Italian Signals


Brief Railway History of Italy

The first railway in Italy was opened in October 1839 between Naples and Portici, 5 miles, in the Kingdom of the Two Sicilies. Subsequent development was rapid, especially in the North, and was carried out by private companies. By 1859, the largest network extended from Venice to Turin and Genoa through Venetia, Lombardy and Piedmont. There were a few lines in Tuscany connecting Florence, Pisa, Leghorn and Siena, a single line in the Papal States from Civitaveccia to Frascati via Rome, and the lines radiating from Naples. Locomotives were furnished mainly by British builders, and British engineers were often used. The track gauge was fixed at 1445 mm (4 ft. 9 in.) in the 1860's. It was not unusual to ease the standard gauge slightly; the Pennsylvania Railroad also used 4' 9". The standard gauge of 1435 mm was adopted in 1934.

Many Italian lines were single-track, unusual for the time. The telegraph block was introduced in 1846 on the single line from Padua to Vicenza, but I do not know the type of instruments used, since it was too early for the Morse system. The Morse telegraph was introduced after 1850, and became predominant. The Morse register gave a physical record of messages, which was considered desirable. Later, the telephone block was very widely used. The first interlocking was a Vignier installation from France, installed at Pontedecimo at the bottom of the Giovi incline in 1865. The first Saxby and Farmer frame was installed at Genova Porto Principale in 1874, probably the new rocker type. This brought with it British lower quadrant semaphores. These were first used on the Turin-Trofarello line in 1882, by that time including yellow distant arms with notched ends.

The widespread adoption of the telephone after the First World War made possible the dirigente unico method of operating secondary lines. There was an American-stle train dispatcher who would direct all operations, the guard of a train calling in at each station for instructions. This was first introduced on the Fabriano-Urbino line with 16 stations in 1926, and spread widely.

As in Germany, it was required in Italy that turnouts not be damaged if run through in the trailing direction if set wrong. This made it difficult to bolt lock facing points. A speed limit of 60 km/h was established over facing points until 1922, when improved locking devices were introduced. This meant that even express trains slowed to 60 km/h at all stations.

It proved very difficult for private companies to prosper, since it was very expensive to construct railways in the mountainous country, with many tunnels and major bridges. The famous Giovi incline on the line from Genoa northward toward Milan and Turin was 4-1/2 miles of 3.5% gradient (1 in 30), with a 2-mile tunnel on 1 in 33 at the top, for example. After the creation of the Kingdom of Italy in 1861, the government was often required to intervene in railway matters. Lines connecting the north and south were encouraged, to unify the new country. In 1885, the railways were partially nationalized and grouped into three major systems, the Rete Mediterranea (RM), Rete Adriatica (RA), and the Rete Siciliana. Private companies administered these systems, while the state owned the infrastructure. The first two of these connected the rich, industrially developing North with the poor, agricultural South with lines down both coasts. The first connection with foreign railways was via the Mont Cenis tunnel at the Fréjus pass into France via Modane in Savoy.

Continuing financial difficulties led to the creation of the Ferrovie dello Stato (FS) in 1905, which continues to exist today, although passenger operations are now operated by Trenitalia S.p.A., to comply with EU requirements. The first general manager of the FS was the engineer Riccardo Bianchi, previously with the Rete Siciliana and before that with the Rete Mediterranea, and a co-inventor of the Bianchi and Servettaz hydraulic power interlocking. The first Bianchi and Servettaz plant was installed at Abbiategrasso near Milan in 1886, and was later very widely used.

Since Italy almost completely lacks domestic energy sources, except for hydroelectricity, there was an early interest in electrification. A government commission was established in 1897 to study the matter. There were few alternatives at that time, mainly just low-voltage DC as used on tramways (the first electric tramway in Italy was Florence-Fiesole in 1890), and the new three-phase AC promoted by Brown Boveri in Switzerland, Siemens and Halske in Berlin, and Ganz & Co. in Budapest. The RM adopted 650V third-rail DC, while the RA chose three-phase from Ganz & Co., who offered attractive incentives as a demonstration project, as well as the services of the brilliant young Hungarian electrical engineer, Kálmán Kandó.

