Nether Madder and Green Soudley Rly

[RedGemAlchemist]

2 hours ago, Martin S-C said:

Now I need to know what a pervert frame is. I think I need to install one!

[Martin S-C]

The very last signal box diagram, Green Soudley. The eagle-eyed among you may have noticed that this is an almost direct steal from John Ahern's Madderport. I added an extra industrial siding and converted his harbour line to a canal wharf, other than that I think its a complete copy.

Once again, comments welcome on the placement of signals and the lever numbers.

[Annie]

I do like your signal box diagrams Martin and if this one is a direct steal (almost) from Madderport then you couldn't ask for a better source of inspiration.  The signalling looks to be appropriately minimalist and I can't see anything wrong with it.

[ianathompson]

The main problem with your diagram, and some of the others, is that you have a large number of "worked" points.

Worked points are those that are controlled from the lever frame in the signalbox and therefore appear on a box diagram.

In reality most of the yard points would be thrown by hand from a lever alongside them and would not be numbered on the box diagram.

In fact it is unlikely that they would even appear on it, as the signalman was only interested in equipment over which he had direct control.

 

Obviously these controls need to appear somewhere on a model railway control panel diagram if they are to be remotely operated by motors.

It might be better to have a smaller separate panel for hand thrown points and to place the "worked" points together to form a frame.

 

Your box diagram shows 14 levers and no spares.

Points 7,8,10,12,13 would be hand thrown and lever 11 would only work one half of the crossover.

The other half, by the tinplate works would be a hand point.

This would give you a ten lever frame and need a smaller panel for the six hand operated points.

 

I assume that your layout was influenced by Midland practice as it is using economic point locks, although these were used on minor railways as well.

Most railways used a separate lever to work the facing point lock and this was numbered into the sequence.

 

Ian T

 

 

 

 

[Titanius Anglesmith]

Both ends of 11 crossover need to be worked because they trap the siding. Likewise 13 crossover. 

 

I agree that the worked yard yard points are unusual, but I can understand the justification on a model. On in my own layout I’ve allocated yard points to otherwise “spare” levers. The levers are white and labelled as “SPARE (HW1)” (where HW stands for hand-worked). I only have a few yard points though so there’s not an unusually high number of white levers. 

 

With regards to shunt discs (or lack thereof), I believe it was a requirement on the “big railways” that any exit from a non-passenger line onto the running line had to be signalled. I don’t know if that extended to light railways though, probably not. In that regard Green Soudley is fine, assuming that trains leaving the tinplate siding do so via 13 crossover and 14 signal. 

[Nearholmer]

“I believe it was a requirement on the “big railways” that any exit from a non-passenger line onto the running line had to be signalled.”

 

I’m fairly certain that there was no BoT requirement to that affect, but I will check later, and practice seems to have varied hugely, both in the extent and form of such signals.
 

About the one generalisation that it is probably fair to make is that nearly all shunt, and departure from ‘dark territory’, moves that were made on a fairly frequent basis in a reasonably busy area were given fixed signals, for efficiency as much as safety. As is well known, some railways, the GWR and SER to my knowledge, were more generous with such signals than most. The LBSCR seems to have vacillated a bit, odd locations having lunatic quantities of signals, before settling down. Others I leave to local knowledge.

[Martin S-C]

Thanks all. Yes, the discussion about hand worked yard points has been raised a number of times. I'm going with all points and signals worked on the one model railway control panel per station - some of the levers on which represent the signal box and some represent yard points but I don't distinguish. When in full cry the layout will most likely be worked by a team of more-or-less casual visitors and simplicity/ease of use of the control panels outranks historical accuracy on my list of modelling priorities. Its not ideal from an accuracy point of view but I need to make compromises.

The EFPLs were suggested by another expert here (sorry, I forget whom) and their origin wasn't specified but I am not intentionally using Midland practice. I'm using the fictional practice of the NM&GSR which takes its inspiration from a wide range of sources

My model is not intended to be a light railway but a fictional medium sized independent line. The pre-joint management Severn & Wye comes to mind in terms of likely track mileage. It does however have a number of obscure practices that I am sure would give a real BoT inspector fits. I've retained stop boards and hand signals from the bobby on the branch and disc and crossbar signals at the lesser stations along the main line. Only the two termini have "proper" semaphore signals as these I assume would be where any available money would be spent first. The Witts End branch is classed as a tramway or at the very least a light railway though it doesn't protect the motion of its locomotives with skirts. As I said... a bit of a beast as historical accuracy goes (*) though I do like to try and get most things right as regards British railway pre-grouping practice in most areas.

