I`d like to know what type of detection you use for your signal systems. I have 10 blocks to create and am not too crazy about cutting gaps in my track and was leaning to phototransistor detection that would not require me to cut gaps.. If you use current detection do you gap both rails or just one. I have a 1 booster layout with 4 power districts ( which I did cut gaps in both rails for and is the reason I am not to inclined to cut more gaps) HO DCC and will be using ISS position light signals. If I decide on optical detection I will probably use Rob Paisleys` PNP 3 boards the 4 circuit type.
Thanks
Bob D As long as you surface as many times as you dive you`ll be alive to read these posts.
For current detection you have to gap 1 rail. If you have multiple DCC boosters then you still need the double gaps between sections pwoered by different boosters (or circuit breakers, for that matter).
Detectors like phototransistors are 'point' detection - they can tell when a train is at a certain specific point and that's it. With multiple sensors and some additional logic they can tell when a train enters and exists a block, but will become confused if you run a train into a block and then it never leaves - ie, you take the loco off the track and don't return it to the same block it just left. Inisitalization of the current state of occupation is tough - like when you shut everything down for the night and go back the next day. This type of detection does have its place though - things like activating a turnout for automatic route control or control of a reverse loop, or activating an effect.
Current detection is an 'area' detection - as long as one detected wheelset is within the block, it can show the block occupied. It automatically resets when you power the layout back on. What it can't do is tell you that the train is at the beginning, middle, or end of the block - just that there is somethign in that block.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
As Randy has mentioned, you need double gaps between power districts and should have some type of opti-coupled interface between detection circuits is different power districts.
Where I will differ with Randy is on optical detection systems. I have 8 blocks of ABS detection that uses IR detection. The logic detects a train entering the block and keeps it detected for about 2 minutes, even if I pick up the engine and walk away! There is a detector at each end of the block and it operates like an entrance/exit system. The advantage is that I have to do 'nothing' to my DCC track system, I do not need resistor wheel sets for detection, and the entire system can be a 'stand-alone' operation. The advantage of IR over photo-cells is low light condition do not affect the system. I bought the parts from South Bend Signal Company:
http://www.sbsignal.com/
Jim
Modeling BNSF and Milwaukee Road in SW Wisconsin
Right - the 'extra logic'. In your case, the system waits a time delay OR the 'exit' sensor is blcoked - correct? In that case, what happens when a train enters the block, and just sits there, but is not blockign either the entry or exist sensor? Does the block go unoccupied after the delay expires, even though a train is still sitting there? My version of the 'logic' would have the block perpetually remain occupied if the 'exit' sensor never got blocked - say you take the loco to tthe bench, then put it back in some other block.
You just need to be aware of and deal with these possible conditions. There definitely are advantages to not having to modify any rolling stock. Even if you already have all metal wheels, as I do, adding resistors is not the easiest thing to do. Buying wheelsets with resistors already installed is possible but they are NOT cheap - mainly due to the effort involved in making them.
My basement layout uses optical detection. It also has the 2 minute time out. After 2 minutes the block will show clear unless a train happens to be parked over a sensor.
There are a couple of things I don't like about it. 1. In low light conditions the sensors will be constantly tripped. 2. If the light source is not directly over the sensors, tank cars will cause a falst trip since they are rounded and allow light to hit the sensors from an angle.
If I ever get around to it I'll switch to IR or current detection. The gaps aren't hard to do and the extra time invested is well worth it in the long run.
Springfield PA
Jim , Randy said that I needed to double gap the rails between BOOSTER NOT POWER. districts is that what you meant? I have only 1 booster on my layout. Also Jim did you mean that you used SBS detector/signal boards or just their IR sensors and someone elses signal driver boards.
I like Rob Paisleys` 4 photo detector signal driver boards with delayed release & bi polar output. This board according to his literature eleminates the signal flicker between cars and is very reasonably priced built up and will work with common cathode signals that ISS makes. My biggest problem is making a decision on the type of detection. Jim do you have to cut ties to fit your IR units in?
I currently have 1 of Bruce Chubbs` DCCOD boards & 10 TCH technology current boards which are close to the DCCOD boards with a few exceptions + a motherboard that I would put up on HO Yardsale if I go optically.
Those may reduce flicker from between trains but the lighting and tank car issue will probably still be there.
Since I have never implemented a current detection system with DCC, all I can go on is the info I got from two vendors concerning the power districts. Power districts have double gapping as well(at least with my Digitrax PM42's). I did have two blocks of current detection on my DC layout back in the 'older' days.