Three-phase locomotives used large, simple induction motors without commutators that could, even at that time, be insulated for high voltages, with a great savings in transmission losses in comparison with low-voltage DC. High transmission voltages could be stepped down at static substations to the traction voltage, instead of the expensive rotating converters of DC substations. To limit the starting current, liquid rheostats were used that were filled with 5% sodium carbonate solution as the locomotive accelerated. Some observers were surprised to see electric locomotives stopping at water cranes and emitting clouds of steam!

The complication and expense of two contact wires (the third phase was supplied through the rails) was not the principal disadvantage of the three-phase system. This was the fact that the induction motors ran at a fixed speed depending on the frequency of the supply. By changing the pole connections, the effective number of poles could be halved, doubling the speed. Even with this possibility, the frequency had to be lower than the usual industrial frequency of 42 Hz. On the first line, in the Valtellina north of Milan, the supply was 3400 V at 15.8 Hz., and the locomotives were geared for 35 or 70 kmph. This line was opened for trial in 1901 and for public use in 1902.

Locomotives had double collector poles at each end, with a roller to contact each wire. At first, only one set was raised at a time. It was later found that if both sets were raised, the necessary gaps in the overhead at turnouts were effectively bridged, and the current was not interrupted, which could be very inconvenient. Another advantage of three-phase was very simple dynamic braking, with the energy recovered returned to the overhead. When descending a gradient, if the locomotive drove the traction motors above the synchronous speed, they automatically functioned as alternators.

The Giovi was electrified in 1910 from Pontedecimo to Busalla, and through to Genoa in 1911, at 3300 V, 15 Hz. The famous E550 class with 2100 hp output was developed for this line, of wheel arrangement E and rod drive, with speeds of 14 and 28 mph, very suitable for this heavy gradient. Each "lttle giant" did the work of two steam 0-10-0's. Locomotives of this type became the standard three-phase freight locomotive. The alternate line, the succursale, was electrified in 1914, followed by the main line to France through the Mont Cenis tunnel to Modane in 1915. After this, three-phase spread widely in northern Italy, even on light-traffic branch lines. This was certainly the most widespread use of three-phase traction in the world, with over 1240 mi. in service.

The problem of a low top speed remained, and was very troublesome in passenger service, even though Italy had very slow trains for many years. Some four-speed passenger engines were developed, giving 23-1/4, 31, 46-1/2 and 62 mph. Experiments were made in the late 1930's with 10,000 V and 42 Hz, but these were abandoned after World War II. It was possible by then to produce gears for a greater speed reduction, incidentally. Sadly, the three-phase is now gone, but it was always a distinctive feature of Italian railways.

Principally to overcome the speed problems, 3000 V DC was adopted in the 1930's, and slowly superseded the three-phase. 50 Hz AC electrification was not yet fully developed, and Italy did not want to adopt yet another frequency, the 16-2/3 Hz used in Germany. DC motor design had advanced enough that 1500 V traction motors were now available. If they were used in pairs permanently connected in series, an economical 3000 V overhead could be used. Four traction motors could be connected in series, or in series-parallel, giving two equilibrium running speeds. Although this was not much of an improvement on the three-phase by itself, the DC motor speed could easily be controlled by resistors, and also by weakening the motor fields for higher speeds. There was the much simpler overhead, but the advantage of dynamic braking was lost. DC locomotives (and diesel-electrics) can use rheostatic braking, but the excess energy cannot be recovered, because of insuperable control problems.

Most of the Italian network is now electrified at 3000 V Dc. Recently, advances in solid-state power electronics allow the application of chopper control of the traction motors, which completly overcomes any speed control problems. The motor is connected in series with an electronically controlled switch and a choke inductor, and the fraction of the time the switch is on controls the average voltage applied to the motor, without any loss in resistances. Also, solid-state rectifiers have superseded mercury arc rectifiers, which themselves superseded rotary converters in the power substations.

Modern Italian Signals

For distinguishing directions relative to a signal along the line, the Italian term "a monte" is used where in British and U.S. practice one says "in rear of", and "a valle" instead of "in advance of". Apparently the common words for "rear" and "advance" are used in exactly the opposite sense as in U. S. and British practice.

Modern Italian signals are light signals (segnali permanentemente luminosi) of "searchlight" type that show red, yellow and green through a single aperture. These were made standard in 1947, replacing all other types of light signals. Signals are normally placed to the left of the track and have a circular background that is black with a white border. When signals are placed to the right of the tracks that they govern, the background is square, black with a white border. The lights may be fixed or blinking.

