I received my copy today. After a quick scan, I will say nicely done for such a complex suject in only 141 pages.
Even before an in depth read, I feel it was a worth while addition to the library that will prove useful.
I found the opening line of Chapter 6 interesting "The heart and soul of mainline railroading is the interlocking" - seems I said something similar earlier in this thread.......
I have seen articles about building working interlocking machines before, also an interesting concept.
My take on that is this - I have my turnouts controlled by pushbuttons that in many cases select a whole route, rather than just operating one turnout.
The wiring schemes I do this with involve some simple relay logic, and interestingly provide route interlocking just like a mechanical machine.
Conflicting routes cannot be selected and signals will only display the selected route.
More later,
Sheldon
According to my GRS book "Elements of Railway Signaling', an interlocking signal can be cleared to yellow if the block ahead is occupied and the stick relay is up (i.e., the traffic entering the interlocking is following traffic moving the same direction.) This is shown in Figure 924 by relays 45S (stick) and 45YGP. The latter relay would be down in this condition, where APB is operative:
Ed, now you have it, and yes it is simplified from the prototype.
And yes it has always been my understanding of the prototype that an interlocking signal will not clear if there is a train in the block ahead if ABS is imbeded over the the interlocking CTC.
And understand, this is based on my understanding of a number of systems from my era and region.
The primary reason there is even two blocks between some interlockings is to allow switching to happen at both ends at the same time.
Cabs are assigned to throttles by pushbuttons that are duplicated at the tower panels at each end of the block and on the CTC panel. Route, cab assignment (permission) and occupancy control the signals.
And the system also has ATC (automatic train control), if you run a red signal, your train just stops.
Sheldon,
You mention having one to two blocks between interlockings. As you say, those block affect the interlockings. And the other way. If one of those blocks has a train in it, signaling should keep the interlocking from showing a green for a following train to also enter that block.
And since there are so few blocks between interlockings, it does look like you'll have an ABS system overlaid onto your interlockings.
I've had a quick look at your track plan, and I do see sections of track between interlockings that I would detect. For a single block, the signals themselves would surely be a part of the adjacent interlocking signals.
For that one two-block section, it looks like you'll need a single pair of stand-alone ABS signals. Which, I agree, is pretty close to zero.
I also get your point about layouts being so compact that they don't need very many signals between interlockings. Around here, where all Free-mo modules are expected to be MSS (ABS) equipped, there's been talk of having signals too close together to be believable.
For my signaled module, I've set it up so I can insert either a working signal pair, or an abandoned location.
Ed
Ok, maybe we have a misunderstanding of terms.
I'm only modeling the ABS as it effects the CTC interlockings, with no intermediate blocks, and without any permissive signals.
The fundamental problem with signaling is it is always tailored to the situation, terms and definitions aside.
7j43k ATLANTIC CENTRAL Ok, I have a other minute or two. Yes, where distances are great, ABS can protect long stretches of track with automatic control point passing sidings with bidirectional traffic in both directions between CTC interlockings on single track. Show me a layout where you can really model that with 35 car trains? Sheldon ABS does protect long stretches of track. It also protects interlockings that don't have long stretches of track between them. When ABS is present, it is also applied to interlockings. You don't need to run 35 car trains to have and use ABS. If you want signaled interlockings but do not want ABS, you will be operating with timetables and train orders. For a model train layout, that is typically done by asking "someone" if it's OK to go on this track to the next town or whatever. I suppose you can give this job to your towermen. Ed
ATLANTIC CENTRAL Ok, I have a other minute or two. Yes, where distances are great, ABS can protect long stretches of track with automatic control point passing sidings with bidirectional traffic in both directions between CTC interlockings on single track. Show me a layout where you can really model that with 35 car trains? Sheldon
Ok, I have a other minute or two.
Yes, where distances are great, ABS can protect long stretches of track with automatic control point passing sidings with bidirectional traffic in both directions between CTC interlockings on single track.
Show me a layout where you can really model that with 35 car trains?
ABS does protect long stretches of track. It also protects interlockings that don't have long stretches of track between them. When ABS is present, it is also applied to interlockings.
You don't need to run 35 car trains to have and use ABS.
If you want signaled interlockings but do not want ABS, you will be operating with timetables and train orders. For a model train layout, that is typically done by asking "someone" if it's OK to go on this track to the next town or whatever. I suppose you can give this job to your towermen.
My system is designed to work as CTC with the dispatcher on duty, and act like timetable and train order with no dispatcher on duty. Operators actually can become their own tower operators. I'm modeling 1954.
Each interlocking is on the CTC panel, and has its own local tower panel as well.
Further back in this thread is a fairly complete description.
The double track mainline will be signaled for bidirectional traffic.
At no point is there more than two blocks between interlockings, in some cases only one.