As far as the concern that after the 2 mintute time-out, the block will clear. I have all turnouts wired so that if a turnout is aligned to it's reverse route, the signal on each end of the block will be knocked down and stay red until the turnout is aligned for the through route. Most trains get between blocks within 2 minutes, so the only trains that would be stopping between sensors are locals doing 'short work'.
I have been using the detection boards from SBS(about $30/block) and their signals. A 'combo' deal of a board and 2 signals is something like $56 - pretty cheap for signaling. The other advantage of IR detection is not trying to fit the 3 or 4 detection wheel sets available into different truck side frames. A lot of folks only fit the caboose or last car with detector wheel sets. And that may leaves a 'gap' in the 'fail safe' as well. If this was 'real world' - I would use current detection. For my trains, I can live with it.
I use current detection. Specifically the Digitrax BDL-168. Even if you are not going to use the Digitrax system, you may want to look at the manual which you can download. It will give you some insight to current detection and how to gap it.
Elmer.
The above is my opinion, from an active and experienced Model Railroader in N scale and HO since 1961.
(Modeling Freelance, Eastern US, HO scale, in 1962, with NCE DCC for locomotive control and a stand alone LocoNet for block detection and signals.) http://waynes-trains.com/ at home, and N scale at the Club.
Sorry I confused you, thought it was pretty clear. Gap ONE rail between detection blocks. Gap TWO rails between booster power districts. Since you only have 1 booster and no circuit breakers, it would be all single gaps for each detection section. You ALWAYS have to gap both rails between sections controlled by different boosters or different circuit breakers, even if there is no detection used.
Thanks all. The answers clear up a few misconceptions.
jrbernier I have been using the detection boards from SBS(about $30/block) and their signals. A 'combo' deal of a board and 2 signals is something like $56 - pretty cheap for signaling. The other advantage of IR detection is not trying to fit the 3 or 4 detection wheel sets available into different truck side frames. A lot of folks only fit the caboose or last car with detector wheel sets. And that may leaves a 'gap' in the 'fail safe' as well. If this was 'real world' - I would use current detection. For my trains, I can live with it.
Current detection on every block (to detect engines) and IR detectors at block boundaries (to detect cars strung out over block boundaries - you'll especially want this for short signal blocks like control points) will fairly reliably cover about 99% of cases. The only thing it can't detect (which none of the other solutions can either) is a car or cut of cars with no engines attached and no resistor wheelsets parked in the middle of the block between IR detectors. If all cabooses have lights or resistor wheels, then that will cover about 90% of all return-to-train movements or locals leaving their train on a mainline while picking up or setting out cars. (as the standing caboose will still be detected, even if the engines are off on the siding or industrial spur) Still won't work for modern cabooseless operations though wihout putting at least one resistor on every car.
Chris van der Heide
My Algoma Central Railway Modeling Blog
Hello Model RR Forum.
I have a question about how the signals are used, where they are placed and which signals are used on a railroad - typically. I have bought several SIGNALS books and they dont tell me what I need to know.
What does YELLOW mean? Does it mean that the block ahead is clear, but the one in front of that is blocked (RED)? Where are these signals placed on a real railroad - and where have you all placed them on your layouts? Do most of you use RED and GREEN signals or 3 color (adding YELLOW)?
I am not a total stickler, but I want to make sense on my layout (which is nort yet built).
Thanks!
--Calidus
Southeast Iowa
HO & 1:24 scale enthusiast
Here's some quick reading on it.
http://www.fact-index.com/r/ra/railway_signaling.html#Signals
Signaling can get deep quite fast. There are two basic types of signals:
Permissive - These are the block signals that have a mile post number plate below the signal. They generally operate automatically and show these basic aspects:
Restricting - These are 'Stop & Stay' signals(have no mile post plate below the signal) and many times there are 2or 3 signals mounted one above the other on the signal mast. These are seen at junctions/crossings/crossovers - many times in terminal areas.
I have 8 blocks of signaling on my layout - pretty much a simple ABS(Automatic Block Signal) system with a single RR crossing. The crossing is non-functional and is just wired like the rest of the block signals. And yes, I have the 3 aspect colors. They change from green-yellow-red as the trains move over the layout.
Thanks HamltnBlue for the link - I shall study it thoroughly.
And Thank You Jim for your kind and detailed reply - it is just the type of kindness I have always received from fellow Model Railroaders - I thought I was correct about the YELLOW signal - just needed someone more knowledgeable than I to confirm it.
The link that HamltnBlue sent might have this information - but Jim can you answer the question of WHERE you place the signal (next to switch?) or if a long section of track is separated into more than 1 signal sections? And a following Q is what is used to designate the end of the block - so the engineer can know how far he has to stop? Perhaps I need to graph this all out? I think real working signals can be a great asset/focal point to a layout - but we all like to see trains run!