The aspects displayed by a single light (which we shall call an "arm") are illustrated at the right. Not all aspects may displayed by an individual signal. Signals are classified as protection signals or departure signals. At a station, a train encounters protection signals on entry that lead it to its arrival track. The train is then permitted to proceed by departure signals that lead it to its correct route. If there are more than one protection or departure signals, they are identified by square signs, black on yellow for protection signals, black on white for departure signals. The first protection signal and the last departure signal on a route bear plates marked "EST" (external), while intermediate ones are marked "INT" (internal), and are numbered, if there are more than one. This is a relatively important distinction, since the EST signals are boundaries with block sections on the main tracks between stations. Also, they make identification of a signal clear.

Signals that can display Stop (via impedita), a red light, are called signals of the first category (segnali di prima categoria). The Stop aspect cannot be passed except under very restrictive circumstances. These signals are usually coupled with distant aspects (con avviso accopiato), displaying a yellow light warning that the next signal is at Stop (avviso di via impedita), called Approach in the U.S.. As illustrated on the left, the mast of such signals is painted black below about half-height, and grey or metallic above. When the signal does not display a red light, it is a purely distant signal, and its mast is painted in alternate black and white rings. This painting allows it to be identified when not showing any light, implying that its most restrictive indication is Approach, not Stop. In any case, green means "via libera" or Clear.

In stations, the placement of Stop signals is governed by the location of turnouts (deviatori), so the distance between them may be less than the stopping distance. In this case, a distant signal governing the approach to them displays a flashing yellow when the second signal is at Stop, even when the first signal is displaying a yellow light and would normally be preceded by a green distant signal.


If diverging routes are available, a second arm is added below, as shown at the left. If this arm is extinguished, the upper light displays the same aspects described above, applying to the main route, without speed restriction. When the upper light is red, and lights are displayed below, the signal indicates proceed on a diverging route at the speed prescribed by the preceding distant signal, or 30 km/h by default. This signal does not indicate the maximum speed at which the signal should be passed. The lower arm displays a distant aspect for the succeeding signal. Often the aspect displayed is red over green, meaning proceed on a diverging route at 30 km/h expecting the next signal displaying better than Stop, unless otherwise provided. If several routes are possible, the signal is complemented by a route indicator showing the number of the route in a matrix of lamps. Route 1 is usually the one furthest to the left, route 2 the next to its right, and so on. This signal will normally be seen displaying a red upper light and an extinguished lower light, which is its Stop aspect.

This signal replaced the earlier "candlestick" signals in which there was a high arm, like the upper arm, and one or more low arms, each corresponding to a route or a set of routes. These signals were analogous to the equivalent semaphore signals. The low lights or arms referred to 30 km/h routes, the single high light or arm to the direct route with no speed restriction. If all the lights or arms were at the same level (no high light) they were all considered as low. The new aspects are much more economical.

After World War II it was realized that light signals provided a means for displaying additional distant aspects. This would allow a 30 km/h divergence to be preceded by a distinctive distant aspect instead of by the usual yellow light, which requires a severe speed reduction. Two lights arranged vertically, each capable of displaying either yellow or green, would give four aspects. Using yellow-over-yellow for the normal distant aspect and green-over-green for the Clear aspect was considered at first, but since it would be extremely costly to convert the older single yellow and single green aspects to conform, this possibility was rejected. The next idea was to let the upper light display only yellow and the lower only green, but allowing each individually to be concealed, which would give three aspects, yellow-over-black (black = extinguished) and black-over-green, like the old ones, plus the new green over yellow for a diverging route.

There was now an objection to having two Clear aspects, one with an upper green and the other with a lower green. This was overcome by allowing the upper light to display either yellow or green, the lower one green or black. The yellow-over-green diverging aspect was later modified by flashing to govern approaches at 60 km/h. When 100 km/h crossovers were installed, at first a second green light was added. However, this was soon replaced by the modification of the blinking phases. These aspects present the novel feature of blinking either simultaneously or alternately. As you can see, these aspects are not likely to be confused. In addition to these are the usual single yellow and single green distant aspects.

A new aspect was the double yellow, used approaching a signal controlling entrance to a dead-end track, or to one that may be obstructed by rolling stock. It demands restricted speed (marcia a vista) and extra care instead of simply 30 km/h.

