Jump to content

61656

Christleton Junction - 1986 - It’s back (or it will be)

Recommended Posts

On 04/07/2019 at 18:33, Titanius Anglesmith said:

 

If you wouldn't mind please, it would be very helpful. But no problem if not. I've been on the lookout for cost effective 3- or 4-pole relays (general purpose) without much success; you've made me realise that two 2-pole relays in parallel is actually more economic. 

 

Incidently, that station throat point work is looking very impressive! https://www.rmweb.co.uk/community/uploads/emoticons/default_good.gif

Thanks, I’m happy with how the track is going down. I’ve got another couple of points laid today, including the tricky curved crossover. I was never really happy with it on the last layout, so I’m spending a bit more time on it this time. 

 

The relays are 515-517 RS stock code. Like you I soon realised that multi-contact relays are prohibitively expensive. 

  • Informative/Useful 1

Share this post


Link to post
Share on other sites
On 05/07/2019 at 18:51, 61656 said:

Thanks, I’m happy with how the track is going down. I’ve got another couple of points laid today, including the tricky curved crossover. I was never really happy with it on the last layout, so I’m spending a bit more time on it this time. 

 

The relays are 515-517 RS stock code. Like you I soon realised that multi-contact relays are prohibitively expensive. 

 

Thank you for the info, you’re a star!

Share this post


Link to post
Share on other sites

The trials and tribulations of track laying! I’ve been making reasonable progress with the main junction, managing to find the odd hour to keep nibbling away at it. I’ve been pretty happy with the flowing lines of the pointwork so far - at least until I took a couple of pictures yesterday morning.

 

2BD98492-E9E8-4071-A0DB-86FB0C766BAC.jpeg.62451954b8705190a7e8f9bf589e8909.jpeg

 

The first picture shows the slowly growing junction at the platform ends, I’m increasingly pleased with this, particularly as each additional point adds further challenges in keeping everything looking parallel.

 

7747F7A5-DF30-481D-9547-61D2689B5F59.jpeg.d74222e3c91a1e9da7ce483e587a4d1e.jpeg

 

The second shot was supposed to show the smooth flowing lines of the double junction leading into the curved crossover on the Warrington lines. The more I looked at the photo, the less happy I was with the angle between the diamond and the toes of the curved point. Oddly, it didn’t look anything like as bad with the naked eye!

 

A good hour was spent working out how much of a problem this actually was. With a series of rulers, measures, protractors and expletives I worked out that the curved point was about 2 degrees out of alignment. Simply rotating the points by 2 degrees would result in the other end of the crossover diving off the side of the viaduct. Having spent so long ensuring that this wouldn’t happen almost resulted in the whole board being launched out of the window!

 

Reluctantly I lifted the glued crossover off the baseboard to enable me to realign it. Luckily the copydex comes up very easily from the underlay. I had envisaged that some baseboard modification would be required, but by extending the straight section from the diamond to the point toes, it actually allows it to line up pretty spot on.

 

94EF063B-1678-4A0F-B516-0BCF893C9EEC.jpeg.dac669c7f5607d7a09b77e442e809bd5.jpeg

 

The right hand rail at the point toe (as you look at it in the photo) needs easing a little to the left; it is very slightly wide to gauge and curves a little tighter than required.

 

That’s 20 of 31 point ends laid. I then get to wrestle with trying to get the flexitrack to be flexi again. Slow progress but satisfying nonetheless.

 

 

E3F7709B-8FB1-47FE-BC73-73D0479ACB1F.jpeg

  • Like 7
  • Craftsmanship/clever 3

Share this post


Link to post
Share on other sites

Doing an absolutely cracking job keep it up 

 

david 

  • Agree 1

Share this post


Link to post
Share on other sites
On 07/07/2019 at 20:42, Titanius Anglesmith said:

 

Thank you for the info, you’re a star!

 

I’ve been looking around for other options; these relays are very small and fiddly to work with (you need to be soldering 10 fairly chunky wires within 15 x 7mm). So far the test relay has worked well but doesn’t stand up to much physical abuse at all.

