Drafting the Tutorial Signalling Section

Following on Brians post here is the YMR layout stage I ……….unsignalled

I suggest we have a go at positioning the appropriate signals first then as a second exercise work out the appropriate conditions

For the purpose of the exercise lets assume that the buffer is the line towards Weston and the extremity of the Eastham Branch Box

I will have an attempt this evening  

Regards

 

 

 

 

 

Hi again John.

I'm feeling guilty at disrupting your work plan for the tutorial. I will try to make amends by having a go at a signalling scheme.

I don't think it matters about having a realistic scheme: We will all have to make our own minds up on what we use on our layouts. The key is to agree on the signals we will use here and get them to work from TC.

A small branch like this would probably run on "One engine in steam" and not need any signals, but that defeats the object. The signals I am suggesting here are probably overkill, but I have no fixed ideas - You're the boss on this and I will readily agree with whatever you want to do.  I think this setup could deal with a shuttle, with a DMU arriving at one platform and another immediately leaving from the other.

If this becomes more than a group of two looking at the problem I think we will need to share the YRR file and arrange some way of controlling amendments to it.

This is as far as I've got- please tell me what you think of it so far.

Regards,

Brian

Hi Brian (and John), I have been continuing to read the latest tutorials from John, and am interested in how this signalling one is going to develop. I am unlikely to introduce TC operation of signals on my layout for the forseeable future, but am interested in signalling and will enjoy the intellectual exercise with TC.

Couple of questions in regard to your proposed signalling arrangement for the tutorial layout. Firstly, is the advanced starter viable without any blocks beyond the left hand "buffer"? I ask this because presumably it would require a "free" block to the west yet we don't have any more track in that direction - but maybe I misunderstand the function of the advance starter.

Then, I wonder if it is not too complicated to include some call on signals with the 652/3 homes to make provision for run around locos to back onto coaches left behind in the platform?

At the same time, would it be possible to include a ground signal in 653a to permit a run around loco to come back past the said coaches? Maybe also shunt signals into the goods siding and back out again?

I have put a lot of thought into signalling in several areas of my layout, and would find this tutorial very useful in helping me understand the operating rules for such signalling - albeit under the management of TC.

I'd like to ask a question about your Track Plan, Brian.  It looks like there is track outside of the blocks.  Is that right?  Once the train leaves the block, "Eastham Home," how does TC keep track of it?  If you have a sequence which ends on that line, it seems to me that once the train leaves the EH block, the sequence won't be able to end?

. . . or is it just me being thick again? 

I think you make a good point, Max (by the way, what time is it there, must be very early Thursday morning?).

I wonder if we would be best to go back to John's full tutorial schematic with an intermediate station and Westham terminus on the left. We don't have to signal the whole thing in one go, but at least it will allow trains to move through several blocks away from Eastham and therefore have Eastham signals operating fully - I think!

7 a.m. Thursday, Geoff.  Been up for an hour reading my emails.  :shock:

Here is a pic of the switchboard I am on about:-




I have added what I think would be the essential signals. Not sure if I have used the right formats for call on and ground - I am used to semaphores. I am still trying to work with Silver, so no point in posting the yrr as I don't think you would be able to load it into Gold.

Just noticed that having carefully set the signals from Eastham towards Weston to Red, they have gone to Green in this screen grab!!!!! We haven't even started yet, and TC is doing its own thing:twisted:

First thing I discover, which is blindingly obvious really, and that is a 3 aspect signal requires two output addresses. TC Silver automatically assigns them sequentially - just like double slip points.

So, taking my switchboard above. I set the trigger condition for Weston P1 signal to go green if the route is set from P1 to Weston Home, and then set the trigger condition for it to go back to red if Weston Home is occupied. I set all signals to red.

Then, having placed a train in Weston P1. I start the schedule for Weston P1 to Eastham. The signal goes green. Start the sim and after a short delay, the train leaves P1, enters Weston Home and the Weston P1 signal goes back to red.

Maybe there needs to be more to it than that?

Hope I'm not moving too fast here?

I have made the same assignments for the signals heading towards Eastham for Weston Home, Ashby Halt and Eastham Home.

Very interestingly, although there is clearly a route from Ashby Halt to Eastham Home, the signal at Ashby Halt does not go green until the schedule is ready for the departure from Ashby Halt. This must mean that TC does not set the route internally until it is ready for the train to move off.

This set of signals then work well all the way across from Weston P1 to Eastham Station.