Two separate signal units in a single enclosure, with an oval background, was originally used for these two-light distants. It became more usual to use two ordinary signal units with round backgrounds, especially when the distant was coupled to a signal of first category, so that both sets of aspects could be displayed as required. If all the distant aspects were needed, two lights could be added below the top one, giving three lights in all. An example is shown in the illustration. The red-yellow-green aspect is the "Christmas tree", meaning proceed on diverging route at predetermined speed expecting a restriction to 30 km/h at the next signal. Usually, this signal would show only the red light, the lower lights being extinguished.

The red over green aspect recalls the American Medium Clear aspect whose indication is proceed (on diverging route) at medium speed, which was variously 30 mph or 45 mph. However, the actual speed is not specified by the signal, since it is determined by the preceding distant aspect. This is quite different from practices in other administrations, when the speed is either prescribed by rule (as in the US) or displayed at the signal. In France, a Ralentissement is displayed in approach, and a Rappel de Ralentissment at the point of execution, which fundamentally prescribes a speed of 30 km/h. Signs are used to prescribe higher speeds. Route and speed information are separate from the movement information provided by the basic signals in France. In Italy, a three-light signal displays distant aspects for a diverging route ahead on the two bottom lights, while distant aspects for the direct route are shown on the top two lights, with the bottom light extinguished, while the first category aspects (red, green) are displayed on the top light with the lower two extinguished. It should be noted that in most cases, if a light were unintentionally dark, a more favourable aspect would not be displayed. A top light dark definitely means that the signal is imperfectly displayed, and in that case would be regarded as a Stop signal. This use of extinguished lights is uniquely Italian, and has been well thought out.

Auxiliary speed signals were introduced in 1979 to meet the objections to the lack of a specification of speed at the point of divergence, as shown at the right. If no horizontal bars are illuminated, the speed restriction is 30 km/h. One illuminated bar specifies 60 km/h, and two 100 km/h. Railwaymen call these signals informally "Rappel" after the French example.

At the left is shown the square background of a signal placed to the right of the track it governs, as used, for example with double tracks signalled for movement in both directions on either track, or where visibility or clearances demand. The triangle with black border is used beneath a single-light signal to prescribe a 30 km/h speed limitation, for example over a turnout approached in the trailing direction where there is no choice of route.

The two signals at the bottom display either lunar-white or blue lights, steady or flashing, on the mast of a signal. These are descended from the "calling" signals introduced in 1947, and are intended to permit trains to pass a signal at Stop under certain conditions. Since signalling apparatus must always "fail safe", it often occurs that a signal cannot be cleared from Stop, even though there is no danger to a train that passes the signal with caution, for example due to a faulty track circuit or a poor contact. Then, under certain conditions, these signals can be illuminated to allow the train to proceed without lengthy negotiations. The lunar white lights are used on signals within station limits, while the blue lights also imply that the block is clear and allow a train to pull ahead to a signal that may be more favourable. If the lights are blinking, then the route must be carefully checked as the train proceeds, perhaps operating points by hand to the correct position. These signals are called "Owls" because of their appearance. The illustration shows signals based on the original "calling" signals with one enclosure,a single visor and plain lenses; later examples are two separate lamps with Fresnel lenses.

The signal illustrated at the right shows another way to overcome the difficulty of a signal that cannot be cleared. This is a departure signal with an "A" that can be illuminated. When the "A" is lighted, it shows that the manual electric block or axle-counter block is clear, and the train can proceed at restricted speed. If the "A" is flashing, turnouts on the route must be examined to ensure that the proper route is set. If the "A" is not illuminated, the train conductor should communicate by telephone to the controller for instructions. If telephone communication is out of service, the train must wait 20 minutes, and then proceed at restricted speed. The black-on-yellow "Ns" plate below the "A" indicates that the signal cannot be passed at stop when the "A" is not illuminated (s = superamento). It is analogous to the "Nf" plate on absolute stop French signals. There may additionally be a black-on-white plate below this identifying the block section protected by the signal.

Signals within a station may be equipped with a similar "D" plate that can be illuminated. The required actions are generally the same as for the "A" plate. The "D" plate confers no authority to enter a block.

