Help Required - Signalling

Nimo Sep 3, 2017

  1. Nimo

    Nimo TrainBoard Member

    436
    133
    16
    That's what I have in mind as well. Something close to a real place, but not necessarily an exact place from the book.

    The fun of proto-freelancing is that you also become a story teller. Keeping things grounded on the broad aspects of railroad, era, locale and even accurate history, but not restricted to something very specific to give it the required flexibility. Getting that right balance is actually pretty difficult in my opinion.
     
    Hardcoaler likes this.
  2. Jovet

    Jovet TrainBoard Member

    47
    26
    8
    I want to clarify on this that such things are common on dwarf signals. Most speed-signal dwarf signals can show Clear but cannot show Approach, for example. But high signals usually have a larger variety of aspects available, and so can be truer to an "ideal" system.
     
    Hardcoaler and Nimo like this.
  3. Nimo

    Nimo TrainBoard Member

    436
    133
    16
    One year seven months! Give or take, that's how long it took for me to come back to this thread! But with valid reasons - I am more or less settled in my new apartment where the layout will be built, my N Scale layout is nearing end - 2-3 more months and I can call it done, but the majority of the time was spent on making YouTube videos and running the channel. It was much harder than I imagined, but it is keeping me engaged in making models, so that's good.

    Now, it's time to start planning for the HO stuff. I did get quite a few signals from my visit to the US in Nov 2017 (gosh that was long ago!), and the plan of course has changed considerably based on how I set-up my apartment. I am trying to follow a specific prototype as close as possible, but of course it's easier said than done. The PDF file shows the original Childwold Station in the Adirondacks division of New York Central. The image shows my rendition of it in the space I have available. Of course, this one is bare bones, just to show the signals and track arrangements. I plan to build a small prototype first with all the signals in a much smaller space to practice the programming and build the logic before I start with the actual layout.

    Coming to the track itself - the blue line represents the NYC main. All switches are actually on normal position for this route except switch 7. If these were all straight switches, I wouldn't need a modeler's license to call it a high speed route, but the fact is that they are curved. I am still calling it a theoretical high speed route, but I don't see any train practically running in high speed here anyway irrespective of the switch geometry - the blocks protected by 4E will function as nothing more than stub ends to give engines and small trains to move just beyond the switches - you can neither start nor stop at high speed in that much space, even for the tiny little trains that the tracks can hold. Same story for 24W protecting the staging entrance. The only signal that can show a true clear in 23E.

    Another small quirk in the plan is the position of the signal 5W - ideally it should be closer to the switch 7, but given there is going to be a coal tower in between, I moved the signal before the coal tower - in my common sense that's what a railroad would do as well, but I would like to hear what you guys think.

    Now here is the big question - what do you guys think? The last discussion did help me this time quite a bit, but let's be honest, after over 18 months, I was catching up on things as well. So any constructive criticism and comments are more than welcome. :)

    Module-1April-19-alt-Signals.jpg
     

    Attached Files:

  4. Jovet

    Jovet TrainBoard Member

    47
    26
    8
    Hello again, Nemo. Good to see you're still out there.

    My first reaction to your diagram is you have too many signals. I can understand the siding past the coal tower being signaled, but everything off the mainline (e.g. leading to the turntable) I don't think would be signaled at all. The track marked in red, if itself a main line, could be. But, based on the diagram in the PDF, it's a small railroad likely not equipped. That means that 8W, 11W, 15W, 14E and 20E could be eliminated.

    My second reaction is that signals like 12E, 4EA/4EB, 5W, 18W, and 24W would, indeed, show Clear. The exact sequence of aspects used would depend on the speed limit of the track and designed speed of the turnouts, but the whole point of a main line is to get from place to place. Signals indicate transient speed limits, which are temporary and are limited in time and/or distance. A train only needs to slow down for a turnout when it has to take the diverging route off of that turnout. A permanent speed reduction for, say, a curve is not a transient speed restriction. I can guess that, for example 12E, you see as having to slow down for either turnout #9 position, but I don't see a prototype railroad designing a mainline that way. Several of your curved switches (#7, #3) I see as should diverge to the outside, which could be tricky to model correctly. They might require some extra-pretend on a model railroad.

    Lastly, I think I would put 5W past the coal tower. The interlocking signals define interlocking limits, and most interlockings should be as compact as possible. Eliminating 12E means there should be a dwarf signal added between switches #10 towards #7 to protect the main line. Switch #7 could even be a manual switch, which means 5W would move to between switches #7 and #9, and the dwarf signal just suggested would not be needed.