My simple approach is not sufficient for the bi-directional line. I have now added an additional trigger for the green state, and that is the current block being in the state of having a green signal to the desired exit. Seems to then work okay in both directions.

Guess that I could have used conditions instead of triggers. Have to think about the difference??

Keeping track of all this needs a good memory!

To look at some of the responses: My original scheme assumed the extension to Weston would come later so I put in the outer home (if that's the right term) and the adv. starter. I put in a calling on signal below each of the homes 652 and 653 on the assumption that they would be needed on a run around and perhaps shunting.  I looked at Minehead Station (West somerset R.) in the summer - their release crossover is controlled from a ground frame with a key which I assume is released from the Box. I don't think it matters very much what we decide, the aim is to link it to TC.

On my layout I have set up one starter to work with the engine change schedules. As recommended in the manual I have left the Red trigger blank. If nothing else applies the signal goes red with no need to reset it.

To trigger Green I use this:

Block in advance:  Empty
AND
current block:      Occupied
AND
route between them:   Active
AND
      either
      block in advance reserved for a train moving to the right (if the current block is on the left and next block on the right)
      OR
      the On-Off switch (activating the route for manual control) is On
AND
the signal ahead is "NotRed" i.e. yellow or green.

The trigger for Yellow is identical except the last line reads:

the signal ahead is Red

This seems to work OK for any type of running including manual and run with interlocking and keeps the signal at Red when a train is travelling in the opposite direction.  I don't think it covers every possibility but it covers most.

The reset happens when the start block is vacated. I set the memory to switch off when tail of train has passed: I think that is the rule for Absolute Block.  If the memory is left at auto or on a short timer I think that would comply with modern Multiple Aspect rules.

[user=449]brianpr1[/user] wrote:

Keeping track of all this needs a good memory!

That is certainly true

I am very happy for you to take the lead in this  Brian…………..you are much further ahead than I!

Maybe I can pose some questions to clarify where we are agreed and where we need more discussion 

INTITIAL TUTORIAL LAYOUT

I think the first segment of the tutorial should focus on the Eastham Branch that you posted plus the addition of Ashby Halt, Weston Home and Weston P1. That way we can link everything to a couple of simple schedules.

SIGNAL TYPE 

Three aspect

SIGNAL PLACEMENT

My signalling knowledge is based on GWR Semaphore so be warned……..

I agree we need Starter and advance starter so that Eastham Home is within Station Limits and the loop can be accessed…….I am not sure about calling on arms in a branch. I would have thought that this would be a function of ground signals 

GROUND SIGNALS

Unlikely to be modelled but do we need to include them in the scheme

Our main focus is on TC Signal Operation not a debate about placement……………my inclination would be a PM to Chris and ask him for the definitive version

TC TRIGGERS AND CONDITIONS

Your set up looks great to me Brian and has the advantage of having been tested. I assume "Active Route" mean the points are set? What happens if you run manually?

I think once we our signals placed we can easily determine the specific triggers and conditions for each signal

TC Operation

I know nothing!!!  For tutorial purposes we need to come up with at least some alternatives. I had assumed it would be similar to points with a separate power suppply to the light signals that would be fed through some form of accessory decoder that would receive commands from TC.

 There is a lot of stuff on RMWeb……Two Tone Green who was briefly a member here has his layout fully signalled as of course does Beast6606………….I will do some research 

Hopefully this has covered the bases……..let me know what you think

 

 

[user=434]John Dew[/user] wrote:

[user=449][/user]
INTITIAL TUTORIAL LAYOUT

I think the first segment of the tutorial should focus on the Eastham Branch that you posted plus the addition of Ashby Halt, Weston Home and Weston P1. That way we can link everything to a couple of simple schedules.

I agree we should eventually base it on the whole tutorial layout, but for the "Thrashing it out " exercise I would prefer to keep it very simple at the start. I suggest we confine the first stage to the Eastham station limits, delete the left hand stop block (I believe that's the correct term?) and put on a note "continues to left" .  I think we could add one more block to the left to give the adv. starter and Home a reference block. I hope we can set up the signals to be independent of schedules.  I have only the one starting signal working, but it works on Run with Interlocking etc. and, with care, for manual operation.  Unless we are forced into relying on schedules I think the time to add Weston etc. is after we have dealt with Eastham.



SIGNAL TYPE 

Three aspect 

That's what I had in mind too.

SIGNAL PLACEMENT

My signalling knowledge is based on GWR Semaphore so be warned……..

