Paul, yesterday you needed 50, today you need over 100. Yikes! Do you really need that many? What is your budget for the signals? How much are you willing to spend?
Rich
How large is your layout? How are you using these: CTC, ABS, something else?
That seems like a lot of signals.
That is a lot of signals. My layout is pretty small and I only have a total of 8 blocks on my mainline requiring 8 signals for each direction. I don’t use signals for my hidden tracks so that drops the actual signal count to a total of 14.
Have you figured out how you are going to control them, that could get into big bucks fast. I built up a Rob Paisley controller many years ago and I’m currently working on a Arduino MEGA controller as a replacement. The Rob Paisley controller works great but the Arduino MEGA is much smaller and programmable.
A single MEGA ($15) will handle 16 three color heads and to make it a clean unit it takes two expansion boards for the wiring. I think I have about $28 into my Arduino controller project for 14 signal blocks or about $2 per signal including IR detectors but not incluging the 5 volt power supply. I don’t remember what the Rob Paisley controller cost to make but I’m pretty sure it was closer to $8 a pop back in the 90s using his occupancy detectors.
If your not to choosy and can use under the track detectors the Arduino IR detectors work very good and the price is right too at under 50¢ each (50 for $20) from China. I don’t like current detection and the IR detectors work perfec
I want to put a signal at each leg of every turnout that touches the mainline. My mainline touches 11 single turnouts (3 signals ea), 1 single crossover (4 signals ea) and 7 double crossovers (4 signals ea). That means I need 65 signal mast.
YIKES….I forgot to change my thinking on number of signal heads when I changed from dual searchlight heads over to the single head (three lights). That means I do not need “over 100”. Thanks for raising a flag. I was in the process of shopping for mast and you just saved me a lot of money.
This helps me a lot in locating the signals I need. I wanted to order all I needed (plus a few spares) at one time from a single source to make sure they were all made alike. I found the same mast on eBay from several sellers (including the same seller as on Amazon) for a few dollars less. I also looked at AliExpress and they had the exact same item plus they offered a 10% discount but they did not have enough in stock to invoke the discount level. Problem I found was that none had enough to match my (original) needs. Now with a lower number, I might have more luck.
I also discovered that the Amazon signal light is 150 scale (Japan stuff)…not 160 for N scale. That makes it a tiny bit taller and larger than preferred but is workable. I did find another very similar light at the right 160 scale but I need to investigate it further.
I am still studing control options. Right now I anticipate using LCC with RR-CirKits components but I have a guy trying to convince me to use Arduinos as an alternatate. Trying to look at cost and effort required for both approaches and see what works best for me. A player here is that I have a bunch of lighted building I have not
Why?
Why not? I want red and green for the ususal reasons and I want yellow to signal turnout position. Not trying to replicate 1:1 signal aspects. I am a solo operator and I want something that is useful to ME. Real simply signals. With 3 lights, Green means next block is unoccupied, yellow shows turnout position and red means STOP (turnout wrong or next block occupied). Only 1 or 2 lights lit at a time.
You can do whatever you want, of course, but you seemed to be worried about the expense. A sea of signals where they wouldn’t be typically used in real life wouldn’t look realistic to me – but it’s not my layout.
Because the real railroad wouldn’t? Because it’s expensive? Because there are cheaper ways to report a turnout’s position on the fascia?
How are you planning on handling the logic and detection required? Detection is expensive for all the small blocks you will have (by having so many signals). That’s more hardware for each block besides the signals. Plus resistance wheelsets.
That signal logic is non-standard, so I think you’d need to add programming to off-the-shelf JMRI to achieve that. (But I’m no JMRI expert.)
The real-life railroad signals just what is necessary where it is necessary. We can learn from that.
Good luck.
I would have chimed in on this thread earlier, but everyone kept calling the limited commercially available N scale signals ‘crappy’. Well, not every one, but all it did was make me feel depressed about my situation.