    I'm not sure what to suggest for your desired semaphores, though. You could keep 20E because "reasons," and maybe turn 19W into the other semaphore?
     
    Nimo likes this.
  5. Nimo

    Nimo TrainBoard Member

    436
    133
    16
    I think I was suffering from the very common disease of over complicating things when not required! Of course, nothing here fouls the main, so I see no reason for keeping them signals.

    With the explanation you provided about the curved turnouts, I agree for all. My only question is about 4EA/4EB and 24W. You see, they can definitely show the clear aspect since the line 'goes on' in the assumption, but in the model, the trains will have to stop as soon as they cross the signal (hit the end for 4E, and the staging for 24W), so the train on the layout will have to slow down before they reach these signals. I am not sure how realistic it would look when the signal is clear, but the train crosses it at 10-15 scale mph - that's all. The way I was thinking is that the 'off-stage' section of the railroad impose a speed restriction that requires trains to slow down to the right speed when they approach these signals. Does that make sense? The heads can show the option of a clear (i.e. there will be no blanks), but in the real operation the clear will never be shown.

    Again, thanks for pointing out the fact that 'clear' necessarily means maximum track speed, so I just have to put a speed limit board at the approach of the curves. I am not really sure how I'd get a diverging outward in #7 and #3 though - given they are curved turnouts. But as you said, I guess I'll take help of the extra-pretend part.

    If I convert #7 to a manual turnout and move 5W between #7 and #10, then do I still need the slow speed head, considering the fact that the tracks behind the Depot is not within interlocking limits anymore?

    I'm more interested about getting the heads and the aspects right to be honest. The way I see it, if I am keeping the story that the Semaphores are in the process of replacement, then as long as I keep them fairly close to each other, I think it will remain believable. May be I'll just put a few blank target and poles nearby those semaphores to simulate that the days of those semaphores are numbered.

    Thanks again Joseph for all this. Let me go back to the drawing board and I'll get back with a fresh diagram soon.

    Cheers!
    Kaustav
     
  6. Jovet

    Jovet TrainBoard Member

    47
    26
    8
    Well, it's easy to want to put signals everywhere. (Especially if you've already purchased or made them!) Many of us have done that.

    Yes, I see what you are thinking. There are several options to handling that situation. Given how short and small your diorama is, you could consider a toggle switch that might let you select between Clear/Approach or Clear/Approach/Stop aspects for each final signal. This would let you simulate various conditions beyond the layout and real track itself. In the train simulation games I've worked on, I've always kept the notion that the signaling system continues (the signals show Clear) but the track speed limit drops accordingly, such as to 10 MPH right before the physical end of the track. (But that's also for routes that can be hundreds of kilometers long.)

    Yes, the curved turnouts make things more complicated, just as the flow of the mainline and forced divergences did of the previous plan. I can't claim to know what all is available for the track you've picked out but I'd bet the most-prototypical solution would be to build your own outside-diverging curved turnouts—a significant undertaking and challenge.

    I am curious if you ever considered keeping the mainline as the track closest to the edge, letting the siding be what goes under the coal tower?

    No, not that I can see. (Remember that [5W] would sit between turnouts #7 and #9.)

    But, given switch #21 is a controlled and interlocked turnout, I don't see any harm leaving #7 that same way, either. It's up to you.

    You really have all sorts of things you could do. You could make [18W] a semaphore. Or [24W] and/or [23E]. ([23E] should have a number plate since it should be permissive. [24W] too, if you don't actually intend it as an interlocking signal.) You could keep [20E] as a semaphore, and put another signal alongside it for westbound traffic, also a semaphore. This could represent an aging or feigningly-used ABS signaling system used on that other line. (I don't think [20E] could count as a headblock signal but it could be a final control for trains entering into that area, or depending on your parts available, just show Restricting of some kind.)
     
  7. Nimo

    Nimo TrainBoard Member

    436
    133
    16
    Well, I purchased the signal kits, but I'm yet to build them. Except the 2 Tomar semaphores and one NJ internal Semaphore dwarf, all of them are still lying in their packets in kit form. That's why I'm still focusing on getting the signals right rather than thinking about how to utilize what I bought. A good friend of mine once told me never to let the tail wag the dog and I am trying to practice that. :D

    OK. As long as that's how the prototype operated, I think I will be comfortable doing that. I intend to use JMRI and Digitrax occupancy and signalling, so I guess all those rules can be set in the computer itself. That will be a HUGE learning curve for me and possibly will take a year to master - and that's why I am going for a small layout.