Remember it is DC and the distances and train lengths do not justify any more blocks.
When the the dispatcher sets a route and gives authority, that also connects the correct throttle to the block ahead and provides the correct signal aspects.
Nothing would be gained with an ABS overlay.
ATLANTIC CENTRAL Interlocking signals do indicate both permission and occupancy.
Interlocking signals do indicate both permission and occupancy.
No. Not occupancy. Not for the interlocking, and not for anything a train encounters after the interlocking. That is all ABS.
Note that the indications projecting outwards from the interlocking don't change from just before a train enters, and after it HAS entered. Hence, no occupancy.
Dispatchers can give your clear indications all day long,...
The only way dispatchers can give you clear indications is with a CTC panel, and that ALWAYS includes ABS.
but if another train is still in the block, or a non CTC turnout is set wrong, your signal will stay red.
If you are talking about an interlocking, as you say you are, then yes, you get a red if your route is not cleared. And until the tower operator aligns your route, it will stay red whether or not "another train is still in the block". It will only change for YOU if the tower operator resets the interlocking.
You can point out, reasonably, that the tower operator can examine his interlocking and HE is the occupancy detector. But note that he is NOT the occupancy detector for the trackage AFTER the interlocking. So, without ABS, he can direct your exiting train right into another train that he has holding for a red until you clear.
To deal with THAT, you have to have him also responsible for tracking trains outside his interlocking, which isn't his job.
7j43k ATLANTIC CENTRAL Interlockings really are the heart of how trains are controlled on the big railroads. And, just my opinion, it is the only part of signaling we really need on our model layouts. We don't have long runs of track thru the country side where multiple trains will follow each other for dozens or even 100 miles. We are lucky if we have a scale mile between logical junctions/interlockings/control points. By that reasoning, the "railroad" will be a collection of interconnected interlockings, with negligible connecting trackage. So a train leaves terminal A, and has a pre-determined route through the appropriate interlockings, until it reaches terminal B. Since each interlocking will not allow opposing movements, all opposing trains will be stopped by red signals. Once the train is through the collection of interlockings, another train may be run through the collection. In a sense, it's like one BIG interlocking, allowing only one train through at a time. You CAN run multiple trains through this system at the same time IF you can guarantee each route is separate from all the others. What you cannot do is guarantee that each train will be protected by a red signal at each end of its block--ABS. Interlockings don't indicate occupancy; they indicate permission. So only one train can occupy any route at any particular time. You can't send out a following train until the previous one has arrived. Ed
ATLANTIC CENTRAL Interlockings really are the heart of how trains are controlled on the big railroads. And, just my opinion, it is the only part of signaling we really need on our model layouts. We don't have long runs of track thru the country side where multiple trains will follow each other for dozens or even 100 miles. We are lucky if we have a scale mile between logical junctions/interlockings/control points.
Interlockings really are the heart of how trains are controlled on the big railroads.
And, just my opinion, it is the only part of signaling we really need on our model layouts.
We don't have long runs of track thru the country side where multiple trains will follow each other for dozens or even 100 miles.
We are lucky if we have a scale mile between logical junctions/interlockings/control points.
By that reasoning, the "railroad" will be a collection of interconnected interlockings, with negligible connecting trackage.
So a train leaves terminal A, and has a pre-determined route through the appropriate interlockings, until it reaches terminal B. Since each interlocking will not allow opposing movements, all opposing trains will be stopped by red signals. Once the train is through the collection of interlockings, another train may be run through the collection.
In a sense, it's like one BIG interlocking, allowing only one train through at a time.
You CAN run multiple trains through this system at the same time IF you can guarantee each route is separate from all the others.
What you cannot do is guarantee that each train will be protected by a red signal at each end of its block--ABS. Interlockings don't indicate occupancy; they indicate permission.
So only one train can occupy any route at any particular time. You can't send out a following train until the previous one has arrived.
Interlocking signals do indicate both permission and occupancy. Dispatchers can give your clear indications all day long, but if another train is still in the block, or a non CTC turnout is set wrong, your signal will stay red.
If your layout is big enough and your trains short enough, sure you can make the whole thing work like the prototype.
Single track or double track makes a big difference too.
Out of time now, more later.
richhotrain I couldn't resist. The book sounded way too interesting, so I bought it and Amazon delivered it yesterday. I plan to start reading it in detail today, but I did spend some time yesterday skimming through the book. It is 144 pages, divided into seven chapters, the titles of which I can paraphrase as follows: Prototype History Signal System Basics Manual Signal System Train Detection Wiring Signals on a Layout Interlocking Controlled Signals There is no rehash of old articles. This is all new original content written by the author. The text is augmented by black and white photos from the past, current color photos of the prototype as well as selected layouts, and a host of diagrams and drawings which appear to be excellent. One chapter that particularly interests me is the chapter on interlocking, something of which I know very little. The chapter distinguishes between and compares interlocking and junctions. The coverage is excellent on controlling and signalling interlocks. There is a brief Bibliography at the end of the book that cites books, periodicals and websites. There is also an Appendix on scratchbuilding mechanical interlocking and signal controls. This Appendix can effectively be considered Chapter 8. One thing that really caught my attention is the scratchbuilding of a miniature working interlocking machine, complete with miniature levers. Let me at it! Rich
I couldn't resist. The book sounded way too interesting, so I bought it and Amazon delivered it yesterday.