Thanks again!
Calidus,
Normally in ABS territory the block signals are placed just before the siding switch(both facing the points and looking away). In between sidings, the signal pairs are placed about 2 miles apart(average). The spacing is determined by track speed/stopping distance for a train. Sometime you will see single block signal in the field. This is usually part of a safety 'overlap' scheme, or placed ahead of a curve for improved sighting. The signal marks the end of the block - there are insulated rail joiners(sort of) that electrically seperate the blocks.
Thanks Jim & HamltnBlue.
That makes sense - I guess I just need to map out my layout and then determine the blocks and add signals as necessary. Your help has been great. Thanks again.
- Calidus
jrbernier Signaling can get deep quite fast. There are two basic types of signals: Permissive - These are the block signals that have a mile post number plate below the signal. They generally operate automatically and show these basic aspects: Green - track is clear two signal blocks ahead. Yellow - The block after the one you are entering is occupied or there is a restricting signal at the beginning of the next block(you had this one figured out). Red - Block occupied - move at restricted speed past this signal after stopping. Restricting - These are 'Stop & Stay' signals(have no mile post plate below the signal) and many times there are 2or 3 signals mounted one above the other on the signal mast. These are seen at junctions/crossings/crossovers - many times in terminal areas. I have 8 blocks of signaling on my layout - pretty much a simple ABS(Automatic Block Signal) system with a single RR crossing. The crossing is non-functional and is just wired like the rest of the block signals. And yes, I have the 3 aspect colors. They change from green-yellow-red as the trains move over the layout. Jim
Sorta, although this is completely over-simplified and doesn't even scratch the surface. What is said above is generally pretty accurate, although I would note the following:
- You have "Permissive" and "Absolute" indications. I would NOT use Jim's word "Restricting" as that is a specific signal indication (and one that is actually "Permissive" by the way). The distinctive between "Absolute" and "Permissive" really only applies to the Stop indication. "Permissive" refers to a signal where red means "Stop and Proceed [at Restricted Speed]", rather than "Stop".
- the whole number plate vs. no number plate is by no means universal, but is a common rule for western US railways. (for example there is absolutely no such distinction for signals in the rules used by Canadian railways. "Stop and Proceed" vs. "Stop" signals are differentiated in a completely different fashion. A signal may have an identifying number plate, but it has no bearing on the signal indication). Really need to use the rules for your railway, or at least choose a particular set of rules (even if that's your own simplified system) and be consistent.
- Each signal gives information about what to do now, and what the next signal is displaying - not the next block. It may seem like a subtle distinction but it's important because the next signal may not be a simple block signal (or there may not be a next signal), and there are more possibilties than simply "The track ahead is unoccupied, go ahead"
- You will also see signals with more than a single head, and the _combination_ of lights on each head make up different "indications" * (the meaning/instruction that the signal provides). There are literally dozens of possible combinations and arrangements. And almost every major railway had a different rulebook, or even different styles of signals.
Signalling gets very complicated very quickly, but there are some very logical rules to it all.
For example, the following link includes the signal indications from the 1962 Uniform Code of Operating Rules, which was a Canadian rulebook. This rulebook was replaced in the early 1990s by a new one, but most of the listed rules still exist with different numbers. If you count them all you see there are 17 different indications represented by 60 different possible light combinations ("aspects").
http://www.wrmrc.ca/ucor/281.html
If you're just doing a simple block system on a straight piece of main line, with no switches and fancy trackage, then all you need is "Clear" (Proceed), "Approach" (Proceed prepared to stop at next signal) and "Stop and Proceed" (Stop, then proceed at restricted speed). You can see that this is exactly as Jim described above. Once you introduce junctions, crossovers or sidings and multiple routes, then things get more involved.
Always remember, the most important thing in any serious signal discussion (indeed, just about any operations question) is the rules in effect on the particular railway you're interested in. In any rulebook, you'll see a pretty similar list of indication names, but the aspects (colour combinations) will be totally different. You won't actually want to use the list above if you're modelling American roads, because while some indications are the same, there are differences to be found between different rule sets.
For example, "Restricting", which is what you'll get if lined into a siding or yard that is not part of the signalled track, is Red over Yellow in the above rules, but on other railways may be Red over Lunar (white), Red over Flashing Red, or something else entirely. The above rules make no distinction between signals with or without number plates (as some western US railroads do) and include several aspects with staggered heads on the signal mast (which many US railroads do not).
* Signal terminology basics:
Aspect - the particular combination of coloured lights.
Indication - the information conveyed by the signal. Each aspect corresponds to a single indication. Each indication may have multiple possible aspects.