At the left are shown low or shunting signals. They are called "marmots" because of the movements of the early mechanical low signals. The lamp remained in a fixed position inside the enclosure with purple and lunar white lenses while the enclosure rotated as it was moved upwards and downwards (normally by an hydraulic piston). The enclosure would rise and rotate, like a marmot coming out of its burrow to look around. Later electrically-operated signals and the light signals illustrated did not retain the up-and-down motion, but were still called marmots. If the two lights arranged vertically are blinking, the permission to proceed applies only to trains governed by other signals, such as a departure signal governing several tracks, not to shunting movements. If the signal pertains to the track to its left, an arrow is displayed to indicate this. Otherwise, a low signal applies to the track to its right.

The colours originally displayed by these signals were red and green, and the rotating enclosure was painted correspondingly. Purple and lunar white were introduced in 1931, with black and white painting. In 1932 the modern painting was introduced, but with a purple and white zebra stripe that became black and white in 1949. The now nearly universal light signals were introduced in 1966.

High shunting signals may be supported on their own masts, or may be placed below category 1 signals. As shown in the figure at the right, they display rows of five white lights as shown. "A" is via impedita. If an accompanying low shunting signal shows via libera, it may be ignored. "B" is via libera, but anticipating the next high shunting signal at via impedita, or the end of track. "C" indicates via libera, with next high shunting signal at via libera. "D" indicates via libera, but on a track that may be blocked. These signals do not affect the movement of trains, and do not require the observance of low signals on the same route. If they govern more than one track, the track affected is indicated by a low shunting signal.

Where a departure signal is not visible from the point at which trains stop, or where the departure signal refers to more than one track, it is repeated by a signal consisting of two white lights arranged vertically, either elevated or at ground level. The ground level lights blink when lighted, and apply only to trains. They may be used when the departure signal refers to more than one track to indicate the track involved. In this case, they may not be passed by arriving trains when not lighted. If there is a departure signal for each track, the repeaters may be accompanied by a sign showing an upside-down "V" that can be illuminated. If it is illuminated and flashing, an arriving train can pass the associated extinguished departure repeater.

Hand signals for shunting are given with a red flag by day, and a white or red lantern by night. Each signal is preceded by a sound from a trill whistle, given by the employee doing the coupling or uncoupling, or else by the employee most distant from the engine. A rolled-up flag or a white light, moved side to side, is a signal to go forward, while moved up and down is a signal to go backwards. An unrolled flag or a red light, held steady, is a signal to slow down, while if it is moved up and down, it is a signal to stop.

These signals may easily be criticized as inconvenient or confusing on two counts. First, "forward" or "backward" is less well defined than "towards me" or "away from me" as used in German signals, for example. Several paragraphs are given in the rules for defining forward and backward, and usually it must be agreed upon in each individual case. If a locomotive is at the end of a cut of cars, "forward" is the direction pulling the cut, "backward" pushing it. Second, having to roll and unroll the flag, or change the colour of the lantern light may not be difficult, but it must be very inconvenient when necessary in signalling single or repeated movements. No signals are prescribed for buffering up or for kicking, as in the German code. The signals for "back" and "stop" are confusingly similar. It is probably a relief that these signals can now be given verbally by radio, as is mentioned in the rules as well.

The lineside signals at left are provided to indicate the proper operation of lifting barriers or warning lights at road crossings. Lifting barriers may be operated by attendants, or automatically by trains. The signals with triangular backgrounds are distant signals, and those with square backgrounds are placed near the crossings. These signals may be colloquially called "alberi di Natale" because of the red and green lights and triangular shapes. These signals may protect up to four level crossings in an interval of 2 km or less. A sign below the signal shows the number of such crossings, if more than one.

Proper operation of the flashing lights at level crossings without barriers, automatically operated by trains, is indicated to train drivers by the signal shown on the right, which is preceded by the triangular sign showing "PL".

Historic Development

Three eras can be characterized by the principal type of signals used. The first, extending to about 1905, was dominated by the revolving disc signal. Semaphores predominated in the second, until they began to be replaced by light signals in 1925. The third era, characterized by light signals, has been described above.

