Interlocking signals & points via dcc and/or pc for a lever frame

[The Great Bear]

Hello

 

I want to use a lever frame to control my signals and points but want to interlock them. For my branch terminus I've used DCC concepts Cobalt-S levers and managed just to achieve some interlocking and things like facing point locks through the wiring of the pcb. (The locking doesn't lock the levers from moving but means they don't do anything if locked etc.) This was hard enough for me to do for a small lever frame - see my layout thread. My mainline box is going to have 40ish levers.

 

Is there a way of using the levers, either the Cobalt ones or something similar (switches arranged like a lever frame) and achieve the interlocking through use of computer software?. I don't want to control trains via pc nor have a modern control panel. I only want to use the software to do the interlocking bit as that's going to be a lot easier to set up and test than a whole lot of wiring spaghetti.

 

At the moment my points on the mainlne are DCC controlled and if necessary semaphore signals could be done same way.

 

I haven't yet got my signalling plan sorted that will be soon but I may need to buy more accessory decoders if I can go the electronic/pc route, the cost isn't really an issue here. But if can't do this then I may revert the points in the scenic section to analogue control as otherwise there's no advantage in using dcc for this. Hence why I am asking the question now. (Hope this bit is clear.)

 

Any help or suggestions as always would be much appreciated. If my aim or query is unclear please don't hesitate to ask.

 

Jon

[The Great Bear]

My DCC system is NCE Powercab, by the way, if that makes a difference

[raymw]

I suggest you try http://modratec.com/

 

Download the free sigscribe software, and set up your points/signals there, and test out your interlocking.

 

if you want simpler pseudo interlocking, then diode logic ( or ttl chips) will most likely do it for you

 

Best wishes,

 

Ray

[The Great Bear]

Ok, thanks - the diode logic chips something to investigate, albeit I'm not that good at electronics or electrical stuff. I've heard about modratec frames - very very nice but I don't want to invest so much in something so bespoke to my current layout, I think but I'll have a look at their software.

 

Software solution if I've understood RailRoad & Co website if it can be done wouldn't be cheap either and I'd have to upgrade my DCC system I think too.

 

Thanks

 

Jon

[beast66606]

The difficulty you will face is interfacing the hardware - the levers - and the software.

 

It's probably easier to implement the locking with relays.

[The Great Bear]

The difficulty you will face is interfacing the hardware - the levers - and the software.

 

It's probably easier to implement the locking with relays.

Thanks. A first flick through of RR&Co pdf suggests might be doable (although may well be wrong here) - section 14.10 on control panels/contact indicators, but to use this would require the semi-skimmed version of the software, upgrade to my dcc system and more accessory decoders: a sledgehammer to crack a nut - as you suggest.

 

So I might as well eschew digital control for the points that controlled by the lever frame in the scenic section - I'd have had to change accessory decoders in any case to ones that allow push button inputs - the Switch-its I have do, the Switch-8's don't. The latter have been a bit flakey anyway.

 

I'd better ready that signal diagram for the expert (especially Stationmaster's as it's Western) perusal.

 

Once again thanks

 

Jon

[beast66606]

I've used RR&Co to implement an NX panel with more or less everything the prototype would have, approach control, flank protection, even a SPAD - mechanical signalling is easier as it's all about locks and releases - but getting your hardware (the lever frame) to talk to the software, and more importantly getting the software to communicate back to the hardware will be a challenge - to say the least.

 

PS - I know a bit about signalling too - but proper lower quadrants were LNWR imho and so I leave Mike (SM) to deal with the Wessie stuff. (I can get away with saying that as he's on holiday!)

[SimonME46]

I did my initial control panel interlocking with relays. I've now redone it with a Modratec frame, and it is much more satisfying. I consider it very cheap for what it has achieved, and I've learned heaps about signalling.

[Dunsignalling]

Hello

 

I want to use a lever frame to control my signals and points but want to interlock them. For my branch terminus I've used DCC concepts Cobalt-S levers and managed just to achieve some interlocking and things like facing point locks through the wiring of the pcb. (The locking doesn't lock the levers from moving but means they don't do anything if locked etc.) This was hard enough for me to do for a small lever frame - see my layout thread. My mainline box is going to have 40ish levers.