 

The spec I’ve looked for needs to be latching, with a coil voltage of 5 or 12V dc, DPDT contacts (forget getting anything with more on) and a switching current of at least 5A. I’m looking at around 40 point ends on the main layout, it should be possible to pair up quite a few ends, but still probably 30 drive relays and 40 polarity switching relays. Much more than £3 a relay and this will get prohibitively expensive.

 

Which always ends up at the relays already identified. I’m thinking the probably solution is to mount them on vero-board, to protect the terminal pins. 

 

I probably need to look at the current draw if I put 3 motors in parallel; I think the capacitors should limit the current to a level that will protect the relay terminals, but I need to be sure that they won’t burn out after a handful of operations.

  • Informative/Useful 1

Share this post


Link to post
Share on other sites

Hi, 

Just catching up. Great to see you getting a bit done the trackwork is looking great.

 

Cheers Peter.

Share this post


Link to post
Share on other sites
4 hours ago, 61656 said:

 

I’ve been looking around for other options; these relays are very small and fiddly to work with (you need to be soldering 10 fairly chunky wires within 15 x 7mm). So far the test relay has worked well but doesn’t stand up to much physical abuse at all.

 

The spec I’ve looked for needs to be latching, with a coil voltage of 5 or 12V dc, DPDT contacts (forget getting anything with more on) and a switching current of at least 5A. I’m looking at around 40 point ends on the main layout, it should be possible to pair up quite a few ends, but still probably 30 drive relays and 40 polarity switching relays. Much more than £3 a relay and this will get prohibitively expensive.

 

Which always ends up at the relays already identified. I’m thinking the probably solution is to mount them on vero-board, to protect the terminal pins. 

 

It’s reassuring to see you’re going through the same dilemmas as I am, albeit mine are just an imaginary exercise at this stage. My intended layout is quite modest compared to yours; nine ends, eight of which are double-ended. Even then the number of relays needed for interlocking and point drives amount to a fair few. I considered doing the interlocking electronically via an Arduino (just mimicking the locking of a lever frame), but the inductive point motor loads still require an interface of some sort, rendering the idea a bit pointless. 

 

I haven’t looked at the pin-out of the relays yet. I wonder whether it would be practical to plug them in to a common IC socket?

Share this post


Link to post
Share on other sites
2 hours ago, P.C.M said:

Hi, 

Just catching up. Great to see you getting a bit done the trackwork is looking great.

 

Cheers Peter.

Thanks Pete,

 

The more I do on my layout, the more I appreciate what you’ve achieved!

 

Andy

  • Thanks 1

Share this post


Link to post
Share on other sites
17 minutes ago, Titanius Anglesmith said:

 

I considered doing the interlocking electronically via an Arduino (just mimicking the locking of a lever frame), but the inductive point motor loads still require an interface of some sort, rendering the idea a bit pointless. 

 

I haven’t looked at the pin-out of the relays yet. I wonder whether it would be practical to plug them in to a common IC socket?

There are a lot simpler ways to control a model railway, but I’d like the interlocking to be ‘right’. I have at least managed to let common sense allow that right can be eventually and not immediately!

 

I’ve taken a couple of pictures of the relay, including loosely placed in vero board. There may be a IC socket in that size. C20B50A0-B71D-40D2-8899-FEDB40978B96.jpeg.254065dfe33735f149402a8d9dc5c410.jpeg

 

D9203B35-8076-4118-AA79-9D22149E6DF0.jpeg.5139eab18562f8b914eb58737f3ed12b.jpeg

 

C2D2563F-CD20-4870-8E18-2F2FCA7DFD4C.jpeg.4c3a3d8a0b90ececc87a30522a76f6f2.jpeg

  • Informative/Useful 1

Share this post


Link to post
Share on other sites
Posted (edited)

Unfortunately, standard DIL sockets are ~7.5mm (two spare tracks between) not ~5mm.  Veroboard (at 90 degrees to your photo) is your best bet.

Paul.

P.S. I drill out veroboard holes to 1.5mm to get 16/0.20 wires in. 