I agree we need Starter and advance starter so that Eastham Home is within Station Limits and the loop can be accessed…….I am not sure about calling on arms in a branch. I would have thought that this would be a function of ground signals

GROUND SIGNALS

Unlikely to be modelled but do we need to include them in the scheme

Our main focus is on TC Signal Operation not a debate about placement……………my inclination would be a PM to Chris and ask him for the definitive version

We may as well keep it in the open - I will ask in the Signals section.

TC TRIGGERS AND CONDITIONS

Your set up looks great to me Brian and has the advantage of having been tested. I assume "Active Route" mean the points are set? What happens if you run manually?

Tested means two blocks and one signal, it is definitely not thorough.

I use an On-Off switch labelled depart 1. If the block in advance of the signal is not reserved by automatic operation I must manually turn the switch on or the signal stays red. An operation of the switch uses the "Routes" sub menu and sets the route forward to active. This not only sets the points but also locks them. After that the signal changes and allows the driver to leave. Arrival at the stop marker in the block ahead turns the switch off and the off state of the switch turns the route off.



I think once we our signals placed we can easily determine the specific triggers and conditions for each signal

I'm not convinced about "easily", but yes.

TC Operation

I know nothing!!!  For tutorial purposes we need to come up with at least some alternatives. I had assumed it would be similar to points with a separate power suppply to the light signals that would be fed through some form of accessory decoder that would receive commands from TC.

I think that's about right, but compared to point solenoids the power required is very low. Each lighted LED takes in the region of 10mA. I don't know much about the commercial options.  I think the LDT and traintronics drivers include decoders, perhaps some one has specific details?
 
 There is a lot of stuff on RMWeb……Two Tone Green who was briefly a member here has his layout fully signalled as of course does Beast6606………….I will do some research 

Hopefully this has covered the bases……..let me know what you think

 

 

I hope this moves us forward a little. We need to decide on the extent of the layout we are going to ask Chris to look at: Another reason for keeping it simple at this stage.  If Chris comes up with a scheme I suggest we accept it in full without further debate and concentrate on implementing it in TC.

One more point occurs to me: I have shown two Home signals side by side on my original diagram because I could not see a better option in TC, but in practice I think it would be one signal with a feather. So far as TC is concerned I think the second signal could be forced to operate the feather.

See below in Red

[user=449]brianpr1[/user] wrote:

[user=434]John Dew[/user] wrote: [user=449][/user]

INTITIAL TUTORIAL LAYOUT

I think the first segment of the tutorial should focus on the Eastham Branch that you posted plus the addition of Ashby Halt, Weston Home and Weston P1. That way we can link everything to a couple of simple schedules.

I agree we should eventually base it on the whole tutorial layout, but for the "Thrashing it out " exercise I would prefer to keep it very simple at the start. I suggest we confine the first stage to the Eastham station limits, delete the left hand stop block (I believe that's the correct term?) and put on a note "continues to left" .  I think we could add one more block to the left to give the adv. starter and Home a reference block. I hope we can set up the signals to be independent of schedules.  I have only the one starting signal working, but it works on Run with Interlocking etc. and, with care, for manual operation.  Unless we are forced into relying on schedules I think the time to add Weston etc. is after we have dealt with Eastham.

Agreed

SIGNAL TYPE 

Three aspect 

That's what I had in mind too.

SIGNAL PLACEMENT

My signalling knowledge is based on GWR Semaphore so be warned……..

I agree we need Starter and advance starter so that Eastham Home is within Station Limits and the loop can be accessed…….I am not sure about calling on arms in a branch. I would have thought that this would be a function of ground signals

GROUND SIGNALS

Unlikely to be modelled but do we need to include them in the scheme

Our main focus is on TC Signal Operation not a debate about placement……………my inclination would be a PM to Chris and ask him for the definitive version

We may as well keep it in the open - I will ask in the Signals section.

Good Idea I hadnt thought of that……you may want to make a decision on Ground Signals…….I will bet you a Pint of Speckled Hen (yes we have it here) that Chris will include a number of them unless you suggest otherwise

TC TRIGGERS AND CONDITIONS

Your set up looks great to me Brian and has the advantage of having been tested. I assume "Active Route" mean the points are set? What happens if you run manually?

Tested means two blocks and one signal, it is definitely not thorough.

