Railway Signalling and Operations: Significance of Signal Components (N. America)

Railway Signalling and Operations

Significance of Signal Components (N. America)


[ Note to Contributors | Copyright & Licensing Agreement ]

Acknowledgements

Dave Pierson, for insightful comments.


Notes

Learning the aspects obviously requires knowledge of what components of signals are significant, and which not. Here are some comments. Note that the term "arm" is used throughout. This term continues to be used on at least some railroads, even after semaphore signals were long gone. The main body of this page deals with US/Canadian signalling practices only.


Orientations of lenses on the signal face.

Various types of color-light faces are used in the U.S. and Canada. Some have one lens, the various colors being created by colored filters between the bulb and lens. Others have no moving parts, but instead have 2, 3, or 4 bulb-filter-lens assemblies per signal face. These latter assemblies may be aligned horizontally, vertically, or in a triangle. Such orientations play no role in interpretation of the signal. [BN and UP have some signal heads with 4 bulb-filter-lens assemblies. - Steve Ondic]

Zero or one lights per arm.

In most situations, each arm can have only one light. If more than one lamp lights, the signal is improperly displayed. An individual arm may be dark, however; whether this is allowed or not varies from road to road.

The only exception that comes to immediately to mind is that of a horizont lly oriented double-red or double-yellow light on the lower arm to indicate a Manual Block aspect on some railroads.

Absence of an arm.

Some railroads (e.g., Erie, modern Wisconsin Central, at minimum) installed signals with the middle arm "missing", i.e., with a large space between the upper and lower arms. Some RRs (Erie, WC) even specifically show this arrangement in the book of rules. The engineer was apparently supposed to tell, from the very wide spacing between lights, that the upper arm was a top arm, and that the lower arm was a bottom, but not a middle, arm. Often other signals in the vicinity (e.g. on the same signal bridge) helped one make one make this distinction.

In the rules I've seen to date, all cases of upper-missing middle-lower arm aspects would be interpreted identically if one mistook the lower arm to be a middle arm. If anyone knows of a specific case where this is not the case, please email me.

Shape of the signal face.

The signal face is nearly never affects the interpretation of that signal. Granted, there are great differences between a PRR signal and a B&O signal and a NYC color-light signal. But this is not simply a difference of signal faces. In the U.S., some roads use round signal faces (of various sizes), some use oblong ones, and with railroad mergers, some now have both.

One exception to signal face shape is the B&M, which used searchlight-style signal heads (one lens assembly per head) for mainline signals and horizontally oriented 3-lamp color light heads for train order signals.

Staggering of arms.

Many railroads that used color-light signals distinguished automatic signals from interlocking signals by staggering the arms. E.g., if the signal face on the upper arm was to the left of the signal mast, the middle arm was to the mast's right, and the lower arm to the left again. The left-right-left versus right-left-right pattern varied from road to road and was inconsistent on some roads. In contrast to automatic signals, interlocking signals on these roads were installed with all signal faces in a vertical line.

Staggering appears to have been most commonly implemented with the upper arm to the right of the signal mast, the middle arm to the left, and the lower arm (when present) again to the right. At least one railroad (B&M) appears to have followed the opposite convention. (Did other conventions exist, e.g., outside-inside-outside relative to centerline of double track?)

The position-light signal systems (PRR, B&O, N&W) never staggered the signal arms and, consequently, never showed it in the rules.

Significance of number plate.

Number plates are significant on most railroads. These are used to identify an automatic signal, and to distinguish it from an interlocking signal. The signal-numbering scheme varied from road to road but was frequently based on the milepost. A common scheme was to use the milepost, including tenths of a mile, times 10. Additional provisions had to be made for multitrack or bidirectionally signalled lines; these varied markedly from road to road. On the PRR, e.g., E/B signals were numbered even, W/B odd, with one of the signal numbers bearing the number of the next nearest tenth of a mile (though which was the larger number was inconsistent). On signal and track diagrams, but not on the signal itself, the track was denoted after a hyphen, e.g. "178-2" was the eastbound (even) signal near milepost 17.8 on track 2. Conrail uses a scheme based on the full milepost, with the track number and "E" or "W". That same signal would be numbered 182E, i.e., milepost 18, track 2, eastbound.