 

Quote

With regards to shunt discs (or lack thereof), I believe it was a requirement on the “big railways” that any exit from a non-passenger line onto the running line had to be signalled. I don’t know if that extended to light railways though, probably not. In that regard Green Soudley is fine, assuming that trains leaving the tinplate siding do so via 13 crossover and 14 signal.

 

That's correct, goods workings departing that side of the station exit via 14. However shunt moves that require a loco to run around via the tinplate siding would use crossover 11 but in such cases signals 4 and 6 would be at danger. Question - if a goods train arrived via 13 and 12 to stop on the canal wharf and the loco then ran around via 11 and 5 to halt before the signal box in order to propel wagons onto the tinplate siding I presume that move would require a signal. If so would it be before the platform end of the crossover 11 so that 5 and the running line are protected? Is that right?

 

 

(*) As a supporting example I understand the Wantage tramway habitually didn't enclose the motion on its roadside locomotives in later years.

Edited October 15, 2019 by Martin S-C

[Nick C]

On 15/10/2019 at 11:56, Martin S-C said:

 Question - if a goods train arrived via 13 and 12 to stop on the canal wharf and the loco then ran around via 11 and 5 to halt before the signal box in order to propel wagons onto the tinplate siding I presume that move would require a signal. If so would it be before the platform end of the crossover 11 so that 5 and the running line are protected? Is that right?

 

I think it would be on the wharf, at the toe end of 11b, to protect the whole crossover. But it would also be a shunt signal, and you've said you're not using them. I think that generally a freight at a small station like that would arrive in the platform road anyway, then run round using the loop and shunt.

[Nick C]

On 13/10/2019 at 12:57, Martin S-C said:

 

 

I'd put 24 and 25 on one lever. There ought really to be a trap between the slip and 9 to protect the running line from the loco yard, and also one on the timber siding. I'd be tempted to put the homes back by the box, and add an outer home where they are now, and an advanced starter under the word down, otherwise you've got to use the section signals for moves into the works. If you've got room, the advanced could even go outside of 12, but you want an overlap between the outer home and the advanced starter.

[Martin S-C]

Thanks for the comments Nick. I haven't used trap points anywhere on the layout, so using them here would look odd. I had a long think about whether I wanted to include them, even dummy ones but in the end decided against it. How does this look?

 

The other couple of items I have been working on over the last day or two is small bits n bobs shelves beside the control panels. As these are all temporary I haven't gone OTT on design or looks. Here is the Catspaw and Exchange Sidings panels and the Colliery, the mimic diagram for which I don't think I've shown here before.

 

 

 

If you squint a bit the colliery panel just reveals the blue lamps on the DCC Concepts alpha switch buttons. 7 and 10 are reversed. 12 is spare. I laid the paper printed mimic diagram over the buttons and then laid clear adhesive plastic sheet over that to keep dirt off. The blue lamps shine through and leave a smooth wipe-able surface which is useful though the whole feel of the thing is too modern. This is all a stop-gap until I can source some proper levers.

Now that many points are powered and operated from the panels testing of train movements has become a real joy and I'm fairly happy with the arrangements of what I've determined are the normal and reversed positions for the points. A few things are arising operationally and being changed but that's why we play trains test things.

[chuffinghell]

I like your panel diagrams, particularly the boarders

 

The shelving is a good idea too (an idea worth pinching )

[Harlequin]

Hi Martin, To my eyes the NM panel looks a bit cramped. Could you spread it across two sheets (using about 1.5)?

 

That would allow room to breathe either side and room for the line to Snarling Jcn to be seen more clearly as the only line leaving the area.

 

Also, the border lines are visually identical to the track lines (same weight and colour) which makes the diagram a bit more difficult to read at a glance. Maybe you could de-emphasise them by reducing size , including the curlicues, or changing colour? I’m not sure what would work best.

 

P.S. I don’t think you need FPLs in loco release crossovers (e.g. 23) because passenger vehicles should never traverse them.