I plan to start reading it in detail today, but I did spend some time yesterday skimming through the book.
It is 144 pages, divided into seven chapters, the titles of which I can paraphrase as follows:
There is no rehash of old articles. This is all new original content written by the author. The text is augmented by black and white photos from the past, current color photos of the prototype as well as selected layouts, and a host of diagrams and drawings which appear to be excellent.
One chapter that particularly interests me is the chapter on interlocking, something of which I know very little. The chapter distinguishes between and compares interlocking and junctions. The coverage is excellent on controlling and signalling interlocks.
There is a brief Bibliography at the end of the book that cites books, periodicals and websites. There is also an Appendix on scratchbuilding mechanical interlocking and signal controls. This Appendix can effectively be considered Chapter 8. One thing that really caught my attention is the scratchbuilding of a miniature working interlocking machine, complete with miniature levers. Let me at it!
Rich
Anyway, I'm getting ready to order a copy to add to my extensive library on signaling....
richhotrainOne thing that really caught my attention is the scratchbuilding of a miniature working interlocking machine, complete with miniature levers. Let me at it!
Yes I thought the same thing! What a cool idea. Would be a lot of fun to have on a layout - for people without little kids and/or cats.... (Ed)
Andy
--------------------------------------------------------------------------------------------
Milwaukee native modeling the Milwaukee Road in 1950's Milwaukee.
https://www.flickr.com/photos/196857529@N03/
I've started reading it now, I'm a couple of chapters in. Lots of good information.
BTW the Appendix on building a miniature mechanical interlocking says it's from Jeff Wilson's The Model Railroaders Guide to Junctions, although it seems to me it had been an MR article at some point in the past too(?)
I just picked up that new Kalmbach book and am looking forward to reading it. I'm just about through a 1979 General Railway Signal 'Elements of Railway Signaling' book that is chock-full of cool information on the relay designs for wayside signals, cab signals, ATC, ATS, crossing flashers/gates, and more. Fun stuff.
7j43k The photo on the cover reminds me of something: Signal heads are not installed aligned with the adjacent track. They are aligned with the line of sight of approaching operators. Here's a pretty good example: https://www.wsws.org/asset/8d405fd9-728d-48e2-b581-e7d699f91a2J/rail-signal.jpg?rendition=image480 It's pretty unusual to find signals placed on curves on the prototype. I had a bit of trouble finding the above example. My recollection is that there's some on the Cajon Pass route. Winding trackage STILL needs to be signaled. But. Model railroads have a LOT more curved track than real railroads. So you might be placing a signal on a curve. As I have. Ed
The photo on the cover reminds me of something:
Signal heads are not installed aligned with the adjacent track. They are aligned with the line of sight of approaching operators.
Here's a pretty good example:
https://www.wsws.org/asset/8d405fd9-728d-48e2-b581-e7d699f91a2J/rail-signal.jpg?rendition=image480
It's pretty unusual to find signals placed on curves on the prototype. I had a bit of trouble finding the above example. My recollection is that there's some on the Cajon Pass route. Winding trackage STILL needs to be signaled.
But.
Model railroads have a LOT more curved track than real railroads. So you might be placing a signal on a curve. As I have.
Yes, rare, but I have seen prototype signals set farther away from the track on curves and angled to be seen from farther back down the line.
I think I might have to spring for a copy of this.
wjstixIn any case, this is NOT a collection of prior articles
Thank you for the answer.
I am going to buy the book.
-Kevin
Living the dream.
Someone used to sell those little levers but I think they are out of business. I did watch the video, Ron does a good job. His day job is a pastor.
Henry
COB Potomac & Northern
Shenandoah Valley
Alton Junction
BigDaddy If it was a compendium of articles, the author would have to give all the previous authors credit and perhaps royalties.
If it was a compendium of articles, the author would have to give all the previous authors credit and perhaps royalties.
So I watched the video, which is hard for me, and I read the info on his site.
I might pick up a copy just because it looks well done, but I seriously doubt it would change the way I build my CTC or signaling systems.
After reading everything about signals published in MR or RMC, and reading a number of other books and articles, I developed a signal system largely based on Bruce Chubb's original relay based signal system before he developed CMRI.