Significance conflicts

For a number of roads which used number plates and staggered signal arms, it is not at all clear from the rules which of these 2 items is, or if both are, significant.

Marker lamps, etc.

On some roads, additional lamps or other elements were used to underscore the signal's status as an automatic signal. Examples are B&O and PRR. These marker lamps were redundant only at automatic signals; at interlocking signals they could be used to change the meaning of the signal.

Some railroads placed train order signals near interlocking signals. These are generally not considered part of the interlocking signal.

Other marker lamps were (and are) occasionally used, to indicate route, or for various other purposes. Such additional marker lamps are generally shown in the rulebook under the heading "Signals not in Conformity with Typical Aspects" or some substantially similar statement. Examples include route indicators on the Northeast Corridor; a 3-amber-light N&W signal which never got "updated" to the colorized, 2-lamp variety; square-headed signals at Kiski Jct. near Pittsburgh; among others.

Vertical placement of signals relative to one another

Some railroads denoted a signal governing a spur or siding by placing it lower with respect to the signal(s) governing the main track(s). On some railroads, notably N&W, this appears to have been a sanctioned and possibly consistent practice. How "official" this was remains to be determined. No rulebook example has been found as yet.

Assumptions made in signal diagrams in books of rules

Number plate
Many railroads showed the number plate in their rulebook only when it made a difference in the interpretation of the rule. Examples include PRR, NORAC. Others noted, in text, that the presence of absence of a number plate does not change the meaning of the rule unless the diagram shows the number plate (e.g. BN). Still others showed both forms in all cases (e.g., SOU), and then frequenly showed the arms staggered.
Staggering of arms
The philosophy on whether to show this in the rules varied as well. As mentioned, SOU specifically showed staggered and aligned forms of its signals, even when the color sequence was the same. Other railroads showed only nonstaggered forms, the staggered forms implicitly bearing the same meaning -- this, despite staggering their automatic signals just as consistently.
Dark arm
A dark signal arm is assumed in certain PRR and N&W signal rules. Dark marker lights (but not the main arm) in B&O signalling are assumed. In the color-light systems, rulebooks appear never to assume a dark arm, but rather, always to show a dark arm if such is intended.

How conflicting aspects are notated

Most railroads had at least some aspects that were either not noted in their rulebook or aspects that conflicted with the standard aspects. These were shown, usually division by division, in a special section of the Employee Timetable and Special Instructions. Some of these notations were unusual installations (e.g., color lights in Pennsylvania Station, New York). Some were older signal installations (semaphore or other signals). Some resulted from trackage rights over another railroad.

Rule numbers

The AAR [American Association of Railroads] set forth a standard set of aspects and rule numbers. Many U.S. railroads followed this rule numbering scheme, which generally proceeds from the highest-speed aspects to the lowest-speed aspects. Additional aspects, probably added after the AAR scheme was developed, were notated with a rule number and a letter. The latter lettering scheme was not entirely consistent from road to road.


Special Cases

Boston & Maine apparently used searchlight-style signals exclusively for wayside signals, and multilamp color-light signals for train order signals. A lingering question is whether the relevant distinguishing factor is the number of lamps per head, or the placement of the train order signal. [DWP; MDB]


Comparison with other systems

Speed of flashing

The Netherlands railway system rules have a yellow aspect flashing at 75 flashes/minute carrying a different indication than yellow flashing at 180 flashes/minute. No other cases of speed of flashing being significant are known. This signalling system may be obsolete. (Eddie Oliver)

White light

White was initially used as a "proceed" indication, until signal engineers realized their lack of foresight. Clearly, a damaged semaphore roundel caused a false proceed to be given. White thus became rarely if ever used, particularly as a sole lamp. Notably, the Netherlands railway system used (?uses) a system where white, instead of red, is used as a placemarker, in a CROR-like signalling scheme. This use of white gets around the sticky problem of red having two meanings (stop; and placeholder of no significance), and trains passing signals at speed that have any red lamps lit.


Home
Mark's Railroad-Related Stuff
A Pennsylvania Railroad Home Page
Railroad Books of Rules and Signalling: A Home Page
Significance of Signal Components (N. America)


Mark D. Bej, M.D.
bejm@eeg.ccf.org
+216-444-0119
1997.04.16