Edited by 5BarVT
Add post script.
  • Informative/Useful 3

Share this post


Link to post
Share on other sites
On 07/07/2019 at 20:42, Titanius Anglesmith said:

 

Thank you for the info, you’re a star!

Quite a dim star as it turns out. The stock code is 515-587....

  • Informative/Useful 1

Share this post


Link to post
Share on other sites
Posted (edited)
18 hours ago, 5BarVT said:

Unfortunately, standard DIL sockets are ~7.5mm (two spare tracks between) not ~5mm.  Veroboard (at 90 degrees to your photo) is your best bet.

Paul.

P.S. I drill out veroboard holes to 1.5mm to get 16/0.20 wires in. 

 

How about two single-row versions of something like this? (I did find the single row version, then lost it...)

https://uk.rs-online.com/mobile/p/pcb-sockets/7022959/

 

Edit: ok, the pin socket size on that example is wrong too, but you get the gist....

 

Edit 2:  Maybe I’m overthinking it? For me I definitely think veroboard is a good idea, I’m just wondering how much of a pig it will be to swap out a relay when one fails? 

 

1 hour ago, 61656 said:

Quite a dim star as it turns out. The stock code is 515-587....

 

:lol: noted, thanks

Edited by Titanius Anglesmith

Share this post


Link to post
Share on other sites
21 hours ago, Titanius Anglesmith said:

 

How about two single-row versions of something like this? (I did find the single row version, then lost it...)

https://uk.rs-online.com/mobile/p/pcb-sockets/7022959/

 

I was thinking something like that, but I think the pin size is too small on the relay. 

 

I’m also concerned, even with the vero-board solution, about the current capacity. Potentially I could reinforce the copper with either solder or some wire, but it seems like quite a lot of work for 40 point ends!

 

Having seen how frail the relay pins are, it’s clear that I’ll need to do something to protect them. 

Share this post


Link to post
Share on other sites

How much can you modularise your circuits (Geographical!!!)?  OK I'm doing it with electronics but the principle is the same.  Here is (blurred 'cos of blowing up from a larger photo) one of my point drive units which I connect together with 0.1" pcb connectors.  I then build one spare of each type to allow swift swap out.  May be more difficult for the actual interlocking part, but I'm sure you could do the 'loc' circuits that way. 

PDU.jpg.6cd3cd6b8ac16974341e641b8a856a3e.jpg

I also use stripboard for my traction current distribution to individual track circuits and I did sums on max stripboard current for that.  I think it might have been about 1A which I can see might worry you for CDU peak current.  However, there is a thermal lag component that you can use - 1A continuous, but for the duration of point drive pulses you will get away with more.  I don't think the volts drop will be significant (add more 'batteries' if so!!!).

Paul.

Share this post


Link to post
Share on other sites

Just found your topic, Andy, and I'll be following now.

 

I think that you and Paul should both buy me a drink...

 

FIMechE FIEAust

  • Like 1

Share this post


Link to post
Share on other sites

Done.

You fund my air fare and I'll fund my accommodation and the drinks!

Paul.

  • Funny 1

Share this post


Link to post
Share on other sites
39 minutes ago, 5BarVT said:

How much can you modularise your circuits (Geographical!!!)?  OK I'm doing it with electronics but the principle is the same.  Here is (blurred 'cos of blowing up from a larger photo) one of my point drive units which I connect together with 0.1" pcb connectors.  I then build one spare of each type to allow swift swap out.  May be more difficult for the actual interlocking part, but I'm sure you could do the 'loc' circuits that way. 

https://www.rmweb.co.uk/community/uploads/monthly_2019_07/PDU.jpg.6cd3cd6b8ac16974341e641b8a856a3e.jpg

I also use stripboard for my traction current distribution to individual track circuits and I did sums on max stripboard current for that.  I think it might have been about 1A which I can see might worry you for CDU peak current.  However, there is a thermal lag component that you can use - 1A continuous, but for the duration of point drive pulses you will get away with more.  I don't think the volts drop will be significant (add more 'batteries' if so!!!).

Paul.

On the original incarnation I started drawing up the route locking and aspect level circuits. I soon realised with some routes having 15 or more opposing routes that a route relay interlocking was impractical, without using BR930 series relays! 