But, I took the advice of several other modelers, and I only model the interlocking/control point signals, and I do not fully model "yellow" or "approach" aspects. and I do not model any ABS functions between control points.
So, an "absolute stop" at one interlocking does not produce a "yellow" at the next interlocking behind. It does produce a yellow "approach" for an approach signal half way between the two interlockings.
And of course "yellow" indications are used at interlockings for speed restricted routes.
My CTC panel is simplified, no full blown CTC levers with multiple actions required.
Lighted pushbuttons set routes, and assign routes to throttles, which clears the signal if occupancy is clear. So moving a train along to the next block or two only requires pushing a few buttons, wihch lightup to show routes and assignments.
All the logic is done with detectors and relays, the same relays that operate the turnouts have extra contacts for the signal logic.
A simple logic chain of relay contacts:
Detector relay contact - in the block available?
Turnout relay or relays - is the route set conecting the two blocks?
Authority test - are the connected blocks set the the same throttle?
When all three test yes, the signal goes "clear", other wise it is "stop"
The dispatcher sees detection indications, and so will operators around the layout on several "statis panels". Hard or imposible to see signals will be repeated on panels, impossible to see signal heads will not be lit but will just apear on panels - kind of like cab signals.
IF there is no dispatcher on duty, the layout can be run locally from the local towers, and the signals will act as if the dispatcher was controlling them.
Since the Railroad is freelance, I decided on simple color light signals, no targets, no position lights.
Detection is Dallee inductive current sensing detectors, with their carrier system for stopped trains (this is a DC layout). All trains have lighted/resistance cabooses/last cars, there are no resistance wheel sets as the Dallee system is not that sensitive. But being inductive it is totally isolated from the propulsion power.
No need for computers, processors, sensors. Just relays clacking, just like in 1954.......
Another good point. Abeles also says on his blog that the book is based on his own experience with his layout.
As I said earlier, I haven't had a chance to sit down and read the book cover to cover, but it doesn't look to me like it's one of those 'collection of previous articles' books. It starts out talking about the history of prototype signals, then adding a manual signal system, then discusses the various detection methods like infrared, track detection, etc., then how to integrate the detection system you've set up into a computer-based signal system using JMRI. Only thing I saw that looked like a previous MR article was the last chapter (actually an apendix), which went over building a working interlocking system using scaled-down manual interlocking levers.
Ed,
You've done a superb job on your signals, especially the PRR ones you show here. While I run modern, I'm still a fan of Pennsy signals and had some custom made for me to install on the layout.
I agree there's a shortage of signals out there. Nice you have the skillset to make them. I enjoy installing them and having them working as I dispatch trains so the signals are controlled manually at the panels. Building them from a kit or 3D is above my pay grade.
Please keep posting photos, I enjoy seeing your work!
Neal
The signals look good Ed!
Dave
I'm just a dude with a bad back having a lot of fun with model trains, and finally building a layout!
Having an informative book is a good step in the right direction but if there aren't decent, affordable signals and hardware available the budding modeler is going to be in a tough spot.
Over the years I've gathered much of what I anticipated I would need to have a fair representation of signals. Most of what I have is no longer available today or if it is it commands a budget-wrecking price.
One of my sources for reasonable and good looking signals in HO is products from Oregon Railway Supply. In fact I'm currently assembling PRR position light signals for a four track bridge with eight heads on it.
IMG_2679 by Edmund, on Flickr
This is another Oregon bridge kit:
IMG_2666 by Edmund, on Flickr
LED_Signal-head-1 by Edmund, on Flickr
Another bridge was finished up last week with three-color SMD heads.
Signal_DB-west by Edmund, on Flickr
My favorite control is the Signal Animator from Logic Rail. For about $25 each you can have three indications and detection through IR LEDs imbedded between the ties.
This is a "temporary" (needing clean-up) assembly for the three-head, two direction bridge. I also tie these into the turnout and also proviide a direction-of-traffic switch. Nothing worse than seeing a green signal with an opposing train approaching.
IMG_6236 by Edmund, on Flickr
Other manufacturers, NJ International, BLMA, Model Memories, Atlas, Tomar have spotty availability at best. Anyone just starting out may have a tough time rounding up supplies or they may have to hone scratchbuilding skills. Shapeways may be a good source for heads and trackside structures, which is a plus.
IMG_2728 by Edmund, on Flickr
Signal_11-21oh by Edmund, on Flickr
NYC_6103_GP20b by Edmund, on Flickr
IMG_8220 by Edmund, on Flickr
Good Luck, Ed
Either way (rehash of old articles, or new material), there must be enough people out there exploring this topic for the first time that the book is now sold out on Kalmbach's site.
One clue that makes me think this is new material: the author notes on his blog that it is his first book. Ever. And it's based on a lot of experience and research. Sounds to me like it's not a book that simply collates a bunch of old articles.