 

Is there a way of using the levers, either the Cobalt ones or something similar (switches arranged like a lever frame) and achieve the interlocking through use of computer software?. I don't want to control trains via pc nor have a modern control panel. I only want to use the software to do the interlocking bit as that's going to be a lot easier to set up and test than a whole lot of wiring spaghetti.

 

At the moment my points on the mainlne are DCC controlled and if necessary semaphore signals could be done same way.

 

I haven't yet got my signalling plan sorted that will be soon but I may need to buy more accessory decoders if I can go the electronic/pc route, the cost isn't really an issue here. But if can't do this then I may revert the points in the scenic section to analogue control as otherwise there's no advantage in using dcc for this. Hence why I am asking the question now. (Hope this bit is clear.)

 

Any help or suggestions as always would be much appreciated. If my aim or query is unclear please don't hesitate to ask.

 

Jon

Jon

 

If you intend to use levers, I can't see any gain from DCC operation of points/signals - it surely just introduces another layer of technology.

 

For a reasonably conventional double-line station, few levers should need to be interlocked with more than two others. This should be achievable (albeit with a bit of trial and error) via unused contacts in the Cobalt levers if my reading of the manual is correct.

 

You have already done this on the branch line and you should be able to use similar methods for most of your main line station except where some part of the layout is more complex (e.g. direct access to down sidings from up main or vice versa).

 

Any levers needing extra interlocking can probably be dealt with using accessory contacts on point motors. A very few might need dedicated relays but, if you think of these as point motors with accessory contacts but no point, you may find the idea less daunting!

 

If you require physical locking/release on the levers, Modratec is more suitable (Cobalt levers would have to be modified quite substantially) but my understanding is that you don't need/want that.

 

John

[The Great Bear]

Jon

 

If you intend to use levers, I can't see any gain from DCC operation of points/signals - it surely just introduces another layer of technology.

 

For a reasonably conventional double-line station, few levers should need to be interlocked with more than two others. This should be achievable (albeit with a bit of trial and error) via unused contacts in the Cobalt levers if my reading of the manual is correct.

 

You have already done this on the branch line and you should be able to use similar methods for most of your main line station except where some part of the layout is more complex (e.g. direct access to down sidings from up main or vice versa).

 

Any levers needing extra interlocking can probably be dealt with using accessory contacts on point motors. A very few might need dedicated relays but, if you think of these as point motors with accessory contacts but no point, you may find the idea less daunting!

 

If you require physical locking/release on the levers, Modratec is more suitable (Cobalt levers would have to be modified quite substantially) but my understanding is that you don't need/want that.

 

Joh

Thanks, John - that's very helpful. I'll abandon the dcc control for the points and make the changes to the wiring; the dcc for the staging yard is great, the use of the macros in the Powercab so I won't change that.

 

All the best

 

Jon

[The Great Bear]

You have already done this on the branch line and you should be able to use similar methods for most of your main line station except where some part of the layout is more complex (e.g. direct access to down sidings from up main or vice versa).

Yep, this is an area of confusion to me regarding the signal diagram, as is the area of the double slip at the other end of the loop and the connection back to the down main: in particular whether every possible route has to be explicity signalled. In the prototype (Kidlington) which I largely follow it appears not from the Signalling Record Society diagram I have. What isn't available from SRS is the real locking table to see if the locking prevents those routes - or could it be they were so infrequently used every permutation doesn't have to be covered?

[Dunsignalling]

Yep, this is an area of confusion to me regarding the signal diagram, as is the area of the double slip at the other end of the loop and the connection back to the down main: in particular whether every possible route has to be explicity signalled. In the prototype (Kidlington) which I largely follow it appears not from the Signalling Record Society diagram I have. What isn't available from SRS is the real locking table to see if the locking prevents those routes - or could it be they were so infrequently used every permutation doesn't have to be covered?

In the example I suggested, I think it would be pretty much inevitable that conflicting routes could be set when all signals were at danger.

 

The function of the interlocking is (1) to make prevent signals leading into them being cleared and (2) to prevent such a route being set once signals had been cleared for another.