The revolving red disc has a long history in France, where it still existed until recently. When "closed" it displays a red disc, and a red light by night, to an approaching train. When "open" it is turned parallel to the track and originally showed the "white" light of the lamp, since the red glass was removed when the disc rotated. It was placed at a distance from the first turnout of a station equal to the stopping distance plus the length of the longest train and perhaps plus an additional safety distance, perhaps about 800-1000 m. Normally, on double track, a lever at the station was moved to pull a wire that opened the signal. When the lever was returned to normal, or the wire broke, the signal turned to closed by the action of a counterweight. A train finding the signal open could run through the station without stopping. In this case a switchman would return the lever to normal for a certain time interval, perhaps ten minutes, then open the signal again. If a train stopped at the station, the closed signal would protect it while standing at the station. After it departed, the signal would remain closed for the prescribed interval, after which it would be opened again. This "normal clear" operation was usual in these days. A signal would be turned to Stop only when the line was obstructed in some way. A train might well encounter a closed disc with insufficient space to stop before passing it, but this was allowed for in the placement of the signal. In France, there was even a post showing the limit of protection, that a train would pull within for safety. These arrangements actually did work quite well for many years.

If a disc was placed closer than 550 m from the point of restriction, this fact was announced by a permanent yellow disc, showing that there was limited stopping distance ahead.

There was, however, a need for a signal that could stop a train at a definite point and permit it to proceed when desired. In France, the carré was introduced for this purpose. In Italy, the plain red disc performed this function, and was called a disc of the first category. The disc operated in the traditional way was called a disc of the second category, and had the figure "2" painted on its face. A disc of the first category was always preceded by another disc, called of third category, at a distance sufficient to ensure a stop before the disc of first category was reached. That is, it served the purpose later assumed by the distant signal. It was probably operated simultaneously with the first category disc, and had a "3" painted on it. This classification of discs appeared by 1905, when semaphores were already becoming important. Category 2 semaphores are said to have been used on the Bologna-Pistoia line in 1889, after their introduction a year earlier. Category 3 semaphores were used with the idea of protecting a train ahead of them.

Semaphores were classified in the same way as discs. The category number was painted on the white stripe of the semaphore arm. As for discs, no "1" appeared. A semaphore without a number was automatically of category 1 and could not be passed when at stop (horizontal). Although in the beginning various forms of semaphore arms were used, the FS had a standard arm that became universal. It was clearly derived from the British semaphore arm, and painted in the same way. It was slightly tapered to overcome the optical illusion of seeming narrower at the outer end. The back was painted white with a black stripe. The arm was depressed at an angle of 54%deg; when open. In 1905, green was offically adopted for Clear, instead of its earlier meaning of Caution, Reduce Speed, so the arms showed red and green lights by night. However, white lights were still in use many years afterward, especially with discs, which it was difficult to modify to show green. A few still existed in 1940. Green for Clear actually appeared as early as 1886, but was rare except on semaphores. The inner end of the arm is counterweighted so that the disconnected arm will return to horizontal, which explains its unusual shape. Semaphores displayed a green backlight by night when at Stop, white when open. The green was finally changed to violet in 1932.

Note the counterweight just beneath the arm that opposed the weight of the operating linkage. When the arm was horizontal, any pressure on it acted on a line through the pivot of the counterweight lever, so the arm could not move. In British and American lower-quadrant semaphores, this was not the case. The safety aspects of this arrangment overrode the disadvantage of pushing the arm off, where the weight of the up-and-down rod did not help a return to Stop. There was a second counterweight at the bottom of the mast to counteract this weight.

The use of distant signals instead of category 3 semaphores slowly grew. In Italian terminology, this was called "double signalling" to distinguish it from co-acting category 3 signals. Double signalling was encouraged after 1914. Rotating discs and category 3 signals were abolished in 1925. Category 2 signals were retained, but restricted to secondary lines. Distant arms were painted yellow and had a fish tail end like British and American distant arms, but instead of a contrasting black chevron had a simple white stripe like that of a category 1 blade. This could not have aided visibility in any way.

A distinguishing characteristic of a category 2 signal was the absence of a distant signal. This meant that the signal would often be overrun. However, the rules seemed to imply that trains would stop before reaching the signal, so that it would protect a preceding train that had stopped and had pulled just inside it. There was no allowance for stopping distance in the placement of the signal, just the length of the longest train plus 100 m from the point protected. However, what are essentially fixed distant signals were now provided, as shown in the illustration. These suggest that a train would ease off preparing to stop until the signal was sighted, so that a stop short of the signal was possible. Since they were restricted to secondary lines with maximum speeds of 80 km/h or 90 km/h, the inconvenience would be minimal. A steady yellow light is displayed by night. Also illustrated is a fixed distant approaching a station with a spring switch, at which a train might have to stop if the switch was not in order. This signal is placed to the left of the track, in spite of its square shape.