I’ve been modeling N for a pretty long while, and I am well aware of the limitations. I bought NJI two-head signal masts, about 40 or so. It was a conscious and considered decision. While I would not call them crappy, I will admit they are a little less than (what my doctor would call) optimum. But they function, and they are fairly well made. Not exactly cheap, but about $20 less than the Tomars, which I could not find anyhow.
I considered rolling my own, and I Iooked at kits and components (some of which are mentioned in this thread), but the time spent fabricating and assembling 40 or so weighed heavily against it. My hands are as steady and my eyes are as sharp (nearsighted) as anyone, but I doubt my results would be much better than the ‘crappy’ ones I bought, and there’s a good chance they’d not even be as good. Others might fare better.
Only my opinion, of course.
Robert
Paul,
I basically do the same thing. All the signals on the double track mainline are at the crossovers and sidings, but are also on the control panels. I use the two color like in the photo. You just have to glance at the control panels and see if you have all green signals for continuous running on both mainlines…running East/West. I use dual coil relays for that, along with powering the frogs, with normally open push buttons at the control panels, with walk around tethered hand helds. I’m a DC user and at the time the relays were 7.50 apiece…6.50 by the case. I needed 45. Tortoises were not even out yet! But don’t care…no problems at all with the relays, since the mid 80’s.
Good Luck on Your project! [:D]
Frank
It seems clear from the above statement, and other statements made in this thread by the OP, that he wants to closely simulate the Santa Fe signal protocol used in the 1970s.
Santa Fe did use searchlight signals in the 1970s along its mainline, so the OP is correct in that regard.
What remains to be decided is the number of signals used at turnouts, crossovers and double crossovers. The layout will have 11 turnouts, 1 crossover and 7 double crossovers. What we need is someone with knowledge of the Santa Fe signal protocol in the 1970s to chime in and indicate the number of signals that would be used by the Santa Fe in such a situation.
Rich
Actually, only the OP referred to certain signals as “crappy”.
Regarding the Tomar Industries signals, they are brass, a soft metal, and they are good looking and durable. They can be purchased directly from Tomar Industries, and the OP does agree that the Tomar searchlight signal is what he is looking for except for the price.
Rich
If so, it appears he’s already off by choosing to place signals at every minor turnout connecting to the mainline.
And which ATSF signal protocol? CTC? Block signals? Interlocking?
You have detection of equipment in the blocks, as the OP desires? What logic are you using for detection?
It’s easy to signal switchpoint position, and that can give a nice model signal solution (non-prototypical).
But that’s not what the OP describes.
And semaphores: Lee, TX, 1974
I linked this photo not to be a jerk, but to note that prototype signaling is complex and varied, even for one railroad. When someone says “I want prototypical signals”, they often don’t take into account (or don’t know) about all the variations.
The route (pun intended) I often recommend to clients is to instead think about what combination of cosmetic and non-prototypical-function signaling (like indicating switchpoint position) can look good and still not be too onerous to install and maintain.
To me, that’s a better place to start and build on rather than taking an overly ambitious (and non-prototypical) goal and trying to shoehorn it into the confines of a modest layout.
[Not to mention less expensive. Many modelers don’t realize the detection and logic implications of detecting many small blocks, but it’s a huge step in cost and complexity.]
Prototypical or not…that is what I use and it works for Me. I can jump on any of the mains with a switch engine with cars, while the two mains are running by themselves and keep track of the time I need to get into the industrial sidings, which is how it was set up for three operators. Set the mains back to green and switch out the industries. No industry is spotted from the main. You have to completely get off the mains to do any switching. I have a three cab set up(45 blocks) with 3 MRC 20’s. Would be a lot easier with DCC, but unfortunately, that was not out yet either when layout was built. Works fine for Me, all these yrs. It was built in removable sections, half of which have already been moved to My Grandsons basement. I will be 77 this yr., getting hard to get around, so I made the decision a couple yrs. ago to move it so I could still help with getting it back together for them.