 

Let us assume the points from the up main to down yard (crossing the down main) are controlled by lever 12, and the the down home signal is lever 3. Assuming there are no facing point locks involved, when all levers are back in the frame, either lever can be moved before the locking comes into play. Then:

 

With 12 reversed (to access the yard), 3 cannot be cleared.

 

With 3 reversed (to pass a Down train), it will be impossible to reverse 12.

 

Thus, in locking-speak 12 reversed locks 3 normal, and 3 reversed locks 12 normal.

 

This is actually no more complicated than for a siding trailing off the down main where operation of the point (or GF release) would lock the home signal to prevent a movement approaching points set against it and clearing the signal would lock the point or GF release.

 

In most cases the 'yard' end would also incorporate a trap point to keep any runaway wagon away from the running line. This would be worked off the lever controlling the main-line point ("double-ended") in the same way a crossover operates.

 

All this applies whether the lines cross via a plain diamond or a slip, though obviously the latter will introduce further complication (extra signals and associated locking) to the mix because it will be possible to access the yard from either running line.

 

Space-starved modellers are often drawn to substituting a double slip for a single to add the benefit of a crossover between the running lines. This was quite rare at through stations on the real thing; they were more common at termini, where room was tight, and speeds low.

 

John

[LNERGE]

I did my initial control panel interlocking with relays. I've now redone it with a Modratec frame, and it is much more satisfying. I consider it very cheap for what it has achieved, and I've learned heaps about signalling.

My Ely North Junction has 40 levers with mechanical interlocking. I consumed around 3500 metres of 0.75 wire to make it work. My Ely Dock Junction has a 36 lever Westinghouse style L miniature frame with electric interlocking. This consumed at least 7500 metres of 0.75 wire. With mechanical locking if one lever locks another the converse is automatically provided, if lever 1 locks 2 then 2 will lock 1.

 

With the L frame every interlock has to be provided by cutting the circuit as required. If i cut the circuit to lever 2 on a circuit controller band to give 1 locks 2 i will also have to cut the circuit for lever 1 over a band on lever 2 to give 2 locks 1. In the simple case quoted that used nearly 4 metres of wire.

 

Locking levers mechanically will cut down any wiring between interlocking relays or keyboard work on the computer and will give the most realistic

result without a doubt. It is also a voyage of discovery. Sliding tappets, lifting tappets, tappets with sliders, duplex tappets, it's all out there to discover and loads more.

 

 

This photo is from the early days when i first had the frame, hence the in line joiner, wires with temporary insulation, and numbered test straps in shot.

[The Great Bear]

My Ely North Junction has 40 levers with mechanical interlocking. I consumed around 3500 metres of 0.75 wire to make it work. My Ely Dock Junction has a 36 lever Westinghouse style L miniature frame with electric interlocking. This consumed at least 7500 metres of 0.75 wire. With mechanical locking if one lever locks another the converse is automatically provided, if lever 1 locks 2 then 2 will lock 1.

 

With the L frame every interlock has to be provided by cutting the circuit as required. If i cut the circuit to lever 2 on a circuit controller band to give 1 locks 2 i will also have to cut the circuit for lever 1 over a band on lever 2 to give 2 locks 1. In the simple case quoted that used nearly 4 metres of wire.

 

Locking levers mechanically will cut down any wiring between interlocking relays or keyboard work on the computer and will give the most realistic

result without a doubt. It is also a voyage of discovery. Sliding tappets, lifting tappets, tappets with sliders, duplex tappets, it's all out there to discover and loads more.

 

ely dock junc 314.jpg

 

This photo is from the early days when i first had the frame, hence the in line joiner, wires with temporary insulation, and numbered test straps in shot.

 

Crikey that's impressive, but way way way beyond my skill or (more importantly) my patience! In addtion to this, one reason I am content with an electical locking solution is that if I use the DCC concepts levers and wiring looms it's recyclable should the layout change. The currently layout is, in some ways, to sate a desire to see trains running round and round after the little branch. I am not sure in the long term whether I will find this completley satisfying or not or want more.

 

All the best,

 

Jon

[The Great Bear]

In the example I suggested, I think it would be pretty much inevitable that conflicting routes could be set when all signals were at danger.

