A Brief Explanation of Railroad Signals - MagnumMacKivler/trakpak3 GitHub Wiki
Before you jump into the fascinating rabbit hole that is adding signals to a map, it's important that you have a good grasp of what exactly signals are for and how they work.
This page covers the theory behind signal systems and is a helpful resource for those interested in designing their own signaling system from the ground up, rather than simply implement an existing system in Gmod. For an in-depth tutorial on how to use SigEdit (Trakpak3's in-game signal system editor), click here. (Link TBA)
- (=Rule vs. Aspect=) (=Signal Aspects=) (=Heads and Arms=)
Introduction
A signal's primary purpose is to prevent accidents resulting from trains hitting things they should not hit. This includes other trains, but also broken rails, workmen and their equipment, or improperly-thrown switches. If a signaling system does absolutely nothing else, it should prevent accidents.
A signal's secondary purpose is to help the railroad run more efficiently. This is a pretty broad category because it encompasses everything from keeping trains from blocking other trains to ensuring trains get through certain stretches of track as quickly as possible. Some really basic signaling systems don't need to do this, but you'd be hard-pressed to find an operating railroad today that doesn't employ signals to help the railroad run smoothly, and not just safely.
Compared to computer systems, signals are often quite simple for the ease of design, use, and maintenance. The systems used by each railroad often varied widely, however, to suit the railroad's needs. Some smaller railroads don't use them at all, due to the traffic volume being too small to justify the expense.
Terminology
Rule vs. Aspect
Strictly speaking, the "Aspect" is what the signal is physically showing (colors, arm positions, etc). The "Rule" is how you're supposed to interpret the Aspect, and is expectedly found in the railroad's rulebook.
For example, take a single red light. The "Aspect" is red, and the "Rule" is (probably) Stop.
If someone is talking about a "Stop Aspect" or a "Clear Aspect" or a "Medium Approach Aspect" or something of that nature, it's okay to assume they're referring to a color combination that translates to the rule with that name, so "Aspect" can be used interchangeably with "Rule" in most cases.
Signal Aspects
Most modern signals transmit information to the engineer by way of a colored light of some sort. Examples are the Color Light (CL), and Searchlight (SL) signals found on many railroads worldwide. You may recognize Red, Yellow, and Green if you've ever rode in an automobile. Note that the Green often has a slight aqua/bluish component to it, and that the Yellow has a noticeable reddish component, making it look more golden than actual, pure yellow. A fourth common color is Lunar White (Lunar for short), which appears as a cool, slightly bluish white. It is noticeably different from "Regular" white, which has a very slight reddish/yellowish tint to it.
Less common signal colors include Blue, Purple, Lemon Yellow, and White (not Lunar White). Blue was used by a few railroads for miscellaneous purposes (track pans, derails, doll posts). Purple was used as a substitute for Red in some signals, though as of the 1950s has fallen out of usage in the US. Lemon Yellow is used in PRR Position Light signals since it cuts through fog better than any other color. White was once used before Green was invented (no really, green glass wasn't reliably manufacturable when signals were first introduced) and as a miscellaneous light in other systems, such as B&O Color Position Light (CPL) signals.
Earlier signals, such as the semaphore, Hall Disc, banner, and ball signals used lamps (or lanterns) only during the night, when their "primary" form of indication wasn't visible. Semaphores used the position of a rotating arm, which could be vertical, diagonal, or horizontal. Hall Disc signals used a disc of cloth that was rotated out of view by an electromagnet. Banner signals used a rotating target (just like a switch stand), and ball signals used a ball that was raised and lowered by a pulley.
Heads and Arms
Most signals are referred to as having multiple Heads. These multiple heads are shown together, often vertically arranged. The combination of aspects shown on the different heads can be used to indicate many different rules.
Semaphores are the exception: their "Heads" are usually referred to as Arms since they are a long, physical object that rotates. The premise, however, is the same.
How Signals Work
Basic Signaling System
There are four fundamental rules that all other signaling rules are based off of. The simplest signaling systems use only two or three of them.
