I'm at the wiring stage on my freelance road and I have two questions about signalling.
1. My line is single track and passing sidings under ATSF 'Flat Pair' ABS, using signals similar to the B&O CPL's just with a NYC style (triangle light) head in place of the B&O position light. What is the best signal system and wiring setup to simulate this?
2. the portion of my layout I have built currently has a double track crossing at the end of one of my passing sidings. I've drawn a rough diagram of the area: http://img12.imageshack.us/img12/4342/jctctysignalplan.png The arrows on the red dots show signal direction. Since the cross tracks are only about 2' long, is the a way that I can use the signals to simulate crossing trains that is seperate from the track detection circuits?
To make the diagram visible:
https://tstage9.wixsite.com/nyc-modeling
Time...It marches on...without ever turning around to see if anyone is even keeping in step.
First, there are many different signal systems available. Some are for DC and some are for DCC. Usually they are not interchangeable with each other. Also, working signal systems are not cheap.
Here is a link that should help you as you start:
http://waynes-trains.com/site/Signals/Model-Railroad-Signaling.html
My layout is DCC and I use the Digitrax signal boards. It does take a computer to operate it.
The signals that you describe using sound like the type G signal. Three color lights in a triangle. If you are modeling a particular Railroad, you should find out what type of signals they use, and then use that type. If you are building a freelance RR, you can use anything you want.
As to your second question, yes, there is a way to change the signals that simulate usage of that particular track. However, different signal systems have different ways of doing it, so you need to select your signal system first.
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.
I'm planning on using a Digitrax system for train control, but I hadn't yet looked at their signal system. Looking at the signal mast kit, it seems it only has connections for two 3 light signal heads. My signal design is the Type G with the B&O style markers, which means a signal can have up to 9 separate lights. Can the Digitrax system handle this? I'm not sure that there is a location on my layout that requires a signal with all markers, but I assume that since each marker lights up for different aspects, a set of 9 contacts would be needed 'in the system' even if a specific signal doesn't need all of the markers.
You can do nearly anythign imaginable with the SE8C, because it is computer-driven. There is no signal 'logic' on the board, it is purely a bunch of signal head outputs drive my switch commands. There are 2 addresses per head, to give a total of 4 possible aspects per head. The examples on their web site and int he manula showing a dual head on the point side of a turnout, a dwarf byt he siding, and a single head mast ont he main track at the frig side is just aht - an example, It does NOT have to be wired like that.
On our club layout we have all sorts of signals with multiple heads, all Digitrax driven. Some use the SE8C< some use the Tower Controller from RR-CirKits. One complex example is a single dual-mast tower which has 3x heads per mast for a complex interlocking. ANother mast has 2 heads capable of showing 4 aspects each, plus a third head with just red and yellow. The signal system is all controleld via JMRI, and follows current contemporary NORAC rules. All the different flashing aspects drive me nuts, as I personally model the mid 50's and things were a bit simpler, but the system was designed and programmed by the head of the signaling department of a local regional railroad, so he knows what he is doing.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
To further refine my question on the signal mast kits; All of my signals have 1 'head', being a normal Tri-Light signal and up to six marker lights, arranged B&O style. Each signal, if equipped with all 6 markers, can display 15 aspects, controlling main and diverging routes. The Digitrax signal mast kit seems to only have contacts for 6 lights. After looking over my track setup, I know at least 3 signals will need 8 lights (the 3 main lights and 5 markers). What I'm wondering is could I use the signal mast kits or do I need to use the SE8C or a similar device and have each individual signal be considered as multiple 'heads'
You should be able to treat the markers, or collections thereof, as heads. 3 lights per head, so for up to 6 markers youw ould need 2 heads, plus another head for the actual CPL. If you are unable to groupt he valid ocmbination of the markers this way, you may have to use additional heads. A little workign out of the logic of valis combinations might find you soem creative ways to reduce the number of signal head outputs needed to acheve all valid combinations for a given control point, the idea being sort of liek a diode matrix for turnouts in reverse. FOr the complication of figuring out the logic, plus the cost of some low current diodes (pennies a piece), you cna reduce the number of (relatively expensive) signal head outputs needed. Signaling is not esy to implement if fully following prototyical practive, but the results are outstanding. EVERYONE remarks on hwo the signals change, even if they don;t understand all the indications.