 

In the short term I’ve decided to wire the points with motors and relays for frog polarity selection (the SEEPs just aren’t reliable enough), but to stop at that. Eventually I think some sort of geographic circuitry might be possible, but DPDT relays (or 2F2B as I think of them) are very limiting. You can get 4PDT relays for under £5, but I think the sheer volume of relays will be prohibitive. That said, even a computer based interlocking needs to drive points and so having local point contactor relays will always be required, at least for solenoid machines.

 

I’ll need to give it some real thought when I fit signals, but the cost of those is sufficient to put it out of mind for the moment!

  • Like 1
  • Informative/Useful 1

Share this post


Link to post
Share on other sites

Suitably inspired by browsing the forum earlier, I found a couple of hours this evening to get some more track down. I think that’s 26 of the 32 ends required for the west end approach (but who’s counting?).

 

I have found taking pictures before the glue goes off really helpful in sorting out the alignment. There are still a couple of areas where I need to use a combination of soldering iron and cutting disk to get it ‘just so’. Most of the angles seem to be neatly flowing through the junction now. There are a couple of places I’ve spent time with, before realising it’s just a shiny soldered joint that misleads the eye!

 

I realised as well tonight that I’ll need a wide to gauge trap point from the centre road. I can’t imagine how to make that work in 00, so maybe there’ll be a compromise somewhere. Many years ago I fitted the rodding to a real one and it was hard enough in 1:1 scale!

 

I also realised I have now got enough track down for a full length trans-pennine rake, so a 45 brings in a typical Heaton lash-up for a relief service. No ETH, so the first class passengers will be sweltering in the air-con at the back!

 

0C543CAC-8505-4D55-9B58-2C5FD73427D8.jpeg.969fab2f4e9a4b3779e87c1c5ec26fad.jpeg

 

C830E72F-C985-4D9B-AF37-F0CD125C2BC0.jpeg.37ac1c14332894f573a7fb9fadfdecd3.jpeg

 

BA00E84B-C196-442C-A283-C2E7776536F4.jpeg.00fbd01b00066827d07af5c89d189fe9.jpeg

 

9B4C29F8-1DAE-49BB-9777-DA690FCBC1CD.jpeg.de39914d9be8cc6f483fea4564e14b66.jpeg

 

5AB1A271-CBB5-4972-9C9D-9A99EFEF8712.jpeg.8b6e4a24e6a94a525bc9931c5c26c278.jpeg

 

 

 

 

  • Like 8

Share this post


Link to post
Share on other sites

It’s fair to say that I have found a bit of passion for track laying at the moment, which is odd given how long I put it off for!

 

I had started with the parallel double slips, as all the geometry seemed to take its cue from there. I worked west first, for what I thought was no reason at all until I started to work east. Here it gets a little more complicated. I had planned to get the west junction and terminal platforms laid first; then, before doing the east (or south) end, I’d crack on with a bit of wiring. That way I’d be able to mix jobs up to avoid the monotony of soldering 700 relay terminals, plus be able to do a bit of carriage shunting with the 25 from time to time.

 

Tonight I realised that the terminal platforms all need to align with the angle taken by the outside through platform (number 2). This in turn takes its alignment from the Down Through. Which aligns to the Up Through, which aligns to Platform 1, the inner platform. The curve for platform 1 is tricky, as it has to align with the South Junction (at the east end - are you keeping up at the back?). 

 

Previously, in a 4m wide room, the south junction had acres of space, not so this time. Plus, the angle of the junction has a big bearing on the radius of the track through platform 1. Now, deep at the back of my brain is unwritten rule 46B: curves will be no tighter than 900mm if visible (57A refers to hidden tracks). What does that mean for platform 3?

 

Suddenly I realise I’ve laid 8 consecutive points in a perfect straight (but not perpendicular to the wall) line, but I have no idea if I need to take down a 300 year old chimney breast to get a curve in to the fiddle yard to allow anything to run on them. Time then to unplug the soldering iron and open a bottle of beer (there’s a brewery within 300m of my house, it’s a wonder I get anything done).