Take Care! [:D]
Frank
Agree – and it works for a lot of folks. I was just contrasting that with what the OP wants and noting the difference in complexity (which he may not have fully appreciated from your post).
As the OP, I need to make note of several things. I do not have any traditional control panels mounted on my facia. My “control panel” is a JMRI panel that I view on a iPad that I use as a walk around control panel. My layout is N scale Kato Unitrack and is is very difficult to see turnout position from more than a few feet away. I have zero interest in trying to use prototypical signal aspects and placement. Speed is a typical component in a protototypical signal indication but I don’t need that feature. All I want to know is how the turnouts are aligned ahead of me and if a block is occupied. I am a solo operator so I seldom have more than one train running although I sometime may have a train running laps on the mainline while I am switching off the main.
I can meet MY needs with a 3 color mast head
Green over dark = block ahead is clear/turnout straight through
Green/yellow= block ahead is clear/diverging turnout
Red = Stop (any reason, block not clear, turnout wrong)
That gives me all I need in 3 aspects and I can do that with a simple 3 LED head.
My objective here was to establish a design standard for MY layout using a signal plan that met MY needs. For the near term, I can put all my signals on my JMRI panel with zero physical signals on the layout. However, I want some on my layout for animation if nothing else. I also have a bunch of lighted building that I need to control via some solution other than a bunch of hand switches. To me, an answer to this is the emerging LCC approach and that is where I am headed. I do not plan to put it in place in one step (too $$$$$) but I do want to move in that direction.
I can put 100% of my signals on my JMRI panel for free and then add the physical side incrementally as I see fit.
Just to call this out a bit, but that’s not true.
Dwarf/ground signals provide slower-speed indications and are often found as the signal on a siding or other minor track to re-enter the main track. That siding could be 100 miles from anything in the middle of the wilderness with no other special complications going on. Not just terminals.
Also partly not true.
Certain signal indications can be displayed on a mast signal only, and not on a dwarf signal, as mentioned above the dwarf signal indications in most signal rule systems are reduced-speed indications. However any indication that is possible to be displayed on a dwarf signal can be displayed on a full mast signal, but often requires an extra signal head on the mast and the mast hardware is more expensive than a dwarf.
Generally dwarves will be used on sidings and other minor tracks where trains will enter the interlocking at slow speed. If the track arrangement is otherwise identical but it’s two main tracks merging into one, then both main tracks will have mast signals and you usually wouldn’t use a dwarf signal on a main track.
However, signal rules and practices differ a bit between railroads, so YMMV and in some situations railroads will go ahead and use bridges and mast signals on side tracks anyways.
How the real railroad does it, and what works for your own layout may or may not be the same thing. Using dwarf signals as switch position indicators or other “simplified” signal systems are p
As the OP, I was suprised at the number of people who considered it heresy that I did not plan to follow the prototype in every way or their own personal solution as a prefered approach. I suspect there are few layouts here or elsewhere that 100% matches the prototype signaling. Life is full of compromises and our small (compared to prototype) layouts are examples of those compromises.
HOWEVER - to make everyone happier, I have changed my strategy after doing more research on signals, availability of suitable parts to construct my signals and the WOW factor for the resultant cost.
First, I now plan to use single searchlight MAST instead of dwarf signals unless a dwarf is suitable for a siding. In several places, a signal bridge (with searchlights) will be used.
Second, I plan to use a MAST from Shapeway since no suitable (affordable) mast exists for N scale like I want.
Third, I plan to use a pre-wired RGY LED. This will give me RGY single searchlight indications for you protypical advocates out there.
Fourth, I have studied my layout and worked hard for suitable signal placement and I discovered that I do not need signals at every mainline turnout. I did this for myself (not you) so I could save a few $$$$ but this should make a few people happier.
Fifth, I am in the process of installing LCC to manage my blocks and signals as well as various animation stuff such as lighted buildings.
This is not cheap so I will do it in steps.
Show me some SMILES out there!