 

The function of the interlocking is (1) to make prevent signals leading into them being cleared and (2) to prevent such a route being set once signals had been cleared for another.

 

Let us assume the points from the up main to down yard (crossing the down main) are controlled by lever 12, and the the down home signal is lever 3. Assuming there are no facing point locks involved, when all levers are back in the frame, either lever can be moved before the locking comes into play. Then:

 

With 12 reversed (to access the yard), 3 cannot be cleared.

 

With 3 reversed (to pass a Down train), it will be impossible to reverse 12.

 

Thus, in locking-speak 12 reversed locks 3 normal, and 3 reversed locks 12 normal.

 

This is actually no more complicated than for a siding trailing off the down main where operation of the point (or GF release) would lock the home signal to prevent a movement approaching points set against it and clearing the signal would lock the point or GF release.

 

In most cases the 'yard' end would also incorporate a trap point to keep any runaway wagon away from the running line. This would be worked off the lever controlling the main-line point ("double-ended") in the same way a crossover operates.

 

All this applies whether the lines cross via a plain diamond or a slip, though obviously the latter will introduce further complication (extra signals and associated locking) to the mix because it will be possible to access the yard from either running line.

 

Space-starved modellers are often drawn to substituting a double slip for a single to add the benefit of a crossover between the running lines. This was quite rare at through stations on the real thing; they were more common at termini, where room was tight, and speeds low.

 

John

 

Thanks, well explained: bit more head sratching to follow. One of the double slips was there in the prototype, the other was a space saving bodge replacing two separate sets of points to get a long enough loop in.

 

Rather than confuse this post with one about signalling for my layout, I'll post separately on that in due course.

 

All the best,

 

Jon

 

[Suzie]

It sounds like just using a feedback module to interface your levers to your DCC system and using JMRI or similar will be the way to go. Completely reconfigurable if you want to change it at any time without having to touch a soldering iron.

 

JMRI software will probably be favourite because it can be run on an inexpensive (<£40) Raspberry Pi rather than needing a full PC or anything and the signal rules are already there to make a lot of the set up simple.

[BrianR]

I have done this recently with RR&Co (Gold):

 

http://www.rmweb.co.uk/community/index.php?/topic/70920-signalling-mechanical-and-electrical-interlocking/

 

JMRI seems a zero cost alternative if you are still looking at DCC. The software/DCC method gives flexibility for changes and needs no more than the standard DCC decoders - wiring is simple.

The advantage of the onscreen "levers" is that they don't work when locked. Mechanical switches/levers other than Modratec type can be moved even when supposedly locked unless you resort to complicated extras.

If you decide to use lever operated switches and non-DCC it may be worth looking at PicAxe chips, easily programmed in Basic and able to operate points and signal drivers directly or through transistors etc depending on power requirements.

[The Great Bear]

Given that the points are already wired for dcc I am going to give JMRI a go.

[Ravenser]

My DCC system is NCE Powercab, by the way, if that makes a difference

Probably the easiest way to mimic interlocking with a PowerCabis to use the macros , either with the signals controlled by accessory decoders or by spare switches on the point motors. Press the macro , and the whole route comes off with interlocked signalling...

 

But that doesn't involve using a lever frame so may not the an approach that suits you. 

[The Great Bear]

Probably the easiest way to mimic interlocking with a PowerCabis to use the macros , either with the signals controlled by accessory decoders or by spare switches on the point motors. Press the macro , and the whole route comes off with interlocked signalling...

 

But that doesn't involve using a lever frame so may not the an approach that suits you. 

 

Thanks. But the lever frame or something akin to it is an absolute must and not sure how this fits into this. I want to have to set the points, then be able to clear the route, not vice-versa.

 

If macros are a solution, I wonder whether the mini panel would be of use: I believe the mini-panel uses macros but I am not sure whether can be made conditional in any way, so signal is only pulled off if one of the other inputs is in certain state. If it could in many ways, that would be the ideal solution avoiding connection to pc. (Not that the pc is an issue, mostly have enough spare bits lying around to build a spare one.)

[The Great Bear]

The "skip" command in the mini panel program looks promising in relation to the above