 

I wouldn’t say I have an immaculate formula-one pit stop style of railway room, this is a hobby after all. I do have a rule about ensuring there’s nothing sharp on the floor though. That said, you’d be surprised at the distance you can travel when you stumble backwards over the case for a power drill, and it will take a lot more speed and practice to get the beer all back in the glass on the downward trajectory. At this point I considered that waggling 900mm of steel tape around trying to work out the radius just wasn’t cutting the mustard.

 

Having reassured the other habitants of the building that I hadn’t in actual fact dropped the Albert Hall through the Crystal Palace, I channeled my inner Heath Robinson. Here we see a length of batten screwed temporarily to the baseboard top, with screws at various radii (high school Latin wasn’t wasted on me) from the centre of the track. I have no idea if radius is measured from the centre of the track, but who’s going to argue?

 

1CE7F3F1-C701-49CD-B195-E4274E6E1DB9.jpeg.ee20599e71b4b86a628905f572db3eca.jpeg 

 

This allows, using some string and a pencil, a nice line to be drawn to show the radii. You can just about see the results on the next picture. The peak and the train (in the first photo) represent a 6 coach Holyhead - Euston service which will swap from diesel to electric power. The diesel has to be able to leave enough space for the electric to back on, whilst ideally having all carriages inside the platform loop. I think I’ll be able to get 7-coach sets in the fiddle yard (I bet you can’t wait for the maths), in which case one coach will have to stand foul at the back, but only of the Up Through and not of the access to Platform 3. Platform 2 should be able to handle a long holiday excursion set, represented by 8 coaches. 8 coach sets will require a small extension of the fiddle yard through the wall into the siting room; to avoid any memos from the domestic fraternity I’ll probably only park 1st class in front of the TV.

 

5168432F-D75D-4B12-807E-D28D1421A52E.jpeg.5206da7a54554c7d92c6041ca1f66caf.jpeg

 

All this means that the south junction will be on a slightly better than 900mm radius through platform 1, leaving a bit of space for Christleton junction power box alongside the electric stabling point.

 

6B3F978F-F791-43F9-8C4F-7479A939933B.jpeg.65bfe72979799fb7602f9b5ad4fa7cda.jpeg

 

Which bring us nicely back to where I started, with a view along the first segment of the track for platform 1! I should be able to get platform 3 down in the blink of an eye.

 

C0D14F80-8A97-4724-B921-DE91C2E2CCF4.jpeg.0832d60bf6bc63c0298db8599bc9067b.jpeg

  • Like 8
  • Craftsmanship/clever 1

Share this post


Link to post
Share on other sites

Looks excellent my friend, a big improvement on Christleton Mk1.  Can't wait for my first chance to inspect and offer 'constructive criticism'.

 

 

  • Like 1

Share this post


Link to post
Share on other sites

The first of the curved tracks for the Up through platform (1) has been laid up to and including the Y point that provides the connection to the electric stabling. Following my previous update I’ve spent a few evenings pondering how the layout will look and how I will operate it. There were a few niggles that I wasn’t fully happy with:

 

1. The south junction encroached a bit too far towards the fiddle yard; getting the pointwork in to allow reasonable train storage would be tricky, unless I really fancied building a scissors crossover and fan of points on a 500mm radius curve.

2. The 4 through roads look like a little too much railway for the space available.

3. There weren’t enough roads for turning freight trains around.

4. The loco stabling was possibly too limited.

5. The junction pointwork I’ve already laid looks great and isn’t moving!

 

I realised that I was perhaps trying to create a replica of the previous layout, rather than the model I’d first conceived. Time to get out the trusty pencil and paper and work a few things out. I started by working what trains I plan to run, and what will happen to them on the layout. The previous incarnation never got further than 4 fiddle yard roads, so I never really got into a lot of the services that could be operated.

 

Realistically (this is a word often used where ‘hopefully’ would be more appropriate) I can get 14 fiddle yard roads. 5 terminal roads for each direction and 4 through roads. 