    You know, that's exactly what I thought at first. I will be using Atlas curved turnouts - the mainline radius of those are 30" and the diverging routes are 22" - so those dimensions made the decisions for me. If I interchange the passing with the mainline in the current design, then the main goes through the shorter radius all the way through, whereas in the current design it at least takes all the broader curves except for turnout 7. That 'U' shaped wall I have is just 7 ft and I can hardly make it 16 inches deep. Even with hand laid switches I think it would still be difficult to make it look right. I spent a lot of time and did a lot of permutation with different manufacturers' curved turnout, this iteration gave the best track flow and I could put at least 10 degrees of easement curves on each curved turnouts - that's why I finalized on this.

    I personally feel that this design is still much better than the previous one. In fact if I had another 7 ft of straight wall, I'd make this whole thing straight and switches #7 and #9, and #2 and #3 would be crossovers (in fact they actually are crossovers even in this design, it just doesn't look like that since it sits on a curve). I am basically twisting a straight line to fit in my space, that's all.

    Right. So if I understand correctly, if I make #7 manual and move 5W between #7 and #9, then I just omit a need for another dwarf signal and an additional head on 5W, whereas keeping 5W as is, I keep 3 heads there and a dwarf to protect the entry to the main. I think let me get the rest of the signals right and check back with my stock to see if I will have enough heads left for 3 headed 5W and a dwarf, if not, I make #7 manual.

    Grasse River RR was actually the biggest logging RR in the Adirondacks and operated for 70 years all the way till 1950s (that's the decade I'm focusing on in the model). However, in no photos or texts I've seen any mention of a signal. The reason I'd put 20E there was because I was using the old design as a reference where the logging RR line intersects with the main, so the protection was necessary from NY Central's safety standpoint. Here it follows the prototype and comes to a siding. So, I think your initial reaction is justified, a railroad like that will not have signals. If there was a signal it would most likely be put there by NY Central to protect the interchange, but now that it is on the siding, I am not sure if the prototype followed the practice. Also, all these switches will have switch stands. The only argument I can see to keep 20E is that the switch stand of #17 will be on the left of the line of the inbound traffic is also curving left, so the engineer needs to know to look left and not right when approaching Childwold, so 20E just adds to the safety of the interchange especially in days when visibility is poor. I honestly don't know if the real railroads thought in those lines, or relied on the training of the engineers of these smaller lines.
     
  8. Nimo

    Nimo TrainBoard Member

    436
    133
    16
    The point about moving the main to the inner most track got me thinking again. That's what is there in the prototype and that's what makes most sense. I am just not sure about considering a 22" curve on the main and put it on the diverging line. I am also looking at the Atlas curved turnouts, the one I intend to use and the inner routes are indeed diverging ones.

    I'm undecided again about which side to sacrifice! :unsure:



    ATL-595-2__38821.1519315182.jpg ATL-596-2__28430.1519315183.jpg
     
  9. Nimo

    Nimo TrainBoard Member

    436
    133
    16
    Revision 2:

    A change in assumption - I removed signal 1W in the previous plan. That part of the track will not be visible clearly anyway - it will be sort of a semi-hidden part with a lot of trees and scenery covering it, so I thought I can leave that technically in the 'backstage'.

    I also made some changes to the labeling - the switches are numbered in red and they will be static going forward. This will ensure that when changes are made the switch numbers will remain constant. For the signals, I changed the nomenclature to E<Number> and W<Number> and followed them in series as per the direction, i.e. starting at the east most point going west the signals read E1, E2, E3... in sequence and vice versa while travelling east.

    Now about the signals itself, I assumed that the top speed of the mainline is 30 mph. The biggest stretch in assumption is that the thrown position in switch 3 is actually the normal - i.e. the inside curve is the normal position for the switch, whereas outside position is thrown.

    In my understanding, this gives me the flexibility to chose variety of aspects depending on the traffic movement (I am still getting my head around all of it), including full clear on all 3 speed limits, except on the route of 4-2-1 since it will be on a diverging route, hence the green lamp is removed from E6B high speed head.

    Coming to the semaphore, I decided to keep the erstwhile 20E, now renamed E3. I somehow like the idea of an added protection for trains reaching Childwold down a 3% grade. I'll place the other two nearby as W3 and W4 (Dwarf semaphore). It will also give a sense that this is the last remaining cluster of semaphores on the line. Might even put a second mast in the process of erection for the upcoming G-type right beyond the W3.