I use an On-Off switch labelled depart 1. If the block in advance of the signal is not reserved by automatic operation I must manually turn the switch on or the signal stays red. An operation of the switch uses the "Routes" sub menu and sets the route forward to active. This not only sets the points but also locks them. After that the signal changes and allows the driver to leave. Arrival at the stop marker in the block ahead turns the switch off and the off state of the switch turns the route off.  As you have gathered I am rather schedule bound…..I will need to do more research but sounds good

I think once we our signals placed we can easily determine the specific triggers and conditions for each signal

I'm not convinced about "easily", but yes.

TC Operation

I know nothing!!!  For tutorial purposes we need to come up with at least some alternatives. I had assumed it would be similar to points with a separate power suppply to the light signals that would be fed through some form of accessory decoder that would receive commands from TC.

I think that's about right, but compared to point solenoids the power required is very low. Each lighted LED takes in the region of 10mA. I don't know much about the commercial options.  I think the LDT and traintronics drivers include decoders, perhaps some one has specific details?

 There is a lot of stuff on RMWeb……Two Tone Green who was briefly a member here has his layout fully signalled as of course does Beast6606………….I will do some research 

Hopefully this has covered the bases……..let me know what you think

 

 

I hope this moves us forward a little. We need to decide on the extent of the layout we are going to ask Chris to look at:

Go with your suggestion above…….Eastham plus tone block 

Another reason for keeping it simple at this stage.  If Chris comes up with a scheme I suggest we accept it in full without further debate and concentrate on implementing it in TC.

Totally Agree

One more point occurs to me: I have shown two Home signals side by side on my original diagram because I could not see a better option in TC, but in practice I think it would be one signal with a feather. So far as TC is concerned I think the second signal could be forced to operate the feather.

Lets see what Chris suggests

Regards

I've posted a request for Chris in the signals section.

This thread is moving too fast for me to keep up. After taking another look at Geoff's plan I realize that I have been assuming without thinking about it that there is an island platform between Eastham station and Eastham loop, and it is much more likely to be a one sided platform on the South side, with a simple loop as in Geoff's scheme.

Geoff:  On your plan you have not shown the block 651 Eastham Goods. On a layout it is not needed, TC keeps the loop reserved and when the engine reappears from the goods siding it is correctly identified.  When it comes back on the sim it leaps instantly to the block ahead but I don't think it matters.  It is probably better to have fewer blocks, but  leaving it out eliminates any automated shunting. Once again, for signalling purposes that does not matter. If Chris uses a ground signal to control the goods exit we may have to think again.

Looking at your triggers etc. we probably need to decide soon whether we are dealing with scheduled running only or making the signals work for all four methods from schedules down to manual.  Should we just drop any attempts to drive on manual? That seems to cause far more difficulties with signals than any of the automatic methods.  I find it very useful to use Autotrain Drag & Drop to move trains around for testing schedules, and it is nice to see my only operational signal change to permit the movement.

We seem to have hit a brick wall on the hardware question. Does anyone know of other options besides LDT and Traintronics?

[user=449]brianpr1[/user] wrote:

I've posted a request for Chris in the signals section.

This thread is moving too fast for me to keep up. After taking another look at Geoff's plan I realize that I have been assuming without thinking about it that there is an island platform between Eastham station and Eastham loop, and it is much more likely to be a one sided platform on the South side, with a simple loop as in Geoff's scheme.

Geoff:  On your plan you have not shown the block 651 Eastham Goods. On a layout it is not needed, TC keeps the loop reserved and when the engine reappears from the goods siding it is correctly identified.  When it comes back on the sim it leaps instantly to the block ahead but I don't think it matters.  It is probably better to have fewer blocks, but  leaving it out eliminates any automated shunting. Once again, for signalling purposes that does not matter. If Chris uses a ground signal to control the goods exit we may have to think again.

Looking at your triggers etc. we probably need to decide soon whether we are dealing with scheduled running only or making the signals work for all four methods from schedules down to manual.  Should we just drop any attempts to drive on manual? That seems to cause far more difficulties with signals than any of the automatic methods.  I find it very useful to use Autotrain Drag & Drop to move trains around for testing schedules, and it is nice to see my only operational signal change to permit the movement.

We seem to have hit a brick wall on the hardware question. Does anyone know of other options besides LDT and Traintronics?

I showed the layout as per the tutorial when I was following John. I stopped when the sessions got beyond me because of my lack of sufficient block detectors in place and not using Gold, so I missed out on the introduction of the Goods siding. I suggest that it is included to enable signalling for shunting.