 

1. Euston - Holyhead intercity. Change of traction in platform 1 or 2. Aircon Mk2s, class 87 changes for 47/4.

2. Trans-pennine Holyhead - Newcastle. Reverses in terminal platform. PV Mk2s, class 45 changes for 45.

3. Trans-pennine Scarborough - Llandudno. Reverses in terminal platform. Mk1 / PV Mk2 mix. 47/4 changes for 47/4.

4. Cardiff to Holyhead. Through service in platform 1 or 2. Mk1s Class 33 or 47/4.

5. Christleton to Holyhead. Terminates, usually connecting with intercity cross country service. Coaches stabled between services. Mk1 / PV Mk2. 47/4 , 31/4, D200. (This service replaces the ‘real’ Crewe - Holyhead, as it connects with the West Coast cross country services using the Christleton loop between Crewe and Warrington).

6. West Coast cross-country service. South West to Scotland through service. Usually connects with the Holyhead service.  Aircon  / PV Mk2 mix. Class 85/86. 

7 & 8. Parcels / Newspapers / Vans. Various options including Midlands - Holyhead, Manchester / Scotland - Holyhead, with coaches shunted on and off at Christleton. Mk1 BG / GUV mix. Class 25, 31, 47, 85. Requires coach and loco stabling.

9. EMU Crewe - Warrington. Through service, connects with DMU. 305 or 310 (subject to availability!)

10. DMU Christleton - Llandudno, Manchester - Holyhead etc. Class 101 / Class 108. Requires stabling, possibly in platform.

11. Freightliner Bescot - Holyhead. ‘Long’ set, through service on through road. Class 45 / 47.

12. Freightliner Warrington - Holyhead. ‘Short’ set, needs to reverse, run round or loco swap. Class 25 / 31 / 37.

13. Speedlink trip service, Warrington - Holyhead. Needs to run round. Class 25 / 31. Freight siding required for VAA / HAA. TTA drop off for loco fuel. May include nuclear flask.

14. Merry Go Round Point of Ayr to Fiddlers Ferry. Needs to reverse, possible loco swap from Class 56 to 2 x20.

15. Oil tanks. Stanlow - North Wales. Needs to reverse or loco swap. Class 25 / 31 / 47.

16. Ballast. Penmaenmawr to Bescot or Warrinton. Uses through road or reverses. Class 25 / 31 / 97.

 

The numerate amongst you will realise that 16 trains won’t fit in 14 sidings, however 3 and 5 are probably the same stock, as are 11 and 12. I think there will usually be at least 1 train ‘out front’, leaving me a whole extra road spare, for which there are any number of options!

 

A review of the operating outline tells me quite a bit about the layout requirements:

- 4 of the passenger services require loco stabling, probably 4 diesels and 2 electrics if you include parcels work.

- 3 or 4 of the freight services need loco stabling, probably 3 diesels. This means my little shed needs a bit more space for stabling.

- I need to be able to store a rake of mk1s, a handful of parcels vans and a DMU or 2. The DMU and parcels can probably share.

- A lot of freight reverses. I can currently only run round on the main trough lines, which doesn’t feel right. I can’t imagine running round a rake of HEAs with a class 86 passing on an express. A class 25 running round a short speedlink rake would look lovely, but not at a platform. I also can’t imagine BR would have been happy with swapping locos on a 4 truck speedlink service due to lack of a run round.

- Only 1 or 2 trains use the through roads. A single bi-directional through road would seem to fit the bill - the LNWR didn’t generally use a single through road, but then I appear to have less space available than them! Having a single through road shortens the south junction by a point length, which helps solve the fiddle yard problem.

 

I think I can have 9 tracks across the main layout. This was originally going to be 4 through, 3 terminal platforms and 2 sidings. I think it would be better to have 3 through, 3 terminal, 3 sidings (which need to include loco run round). Luckily this doesn’t change the junction track already laid.

 

This is a long winded way of saying I’m thinking of making a couple of minor layout changes!

  • Like 4

Share this post


Link to post
Share on other sites

Always good to batten down the operational requirements early on!!

Like your logic, I’m currently limited to 4 yard roads and one as storage.

Paul.

  • Agree 1

Share this post


Link to post
Share on other sites

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...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.