Distant signals were normally located the prescribed stopping distance in front of a category 1 signal, perhaps 800 to 1000 m, but depending on the gradient and maximum speed. At some locations, such as at stations, this might put the distant signal for the departure signal uncomfortably close to the entrance signal. In the US and Britain, the distant signal would then be put below the home signal on the same post. This was not done in Italy. Instead, the "coupled" 1st category and distant signal arms were on the same axis, so that the home arm covered the distant arm at Stop, but revealed it when pulled off. Introduced in 1922, this became the standard Italian practice, and exists today in light signals showing red, yellow and green lights, which is completely analogous.

These signals were arranged as shown in the diagram at the right. The stop arm, in front, had one red roundel and one empty roundel, while the distant arm, in back, had one green roundel and one empty roundel. A swinging yellow roundel, pivoted at "a" in the stop arm and with a pin "b" in the distant arm running in a slot, was located between the arms, which were both pivoted at "c". When both arms were horizontal, the light from the lamp passed through the empty roundel in the distant arm and the red roundel in the stop arm. When both arms were depressed, the light passed through the green roundel in the distant arm and the empty roundel in the stop arm. If only the stop arm were depressed, the movable yellow roundel was drawn in front of the empty roundel in the distant arm, and a yellow light was displayed. When the distant arm was then lowered, the pin "b" pushed the movable roundel out of the way. A lug on the stop arm prevented the distant arm from being lowered when the stop arm was horizontal. The same thing could be accomplished in principle by superimposing the red and yellow roundels in the two arms to display red, but there would be a penalty in intensity, so apparently this was not done.

The only other use of similar signals that I know of was on the New York Central and Hudson River in the early 20th century. Here they were not "coupled stop and distant" signals, but were used at the limits of manual block territory to indicate "Clear but end of block system" when the stop arm was depressed, leaving the fixed distant arm showing.

This idea was carried even further in signals with three superimposed arms. The idea here was to add a fourth aspect, "Proceed on Diverging Route at Restricted Speed", using a yellow arm with a circular crown (as it was called) on the end. The aspects are shown at the right. The colour of the distant and crowned arms was described as "orange" but this is not a good colour for .gif files, so I have substituted yellow, which did indeed later replace orange. These signals were abolished in 1937, as was the orange.

The earliest way of showing signals for several routes was to stack the semaphore arms vertically, as shown at the left. The topmost arm governed the leftmost route, the next arm the next route to the right, and so on, as shown by the track diagram. This method was used nearly everywhere, and had the advantage of compactness. It should be recognized that it applied to the routes available from a certain track, not to parallel tracks. Its outstanding defect was that a principal route, perhaps the one without a divergence, was in no way emphasized and could correspond to any of the arms. This arrangement was the basis for American signal aspects, the different arms referring originally to different routes, and later to different speeds.

A junction signal can give information as to the route a train will follow, or as to the speed at which the train should proceed. Originally, the route was generally indicated, and the driver was expected to know the speed appropriate to the route. This plan is still followed in Britain, where the route is indicated by a line of white lights at certain angles, but the appropriate speed is not specified except by vague hints. Most other railway administrations now use some way of indicating the proper speed, while not displaying the route at all. Current Italian practice, as we have seen, indicates the speed, but not long ago indicated the route instead.

The difficulty of emphasizing the principal route with vertically stacked arms is easily overcome by rotating the display by 90°, and showing the arms side by side instead. These are called "candlestick" signals in Italy, where the individual arms are the candles. These arms governed routes geographically corresponding to the arms; the leftmost arm governed the leftmost route, and so on. If these arms were at the same level, they governed routes with the usual speed restriction of 30 km/h, which applied to all low arms until 1936. The signal for the most direct route could be raised above the others, and then it was not considered subject to any speed restriction. An arrangement such as this is shown in the illustration. If the high arm were open, then the preceding distant signal could be opened as well. If only a low arm was open, then the distant signal could not be opened. This is exactly a simplified form of the British practice. This arrangement was originally carried forward to use with light signals, and may still be found in many places. Often, a low arm may refer to a group of routes, and the selected route may be indicated by an illuminated route indicator consisting of a matrix of lights. Note the common way in which routes are numbered, from left to right.

Light signals began to be considered around 1925. