to particle size, 30mm stones in the full scale yard seem too coarse but Its more about look than real people walking on real rocks and I suspect being somewhat over-scale might give a better realism at 1:76 scale. Your thoughts will be appreciated.

 

 

 

Looks good Colin - are you going to tell us what it is?

[BWsTrains]

On 01/09/2022 at 14:22, St Enodoc said:

Looks good Colin - are you going to tell us what it is?

 

Thanks.

 

Mineral gypsum is literally dirt cheap, there are whole deserts of it in various places. The bags I was buying were $0.75 / kg and one needs a fair bit as a clay breaker. I broke out my useful range of various graded sieves. 1Kg gave me ~200g lump (mostly >3mm) about ~100g ( of 1.0- 2mm) ~200g (0.5-1.0mm) with the remaining ~500g 0-0.5mm as used in my demo.

 

It is interesting to compare the gypsum and brickies sand of the same <0.5mm grade under a lens. The difference is quite striking and easy to see why gypsum is better. It is quite soft and the particle shapes overall far less defined.

 

Edited June 14, 2023 by BWsTrains
tidy up

[Dragonboy]

I’m also building a layout and points using British Finescale track but in N. Just grateful that as a geriatric I still have skills and eyesight to do it, but I was alerted yesterday to an incredibly useful YouTube video from James Hilton Custom Model Railways who video'd himself making a point and showing how easy it was.

 

Brian

[BWsTrains]

British Finescale 'Finetrax' Kits - B7 Single Slip

Much excitement around these parts, and relief, when this kit arrived in the mail in record time from the UK a month ago.

The single slip has a crucial and central position in the Upper Hembury track-plan and so proceedings had been on hold until this latest addition to the Finetrax Code 75 OO Bullhead range became available. A quick check confirmed that the base structure fitted perfectly to my Templot plan (✔️). This is a more complex build than a standard turnout so I took detailed photos along the way. These follow with some comments below:

1.

 

2.

 

3.

 

4.

 

5.

 

6.

 

7.

 

8.

 

1. The Kit components
2. The track plan, printed onto 2 * A4 sheets which are then joined
3. The first pieces - Central "V" rails
4. This shows the tie bars, one assembled with its 2 pins the other en route. More rail elements added
5. About half way. The two frogs are being assembled just before inserting the V Crossing rails
6. Getting there, just the check and stock rails to install
7. Assembled but not yet wired up. The potential for lane extensions can be seen here.
8. Reverse side showing the various wiring arrangements (probably not final, just done according to instructions at this stage)
   Note also the short extensions made to two exit lanes of which more later.
   
   Below
9. The near finished product in situ  (wrong way round as the sharp eyed might note). The slip side should run from Goods Yards to main.
   
   

 

Having first discovered that track could be built by following Gordon S's marvelous site, I never imagined that I might sometime make small steps along this path. These kits make that entirely possible.

 

 

 

 

Edited September 3, 2022 by BWsTrains

[checkrail]

These look like excellent kits.  Excellent work on your part too

Edited September 3, 2022 by checkrail

[BWsTrains]

One more excellent use for the Finetrax Kits; building DIY functioning Trap points.

 

I've two locations where these are needed on the track plan, most importantly protecting the main line from traffic coming from the tannery. Traffic through the trap will be controlled by a ground signal. It occurred to me that the Finetrax sleeper base which comes as two joined halves would provide a convenient starting place to build a working single rail trap (as per GWR practice).

 

A sacrificial half template and a left switch blade / tie bar were combined with the required running rails to give me this.

 

 

Curved to fit the profile of the approach from the tannery to the single slip ahead. Note with only one switch blade and no point motor the blade tip is sitting more to the right than it will after installation.

 

A work of caution: I had a cunning plan to make the unit extend further than shown here by using rail extracted from PECO Bullhead track. However, the sleeper chairs of the template must be microns too tight to accept the rail. Bullhead supplied by British Finescale of course presents no problems.

 

Where it sits in the bigger picture:

 

Trap points are useful as @brossard has recently reminded us!

 

https://www.rmweb.co.uk/topic/174113-catchtrap-point-positioning/#comment-4923508

 

 

Edited September 9, 2022 by BWsTrains

[St Enodoc]

4 hours ago, BWsTrains said:

One more excellent use for the Finetrax Kits; building DIY functioning Trap points.

 

I've two locations where these are needed on the track plan, most importantly protecting the main line from traffic coming from the tannery. Traffic through the trap will be controlled by a ground signal. It occurred to me that the Finetrax sleeper base which comes as two joined halves would provide a convenient starting place to build a working single rail trap (as per GWR practice).

 

A sacrificial half template and a left switch blade / tie bar were combined with the required running rails to give me this.

 

 

Curved to fit the profile of the approach from the tannery to the single slip ahead. Note with only one switch blade and no point motor the blade tip is sitting more to the right than it will after installation.

 

A work of caution: I had a cunning plan to make the unit extend further than shown here by using rail extracted from PECO Bullhead track. However, the sleeper chairs of the template must be microns too tight to accept the rail. Bullhead supplied by British Finescale of course presents no problems.

 

Where it sits in the bigger picture:

 

Trap points are useful as @brossard has recently reminded us!

 

https://www.rmweb.co.uk/topic/174113-catchtrap-point-positioning/#comment-4923508

 

 

That looks excellent, Colin. Let us know how you get on with controlling the movement of the blade now it hasn't got a mate to restrain it.