Remember how I said the primary purpose of a signal is to prevent trains from hitting things? The most basic signaling system has to be able to do that. Back in the old days, when labor was cheap, signals were all manually-controlled. Most train control was done by train orders and timetables, but on heavily-trafficked stretches of mainline, the track would be subdivided into blocks, and only one train would be allowed in the block at once. Entrance into the block was controlled by an operator who would manually set signals to allow or disallow entry. Thus the first two of the fundamental rules were born: Stop and Clear. Stop was self-explanatory: it meant to stop and wait until you received the other rule, Clear, which allowed you to proceed as fast as the track (or the timetable or the train order) permitted.
These stop-and-go signals were termed Home signals. However, there was a problem. Trains are very heavy and can take miles to stop, unlike your car which can probably do it in 100 feet if you've gotten your brakes serviced recently. The faster the trains went, the greater the stopping distance. If the train was going too fast, it might not be able to see the Stop signal in time and might roll right past it! There had to be some way to let the engineer know ahead of time that they needed to stop, and for this purpose, the Distant signal was created. Unlike Home signals, which could only show Clear and Stop, Distant signals could only show Clear and Approach, a third fundamental rule. An Approach signal told the engineer to slow down and prepare to stop at the next signal; Distant signals were placed far enough in advance of their Home signal that the engineer would be able to stop the train comfortably even when traveling at the maximum allowable speed.
But there was another problem. The Distant signal only told the engineer what the next signal was going to be, and didn't indicate what the condition of the track ahead was. So signal designers put a Home signal and a Distant signal on the same pole. The top arm was the Home signal, indicating the occupancy of the track immediately ahead, and the bottom arm was the Distant signal, indicating the aspect of the next signal. Problem solved. There's no official term for these signals, but I'm going to call them Combo signals, since they are a combination Home and Distant signal. Most signals of the Home/Distant/Combo scheme were semaphores or Hall Disc signals, since they could be implemented using only two aspects on each head/arm. Later on, pairs of 2-aspect color light signals could be used, but this was less common.
Automatic Block Signaling (ABS)
Fast forward a few more years, and something called a Track Circuit is invented. The Track Circuit applies a small Voltage to one of the rails at one end of the block, and has a sensor (such as a relay coil) at the other end of the block, which returns current through the other rail back to the Voltage source. Normally, this forms a complete circuit and current is flowing through the sensor coil. Now suppose a wheel axle is placed on the rails. Train axles are solid pieces of steel, so it forms a short circuit from one rail to the next. All the current bypasses the sensor coil and it de-energizes, and that tells the track circuit that something is in the block. If you were to break one of the rails, the same result would happen: instead of being shorted away, the current flow stops entirely, but the coil is still de-energized. Thus track circuits can detect the presence of trains as well as damaged rails, making them fail-safe.
Now that you have a relay indicating track occupancy, you can use that in place of a human operator and have the Home signal display track occupancy automatically! This is called Automatic Block Signaling, or ABS. For the purposes of this article, I'm going to refer to any signal logic that doesn't require human intervention as ABS.
Route Signaling and Speed Signaling
The old Home/Distant/Combo scheme has its limitations. It's really only good for single or multiple parallel mainline tracks with no interruptions like switches or interlockings (named, controlled groups of switches). When a train approached an interlocking, it would have to rely on train orders and timetables in order to know which route to take, and since switches are usually rated for lower speeds when thrown for a diverging route, they'd have to check the timetable/train order to know the maximum speed, too. To alleviate these problems, Route Signaling and Speed Signaling systems were developed. Don't let the nomenclature confuse you: the two are actually the same thing, but with slightly different philosophies. Some systems even combine elements of both.
Most Route and Speed signaling systems ditched the Home and Distant signals, effectively making every signal a Combo signal. This is easy to do on any signal capable of showing three aspects, like a Color Light, a Searchlight, or a 3-position semaphore. Now that you can display three (or more) aspects on a single head, you can use multiple heads to represent modified versions of the fundamental aspects that provide more information about the route the train is lined for. Almost all Route and Speed signaling systems use a maximum of 2 or 3 heads, depending on the number of rules the signaling system has. There are exceptions to this rule, of course; the DL&W, for example, simply used two pairs of heads (4 in total) to combine the Home/Distant/Combo system with Route Signaling.