Edited October 22, 2019 by Harlequin

[Nearholmer]

"I don’t think you need FPLs in loco release crossovers (e.g. 23) because passenger vehicles should never traverse them"

 

This has been much discussed on RMWeb, and it appears that in many, but by no means all, places they did have FPLs, to prevent trouble if a train was propelled right up to the stops prior to departure. If they are absent, you theoretically need a "no passenger trains beyond here" board clear of the crossover, not that you could read it at 1:76 scale, and possibly railings along the platform edge, although they weren't all that common.

[Harlequin]

34 minutes ago, Nearholmer said:

"I don’t think you need FPLs in loco release crossovers (e.g. 23) because passenger vehicles should never traverse them"

 

This has been much discussed on RMWeb, and it appears that in many, but by no means all, places they did have FPLs, to prevent trouble if a train was propelled right up to the stops prior to departure. If they are absent, you theoretically need a "no passenger trains beyond here" board clear of the crossover, not that you could read it at 1:76 scale, and possibly railings along the platform edge, although they weren't all that common.

OK, thanks. I was emboldened to suggest that from the evidence of the GWR terminus trackplans I've been looking at recently but I should have kept my trap shut.

 

(It's possibly academic in Martin's case anyway.)

 

[Nearholmer]

Oh, no.

 

Traps should be maintained open to protect the main line. 

[Martin S-C]

The FPLs in my case are just graphics on the diagram but I'll keep mine due to what Nearholmer suggests.

I agree the NM diagram is too cramped. I will see about spreading it out a bit and giving it more room to breathe. That way the borders may become less invasive and confusable with the tracks. I could also do the borders and curly corners in another colour, brown maybe. It had struck me that I could colour code the diagram borders by original company. NM and SJ would then become bottle green. PB and GS Prussian blue and the three branch stations wine red. The exchange sidings would be brown and colliery would remain black.

[Denbridge]

Hi Martin. I trust you are well and have managed to at least push your depression into the background. As a fellow sufferer, you have my total sympathy and understanding.

My main reason for writing is to enquire whether you have had chance to try out the powerbase gear you mentioned?

I'm considering using it and am wondering how effective it is.

[Martin S-C]

Hi, I am over the worst of the depression now and things are much better, thank you for asking.

I am having a testing session at the moment, first results are disappointing but that I feel is due to the loco I used as a guinea pig - in retrospect a bad choice.

I am using a 1 in 32 grade on a 4 ft radius curve as a test and a DJM Beattie WT as the loco. The problem is the loco is so tiny the DCC Concepts neo magnets won't fit under it. My fear is this is going to be true for a lot of the smaller models. I have therefore used some of my own stock of different sized neo's that I got on e-Bay. These are wider but thinner and not so strong (12x7x1mm vs 12x5x2.5mm). I have found room to glue four of them under the baseplate of the loco but clearance over point switch blades is a problem. I'd hoped to mount these in pairs making a double magnet 12x7x2 but even this doesn't give enough clearance, so it was down to 4 fittings of 12x7x1. Lifting the loco off the track I can definitely detect the extra pull but as far as my tests so far have determined this extra adhesion hasn't increased haulage power at all. There are some pick-up problems with the loco plus, perhaps, as its so light or the motor is low powered the effect isn't apparent. The DCC Concepts thicker magnets are probably more powerful than the thinner ones. I have set this one aside and am starting to test with a Hornby J15 0-6-0.

 

Loco: DJ Models LSWR Beattie Well Tank: 100 grams. Train weight unassisted up 1 in 32 grade: 205 grams. Train assisted with 4x magnets up same grade ... 205 grams.

 

Loco: Hornby LNER J15 0-6-0: 235 grams. Train weight unassisted up 1 in 32 grade: 510 grams. Train assisted with magnets ... still to be tested.

 

I can see that the J15 has plenty of space for 2 of the DCC Concepts magnets so my initial reaction is this system is not aimed at small tank locos. I will need to test more of them - perhaps the well tank was an unfortunate example.

 

 

Edited October 23, 2019 by Martin S-C
added notations to first image

[BWsTrains]

Martin,

 

Two points here.

 

The magnet poles are at both faces in what are highly compressed bar magnets with N/S poles within few mm of each other. Just look how they naturally stack - NSNSNS - to see what I mean. As such their magnetic fields are highly compressed in the plane of the magnets so distance above the attractant is crucial. In this image I've taken the field of a "standard" magnet and compressed the NS magnet axis 10 fold. In a Neo the compression is even greater and the field even flatter.