[BWsTrains]

An Update on the Coming Steps

Nearly two weeks of inaction on my part prompted an update. In between, a minor detour of 3500km over to the West for an important family visit in Perth was one factor while the growing list of Spring garden jobs over here was another.

 

Enough excuses; @St Enodoc ‘s question regarding control of my trap turnout brings me to address an important side job to be done before I get too deeply into track laying.

 

Turnout Control
For the total of 12 turnout devices in my setting I decided to stick with tried and tested solenoid technology with which I’m familiar. My logic; if its good enough for “Hills of the North – The last Great Layout” who am I to argue? There are a few interesting twists in this story that I’ll cover very shortly.

 

Elsewhere - the Station Building
Progress over at Chuffnell Regis prompted some embarrassment action here.  With thanks due to @Graham T for sharing some excellent background material , I’ve done some interesting research and this has led to a design I’m happy to build. Update also to follow shortly, once plans have had some further tweaks.

 

Nothing much concrete done but hopefully progress lies not too far ahead.

 

 

[BWsTrains]

Turnout Control - Capacitor Discharge Unit with some useful features

 

Sticking to the tried and tested, I went in search of a suitable CDU device for my solenoid control. This time I was mindful of the need for suitable frog switching, previously having made do and work arounds to survive using basic insulfrogs.

 

A local kit supplier in Melbourne, viz "Talking Electronics" (no connection ...) who supplied me last time had some new and interesting designs available including one with built in direction indication. As the cost per switch module was less than 4 GBP per turnout for CDU, DPST switch and LED indicators combined - and assembled (!) I was sold.

 

The story between order placement and successful arrival to a final working product matches my very worst online buying experiences, EVER!

 

The culprit was (I'm guessing, now former) employee of a certain Postal Service who decided to cram a package of size X marked fragile into a letter box size 0.5 X rather than drop it off at my door. The switching and other fragile parts were all wrecked. Then followed torturous times claiming compensation, eventually achieved and a rebuild completed. Silver living being I replaced all the SPDT with more robust DPDT switches, immediately providing a route to my frog control. 

 

XXXXXXXXXXXXXXXXXX

 

The concept in this CDU is different. In a standard setup, a momentary switch triggers the fully charged Capacitor to discharge through the solenoid coil in the selected direction. It then recharges quickly for any next action. In the device I'm using, the circuit remains closed, allowing for direction indication from the Module.

The CDU module (blue) is fed by a central capacitor which acts as a charge reservoir. The CDU switching is DPDT ON-ON so there is always one closed circuit in this design.

 

When the switch is in position #1 the CDU capacitor is discharged.

 

When the switch is moved to Position #2, current flows from the Reservoir through one Solenoid coil to charge up the Capacitor on the CDU module. Once the two Capacitor Voltages are matched, current flow through the solenoid is greatly reduced. There is however a minor path via an indicator LED to ground so a very low current continues to flow.

When the switch is reversed to Position #1, the circuit from the CDU Capacitor thru the other Solenoid Coil to ground is closed and the CDU Capacitor discharges via that coil. The design allows for a minor residual current which now flows via the other LED (circuit details not shown).

In essence, the CDU Capacitor is charged up from 0V via one solenoid coil then discharged down to 0V via the other solenoid coil when the switching is reversed.

 

As the switching is always closed in one direction, the usual problems with momentary switching for solenoid operation are avoided and the ability to build in frog switching gained. Not bad for 4 quid per turnout!

 

Edit: PS the input is shown here as 15V simply because that's more than enough for my Finetrax Code 75 turnouts. Using a variable controlled PS I dialed down from the default 25V provided for the CDU (*) until I found a voltage delivering suitable strength of action. Fortunately I have stock of 15V regulated PS bricks rated to various current levels.

 

* one neat part of the CDU kit which I didn't need was a voltage boosting front end, designed to deliver up to 25V output to the reservoir from just 9-12v input. This was put aside for a rainy day.

 

Edit: Much later I learned that solenoid motors are less than ideal for use with Finetrax turnouts which are more delicate than commercial stock. This and other benefits, i.e. integrated frog switching resulted in my moving to use DCC Concepts Omega stall motors. One of those learning curve times.

 

 

 

 

 

Edited April 10, 2023 by BWsTrains

[BWsTrains]

Fine spring weather and grandchildren on school holiday duties diverted my attention temporarily but a return to weather more at home in Old Blighty than these parts today brought a reprieve. Out with the soldering iron and some progress has been made.

 

First to the CDU modules. These come as a bank of 5 units, one for each turnout with a power input / power reservoir cap on the front end. The photo shows the first fully repaired module on my "test bench". The turnout to be tested can be fitted to a test bed and held firmly in situ by the plastic side clamps. Power in this case comes from a regulated variable voltage PS built for me by Graham of Traverser fame.

 

 

Unit 3_5 is shown being tested to ensure correct wiring to the solenoid and the indicator LEDs.

 

Both the switch and LEDs were originally soldered directly onto the module PC board but I removed them as many were damaged. This to repair / replace / make it possible to assemble a decent control panel. Dialing down the input voltage I was able to settle on 15DC as a suitable level for operations on several test turnouts.

 

One further minor but very fiddly step was to assemble a small PS interface. Power input will be distributed to each of the three modules of 5 CDUs; here I repurposed mostly old kit onto a compact board. One tiny step.....

 

but it feels like good progress because now assembly of the control panel will all flow from here.

 

 

 

 

Edited September 27, 2022 by BWsTrains