Route Signaling
In a Route Signaling system, the signal rules include information about track occupancy, next signal state, and the route through the next interlocking. There are "Strong" and "Weak" Route Signaling Systems; Strong Route Signaling systems include precise information about the route, and can either demonstrate the physical direction the track will go (left, right, etc.) or identify primary, secondary, tertiary, etc. routes. Weak Route Signaling simply identifies whether the selected route is Main or Diverging; the primary route is "Main", and everything else is "Diverging." In the US, Weak Route Signaling is prevalent; internationally, Strong Route Signaling is more common.
As stated before, route signal rules are modifications of the fundamental rules (Stop, Approach, and Clear). For example, a train routed for a diverging path might be given a "Diverging Clear" signal, or a "Diverging Approach" signal, depending on the next signal state on the diverging route. The normal Clear and Approach signals when displayed on a route signal typically indicate the track is set for the main route.
Speed Signaling
In a Speed Signaling system, the signal rules include information about track occupancy, next signal state, and the maximum allowable speed through the next interlocking. As stated before, this speed is mostly determined by the route; the main path through a switch, usually straight, is good for the full rated track speed while the diverging path is only good for a lower speed. Pure speed signaling systems don't care about where the train ends up, though it can be implied based on the indicated speed.
Like route signal rules, speed signal rules are also modifications of the fundamental rules. Speed Signaling matches these rules to a speed keyword rather than an exact number; the value of the keyword is found in the rulebook. For example, an interlocking route that can be traversed at no greater than 30 miles per hour will display a Medium Clear or Medium Approach signal. In the rulebook, it will say that Medium means 30 MPH. In the US, signal speeds are typically Slow (usually 15-20 MPH), Medium (usually 30 MPH), and Limited (usually 45 MPH). Clear and Approach are passable at the maximum rated speed for the track.
Approaches, Approaches, Approaches
For both Route Signaling and Speed signaling systems, multi-head signals are typically only used before entering an interlocking; signals on multi-block rights of way are typically only single-headed. Multi-head signals are also usually found one block before the interlocking signals, too. Just like with the Approach signal, trains need advance warning when there is a reduced speed signal ahead so they can slow down in time to pass it safely.
As a result of this need for advance warning, roughly half of all signals in a modern signal system will have the word "Approach" in it, but not every one of them is a modification of Approach. To those unfamiliar with complex systems, it can seem confusing; here is how to tell them apart:
X Approach
Signal rules with some keyword immediately before "Approach" mean the signal is an Approach signal (next signal is Stop) with a maximum speed limit, or a diverging route. So for example, Slow Approach means you must pass this signal at Slow speed and then prepare to stop. Diverging Approach is a regular Approach signal with the knowledge that you're taking a diverging route (speed limit may or may not apply). The same logic applies for "X Clear" signals (Medium Clear, Limited Clear, Slow Clear, and so on), though of course the train isn't required to stop at the next signal.
Approach X
Signal rules with some keyword immediately after "Approach" mean the signal can be passed at full speed, but that the next signal is limited to the keyword speed. In a way, these signals are actually modifications of Clear, rather than Approach, because you're not required to stop at the next signal. For example, Approach Slow means you can pass this signal at full speed, but the next signal is a slow speed signal (like Slow Clear or Slow Approach). An Approach Diverging signal would mean that the next signal is a diverging signal. By this logic, the regular "Approach" signal could be thought of as an "Approach Stop" since you can pass it at full speed but the next signal requires you to stop. I've only ever seen one signaling system (CROR) implement it that way, however; since all of CROR's signals speak entirely in speed terms, they call Approach "Clear to Stop".
X Approach Y
These seem the most confusing, but really aren't; they are a combination of the above two schemes. X is the keyword for this signal, Y is the keyword for the next signal. For example, "Medium Approach Slow" means you can pass this signal at Medium speed, and be prepared to pass the next signal at Slow speed. These are highly specialized signals, used only in places with complex trackwork; normally, the description for reduced speed Clear signals stipulates that once you've passed through the interlocking entirely, you can accelerate to full speed. If for some reason you need to stay at a reduced speed all the way to the next signal, X Approach Y rules are used.