Hence moving them much higher in the loco will greatly reduce the attractive force I'm afraid. This is why positioning of Kadee trip pins and magnets is a fine art.

 

Secondly what is the substrate you're using to attract to? PECO track is non magnetic and any plate under the track further extends the separation between magnet and target.

 

Colin

 

 

Edited October 22, 2019 by BWsTrains
additions and corrections for clarity

[Martin S-C]

Colin, its the DCC Concepts Power Base, so its ferric steel, about 0.3 to 0.5mm thick. It slides under the sleepers. I can definitely feel the magnetic attraction when lifting the loco wheels from the rails.

I am wondering if the steel strips could be cut in half lengthways and laid atop the sleepers between the rails. I have plans to lay an ash ballast effect, burying the sleepers so that's a possibility.

[DonB]

Trying to think "outside the box", It occurred to me that an alternative to the suggested thin steel sheet at track level for the magnets to be attracted to, you could consider mixing the commercial grade of Iron filings (as used in BWsTrains demonstration picture) mixed in your ballast and WELL GLUED in place. The magnets ideally need to have (all, if more than one) of their poles aimed at the ballast and as near as possible to it.

[St Enodoc]

3 hours ago, BWsTrains said:

Martin,

 

Two points here.

 

The magnet poles are at both faces in what are highly compressed bar magnets with N/S poles within few mm of each other. Just look how they naturally stack - NSNSNS - to see what I mean. As such their magnetic fields are highly compressed in the plane of the magnets so distance above the attractant is crucial. In this image I've taken the field of a "standard" magnet and compressed the NS magnet axis 10 fold. In a Neo the compression is even greater and the field even flatter.

Hence moving them much higher in the loco will greatly reduce the attractive force I'm afraid. This is why positioning of Kadee trip pins and magnets is a fine art.

 

Secondly what is the substrate you're using to attract to? PECO track is non magnetic and any plate under the track further extends the separation between magnet and target.

 

Colin

 

 

Martin, I think it would be worth trying some thin magnets with the poles perpendicular to the faces. Small disc magnets are available:

 

https://www.frenergy.com.au/rare-earth-discs.html?p=4

 

I've used magnets for other applications - e.g. loco lamps, as you know - from this firm (usual disclaimer) but I'm sure there will be suppliers of similar items in the UK.

[BWsTrains]

Thx St Enodoc,

 

you raise good points, there's quite some range of powers, sizes and polarisation available and buying from a "serious" supplier gives you ability to customise to your needs but at a price.

 

Buying elsewhere off the Net it's a case of Caveat Emptor! but if you read the Specs carefully I've got very good product from Banggood (usual disclaimer) that's fully met my needs for decoupling and home fixing jobs in general.

 

Colin

[Nearholmer]

Martin

 

I'm still a bit foxed by the haulage problems that you are experiencing, and wonder if an underlying issue might be the rolling resistance of the wagons.

 

it bothers me, because it seems plain morally wrong that a model loco can't do the work of its real-life peer, especially a newly made model loco, because if it can't, that feels like a design failure on the part of Bachmann, Hornby, or whoever.

 

Have you tried measuring the drawbar pull needed to shift the wagons on level, tangent track? Someone on here (it might have been you!) was using some sort of tiny spring balance to measure drawbar pull, accepting that it wouldn't give very accurate measurements, but would allow meaningful before vs after (lubrication, bearing de-crudding etc) comparisons.

 

As a rule of thumb, I think that a loco ought to be able to shift on level, tangent track a load that requires drawbar pull of about one fifth of its mass (others feel free to contribute/contradict), and the effect of gradient is calculable, given knowledge of a load of variables that you probably can't access! But, if you can measure pull on level, tangent track you could also measure it on curving gradient.

 

Kevin

 

 

Edited October 23, 2019 by Nearholmer

[Harlequin]

On another thread (Wright writes?) there has been talk recently of the profile of the wheel treads being important for traction. People have had wheels re-treaded or re-ground and got improved pulling power.

 

That makes sense intuitively: If the tread is angled at all then it will only have a tiny contact patch with the rail but if it's flat then it will make much better contact and the coeff. of friction should be consequently greater.

 

Martin: How do you feel about gluing magnets to your locos and iron plates to your track and having to disguise both???