    So, how does it look now?

    P.S. I am still debating in my mind about moving the main from 4-3-1 to 4-2-1. If that happens then I guess that part will change again. Eliminating 1W in the previous plan leaves me with some extra heads, so I am keeping point #3 within interlocking limits.

    Module-1April-19-alt-Signals2.jpg
     
  10. Jovet

    Jovet TrainBoard Member

    47
    26
    8
    It's my understanding that most train crews don't even bother to look at switchstands... they look at the actual points to determine how a switch is set.

    In the 1950s, you would not have any signals on the left-hand side—that was "illegal." Instead of being put on the left, they would be put on a signal tower, or on a signal bridge or cantilever. Whether switchstands counted in that rule I don't know for sure, but I doubt it. I would think switchstands would be positioned wherever they're safest to be used.

    I feel your pain! I'm afraid I don't have any witty or helpful advice or suggestions. Perhaps some more-experienced modelers can chime in.

    It would be more protoypically-minded to number "1" and "2" as the same turnout, as it is essentially a crossover, and both switches are always operated together. Additionally, it would be more protoypically-minded to label the signals on "E6," which are for separate tracks, as separate signals ("E6" and "E7"). Letters like "A" and "B" usually denote individual signal heads on signals, though it also would not be unheard-of to have the single "E6" signal have heads "A" and "B" on the left, and "C" and "D" on the right.

    On a second thought, while you still may not install the former [1W] signal, you might be wise to reserve its number "W1" for it, and start numbering with "W2".

    I think it looks fine, to me. I am not sure you need all of the lamps on all of the heads you show, but that is a bit of a minor detail.
     
  11. Nimo

    Nimo TrainBoard Member

    436
    133
    16
    Yes, I'm not putting any signal on the left. For the switch stands, the challenge is there is no 'right' or 'left' for a switch stand since trains can approach from both directions for a switch. In case of switch 8, the stand is on the left for the trains approaching from the East, but is on the right for trains travelling from West to East. I believe that makes it even more important to keep signal E3.

    No worries! I have accepted the compromise of using that turnout since it fits the space very nicely. The layout will be nearly at eye level and the layout is visible from the 'inside' of the curve - visually no one will really notice the un-prototypical divergence on the curved turnouts.

    Agreed. I'll make the changes i the next version.

    Hmmm... considering my assumption that the top speed is 30 mph and it can show clear on all three heads, I am not sure I am following. Which lamps do you think I can get rid of?
     
  12. Nimo

    Nimo TrainBoard Member

    436
    133
    16
    I discovered a fantastic blog today and found this: http://position-light.blogspot.com/2014/01/north-american-signaling-dialects.html

    The blog itself contains a ton of information so it will take time for me to go through all that, but at least the above summary is very helpful for me. Still digging...

    What is very interesting is that it seems that certain aspects of speed signaling are indeed position signaling aspects! This is brand new information for me and this certainly makes things easier for me. However, I am not sure how complete this list really is.
     
  13. Nimo

    Nimo TrainBoard Member

    436
    133
    16
    OK. After looking closely at the design, I decided to make some significant changes in the track plan, and finally it seems like I am happy with what it looks like. In the end, I ended up following the prototype station almost as-is, except rearranging it to suit my space and situation. The basic structure however remains the same.

    Coming to signalling, my biggest challenge is to handle the exit from the layout to staging or fiddle because trains that are as long as my longest staging track will have to exit the visible area at a pretty slow speed because they will have to come to a stop (and I will be using realistic momentum for each train length). So I decided to assume required complications off the stage that might require the train to slow down to match the exit speed. So, length of the train decides the exit speed, that decide the exit signal, and in turn govern the signals approaching it. I know it seems like the tail is wagging the dog, but given my longest block length will be a little over 4 feet and the staging will be barely exceeding that length, I found no other alternative.

    We talked about reducing the track speed, and I have reduced normal track speed from 60 mph to 40 mph to denote full clear for smaller trains. Reducing it further will essentially bring the track speed to the range of Medium speed, and don't know what effect it will have on the downgrading signals - couldn't find any clear answers in my research, so tried avoiding that part.

    Bottom line is the biggest stretch in this story is that the real prototype was in the middle of the Adirondacks wilderness, so there wouldn't have been any 'Approach' aspect on either of the exits, nor would be any slow speed exit under normal circumstances. So, essentially I am putting this area which was originally in the middle of wilderness, between two other busy towns perhaps. It is a stretch. But the funny part is installing signals in this line itself is a stretch since there was barely a train order signal at the depot and nothing else! Yes, following a prototype and satisfy your desires at the same time is indeed one of the toughest things in this hobby!