Better Living Through Restricting
So far I've been using the term "fundamental rules" to reflect three common rules nearly every modern signal system has: Stop, Approach, and Clear. But I did mention before that there were four fundamental rules. Say hello to Restricting, and its good friend, Restricted Speed.
Restricting is a special fundamental rule because it does something no other rule does: it can expose trains to a potential hazard, placing more responsibility for safety on the engineer. The definitions for Restricted Speed vary from railroad to railroad, but as a general rule, it says to proceed at a very slow speed; commonly, Restricted Speed equates to 15 MPH, though some railroads go as low as 10, others as high as 20, and some specify different speeds depending on whether or not you're inside an interlocking. What makes Restricted Speed unique is the stipulation that trains must be prepared to stop within half their sight distance to any obstruction. This obstruction can take the form of another train, workers, damaged track, a misaligned switch, or anything the primary purpose of signals is supposed to prevent trains from hitting.
In layman's terms, a Restricting signal means "There might be something in the way. Go ahead anyway, but BE CAREFUL." Restricting signals have a wide range of purposes. You could use Restricting to stack two trains in the same block, say, a passing siding, to allow a third train to squeeze by afterwards. Or you could use a Restricting signal to route a train into "Dark" (unsignaled) territory, such as a yard, branch, siding, or spur, where they won't have another signal to tell them when to stop. Restricting functions as the manual override for the signaling system, allowing trains to proceed in special cases which the system can't otherwise handle.
Restricted Speed's most profound contribution, however, is its ability to modify Stop. For example, suppose you have a double track mainline with two opposing one-way tracks. A train on one track can only move in one direction, and barring any special circumstances (such as an accident, or a dispatcher hold, or some kind of mechanical problem), will likely continue to move in that direction. Instead of making a train stop and wait for a long time because the track ahead is occupied, you could have the train Stop and Proceed at Restricted speed. The definition for Restricted speed includes the warning to watch out for hazards, so if you do actually catch up to a stopped train ahead of you, you can easily stop in time. Most mainline automatic signals (usually denoted by a number plate on the mast/pole) are not capable of showing a true Stop aspect, and a red light on these signals actually means Stop and Proceed. These signals are often called Permissive signals. Interlocking signals do not usually follow this rule (Stop means Stop), and thus are called Absolute signals.
Some railroads, like the B&O all the way up to CSX, took it a step further and equated Stop and Proceed to Restricting, which means that trains aren't even required to stop, and can simply roll on by at Restricted speed.
Another example of how Restricted Speed modifies a Stop signal is the Grade Signal. It requires a lot of fuel to start a heavy train from a dead stop, and fuel costs money. It also takes time to fully stop, then release the brakes, then get going again, and time is money. Rather than waste money^2 having engineers hill start their trains every time they come to a Stop and Proceed signal, the addition of a G plate (G for "Grade") allows high-tonnage freight trains to treat the signal as Restricting instead of Stop and Proceed. As the name suggests, Grade Signals are only used on grades.
Small optimizations like Stop and Proceed and Grade Signals play a big part in minimizing delays and help the railroad run like a well-oiled machine.
External Control
By using simple logic, ABS signals can calculate the exact aspect to show for all track and signal conditions. But as robust as ABS logic is, it can't make decisions the way a human can, to prioritize trains, set routes, or handle other, unexpected situations. Some measure of human control is always necessary at interlockings, either locally (at an interlocking tower), or remotely (by a dispatcher at a Centralized Traffic Control, or CTC center). To ease the task of monitoring many signals at once, signal operators are usually given only a small measure of control over each signal. In the simplest systems, the operator can either Hold a signal (force it to display a Stop signal), or Allow it to set its own aspect based on track conditions. A signal set to Allow can still stop a train if the track ahead is occupied or the route is misaligned, but as soon as conditions improve, the signal will let the train through. A little bit of additional circuitry can allow the operators to set up signals for a One-Time Allow: they will automatically revert to the Hold state once a train passes. Lastly, the operator is sometimes able to Force the signal to let a train pass, usually by giving a Restricting or a Stop and Proceed signal.