I know that I went for triggers based on schedules, but I do like your thoughts around signalling for manual control. In fact, my current intention is to use TC for point control through route setting and not running trains under TC as I do not have occupancy detectors in place at this time. I am interested in signalling, though, and it would be good to run the signals from TC at the same time as setting routes - albeit that I would have to manage without block detection.

As for the hardware, I don't really understand the brick wall. Almost any accessory decoder will drive LEDs or incandescent lamps if set up appropriately. So it should be straightforward to drive light signals. In my case, many of mine are to be semaphores driven by memory wire motors each taking around 350mA when operated, so I would probably need to drive relays from my accessory decoders. I use ZTC ones, and find them very easy to set up and operate.

Geoff,

If everyone else agrees then I suggest we keep the goods block and, by hook or by crook, we make the signals work for all forms of control.

I think the hardware issue is probably cost.  The commercial decoders which can switch to continuous output seem to have four outputs with wiring for solenoids, i.e. 3 terminals per output.  So far as I can see a 3 aspect signal will need two of the four, so two signals will cost about £15 each for conventional decoders.  I think the LDT and Traintronics solution may be  slightly cheaper. I haven't looked in detail at this so I may be completely wrong.  I've chosen  a DIY approach with MERG decoders so the commercial gear is a mystery to me.

Regards,

Brian

[user=449]brianpr1[/user] wrote:

Geoff,

If everyone else agrees then I suggest we keep the goods block and, by hook or by crook, we make the signals work for all forms of control.

Hi Brian

I meant to mention that I visualised Eastham as having a single branch…….I added the Goods branch because I intended to develop a shunting tutorial but I never got round to it……and now Max's is much better

I am happy to stick with the layout as you posted it…..ie with the goods siding

Thinking about control……am I correct in assuming that for anything other than fully automatic (schedules) the signals would control the operation (as in the prototype) ie if the conditions specified for a given signal were not met it would remain at Red and a train would be unable to pass?  

Conversely, with Schedules, TC has its own internal signals and it is those signals which control the operation. The signals the user sets up are purely decorative……functioning cosmetic if you will………you could of course set up conditions to prevent a train passing a signal at Red but I think it would be redundant bearing in mind all the internal checks that TC operates 

That is not to say that users committed to the automatic route would not wish to have working signals correctly linked to their automatic operation (in fact I am beginning to invent a scenario for GWR light trials starting in 1947:shock:)

If the above is valid then I suspect the conditions (or some of them) and the triggers will be different depending on the method of control…………nothing wrong with that…………….it may help us determine where to start…….ie focus initially on one method of operating…..probably yours seeing how far you have got……get that documented and then move on to the next method and dtermine what changes need to be made.

Regards

Hi John,

[user=434]John Dew[/user] wrote:

[user=449][/user]

Thinking about control……am I correct in assuming that for anything other than fully automatic (schedules) the signals would control the operation (as in the prototype) ie if the conditions specified for a given signal were not met it would remain at Red and a train would be unable to pass? 

That's not what I had in mind, although I am working things out as we progress and may be forced to change my mind if this approach hits problems I can't overcome.  I see our signals as a sort of DeLuxe version of the internally calculated block signals, which tend to go grey too often, so that if TC decides a train can go the signal will reflect that. For Manual the signal will change on the same basis, but it will be up to the driver to obey it.
 
Conversely, with Schedules, TC has its own internal signals and it is those signals which control the operation. The signals the user sets up are purely decorative……functioning cosmetic if you will………you could of course set up conditions to prevent a train passing a signal at Red but I think it would be redundant bearing in mind all the internal checks that TC operates 

That is not to say that users committed to the automatic route would not wish to have working signals correctly linked to their automatic operation (in fact I am beginning to invent a scenario for GWR light trials starting in 1947:shock:)

If the above is valid then I suspect the conditions (or some of them) and the triggers will be different depending on the method of control……

I'm trying to avoid that by making the same triggers fit all methods of driving.

……nothing wrong with that…………….it may help us determine where to start…….ie focus initially on one method of operating…..probably yours seeing how far you have got……get that documented and then move on to the next method and dtermine what changes need to be made.

My approach is developing into a fix for all auto methods first, then add-ons to cope with manual. It remains to be seen if it works.

Regards

Chris has asked on the Signals section for more info. on the types and methods of working both Eastham and Weston. I think you are probably the best person to answer that, I have already made one false assumption about an Island platform at Eastham!

Regards,

Brian