Also, I am glad I don;t model the B&O, the most complex B&O signal I was aware of had just a marker above and one below the main signal. I was not aware they got so complex as to have SIX markers. Yikes! The Tower Control might be you friend here - use a combination of direct connections (for the markers) plus the multiplexor boards (for the main heads). Or a combination of SE8C's for th emain heads plus Tower Controller outputs for the markers. The SE8C by itself doesn;t really lend itself to single outputs, since all outputs are multiplexed. Teh Tower Controlelr allows access to 64 individual outputs that can be seton or off, but there are also multiplexor boards to produce outputs for multiple heads from one set of output bits. Any combination of boards plus direct access to the output bits is allowed on the Tower Controller.
If I'm understanding this right, I'm going to need to have a Tower Controller for each passing siding end (only place with complicated signals for the most part, intermediates are only tri-lights and 1 marker), and each "signal" would have a 4ASD-4 controlling it with commands from JMRI. Is that correct?
Our Club is modeled after the B&O and I built quite a few of the signals and installed the Digitrax signal system.
You are correct in that a signal with all the markers is a lot of lights. However, what you should do is figure out how many you can actually use. With a three aspect main head and then six additional markers, gives you about 18 overall aspects. (In the B&O system, only one marker at a time could be lit.) Can you use them all? Is your RR that large?
I narrowed our club signals down to the main head with three aspects and two markers. I can have an upper marker on (green second head), a lower marker on (yellow second head), or both markers off (red second head not connected). The primary reason that I did this is that you will never get a bunch of modelers to learn every aspect possible that a B&O signal can be set for. Most of them don't know the difference between the upper marker and the lower marker and what they indicate as simple is that is now. Also, on the B&O system, the markers can be considered as modifiers to what the main head is indicating. This type of thinking tends to make things on the signal simpler.
So think about what aspects you really need your signals to indicate and how you can do it with less lights.
Oh, and one more thing. How many wires do you think you can get in a 3/32 diameter brass tube that is going to be your signal mast? About the max I have gotten in there was six.
Before you start to figure out what JMRI can do, you need to decide on all the aspects you are going to be using. And for JMRI help, you should sign up at Yahoo Groups and join the JMRI user group. You will get better and more specific help for JMRI there.
Reevaluating my setup, I probably could get away with using 3 markers - top middle, bottom middle, and bottom left - for my siding approach signals. All the other signals are either top middle marker only or bottom middle only.
After re-reading the initial post - since you are freelancing, perhaps a rethink on the B&O signals idea? Since you say you want to use Type G heads instead of the B&O CPLs, I'm not sure if anyone who used type G's actually used B&O style indicators with them, preferring multiple heads for alternate aspects - much like we have in some palces (Readign also used type G heads). One basic junction has 2 full G heads plus a 2 aspect head with just the red and yellow for the more restricting indications. Or really go out there - at some locations on the Reading,t eh G head was combines with a semaphore! No kidding.
As for part 2, I've been mentionign 'computer' all along, because thsi is WAY simpler and more flexible int he long run than those signal boards that cna operate without a computer. Just read the manual for the Team Digital boards and look at the CV settings to build the logic int he board. Of ocurse, JMRI makes it easier, but wait, now we have a computer involved ANYWAY... So, with a computer invovled, it wouldn;t be terribly difficult to have a random, or scheduled, event where a train on the connecting railroad would be coming through the crossing, dropping the signals for your line. This can get even more complicated, depending on which line has priority. If it's yours, then if a train happens to be approachign on your lien it coudl lock out the other one so your trains are cleared through, but if the scheduled time ticks off and you have no trains within two blocks, it will clear for the crossing line and any trains on your line arriving after this will have to wait. I'd include signals for the other line there as well, just for the effect of them changing. Everythign but the actual train moving on the rails. Definitely an interest adder for operations.
I think I'll probably end up with straight Type G's now after realizing how redundant the markers are for my situation. The idea of using the computer interface to generate the cross traffic is one that for some reason had never come to my mind.