    So, accepting all the assumptions, how does it all look now?

    Phase-1-April-22.jpg

    Phase-1-V22-Signal-Schematic-2.png
     

    Attached Files:

  14. Nimo

    Nimo TrainBoard Member

    436
    133
    16
    3 years since the first post on this topic! Look at that! This is why they say time flies.

    Well, one thing led to another and I still haven't started with the project - I guess when you really want to start right, you gotta take your time (Upto a point). The reason I came back to this post is because coming November I am finally kicking it off - exactly 10 years from the time when I started Wrightsville Port. A lot changed in last 3 years and the clarity I needed to get started is finally here. In the meantime I researched about my prototype to the best of my abilities, multiple versions of the plan have been made, strong negotiations were done with self and the 'home ministry' and finally the plan was 'rubber stamped' a couple of weeks ago. All that will be coming up in a new thread as the project starts. But now to signaling.

    Since the last plan, I decided to simplify things and bring it as close to the prototype as possible. But before that some refresher context - The layout will be primarily logging - a large logging operation based in Conifer, NY in the matured years of Emporium Forestry Company. Then the line interchanges with the NY Central's Adirondacks Div. at Childwold. Everything had to be compressed to fit my 'givens and druthers' but the track plan at Childwold pretty much follows the prototype, with just one compromise: one passing siding instead of two. However, the station had to be compressed further and moved far to the right to make room for Conifer, the main subject of the layout. That means, there is a long stretch of hidden tracks till the main reaches the small 2-track staging (Herkimer, or towards Herkimer), intentionally kept two tracks to denote a passing siding and incrase operational value. The other side has a smaller 2 track staging (denothing another passing siding 'towards Tupper Lake') crammed in the corner of my workspace so that I can at least turn a loco around and put it in front of a train for a return journey. The trains on the line will be short - upto just 4 feet in length. If it's any consolation, the real trains back then on that line were not long haulers either - 3-5 car passenger trains and 8-10 car freight used to be common sightings.

    All 'diverging on main' and non-prototypical 'slow speed souble-slip' type compromises in the previous plans have been eliminated - the main follows the 'Normal' on all switches, including the curved ones where used. All switches on the main are medium speed (#6 or higher). The mainline coal tower is included at the entrance of the Tupper Lake staging to justify a train 'starting' at that point despite a clear route ahead in certain scenarios. I have used some cleverness of typical micro-layouts here to make 'abnormal' things look normal even if that meant introducing a mainline coal tower near Childwold even though there were none in real life. Well, you can follow a prototype only upto a point, unless you own at least 1/87th of the land that the real owners had for a model railroad!

    With all that preamble - here are the plans -
    1. Overall
    2. Childwold+staging
    3. Zoom in version of Childwold.

    Childwold-and-Conifer.jpg
    Childwold-Staging.jpg
    Childwold-Closeup.jpg
    Childwold-3D.jpg

    Next-up the signal schematic.
     
    Last edited: Sep 14, 2020
  15. Nimo

    Nimo TrainBoard Member

    436
    133
    16
    Now coming to the actual signalling, I simplified the concept there too. However, imaginging how little the station is, and the fact that the trains actually travel nearly 3 times the length of the visible track in the hidden tracks, I thought it would be interesting if I actually signal the hidden tracks too, and add those signals on the fascia, so you don't see the train, but you can pretty well see the train 'moving' by following the signals. This actually presented some unique opportunities to do decent degree of signalling even within a relatively small set-up. It will also give that sense of distance.

    The mainline operations will be mostly computer controlled (CTC) with 'local' switching happenning at Childwold where I will have 'dual control' switches just like the prototype. Dispatcher will authorize the control of these dual control switches.

    Here is the signal schema, as well as signalling for main operational scenario. Yes, it is small, but as you will see, the operation potential is pretty engaging.

    As always any critique is most welcome. Though my knowledge of NORAC signalling and NYC signalling before 1950s is much stronger than what it was 3 years ago (Thanks to all the wonderful helps that I have received here, mainly from Jovet, plus all the great documents that I unearthed in the meantime), I always keep room for improvement.

    Cheers!
    Kaustav
    Signal-Schema-Sep2020.png
     

    Attached Files:

    Last edited: Sep 14, 2020
  16. Jovet

    Jovet TrainBoard Member

    47
    26
    8
    Hey, how's it going here? :)
     

Share This Page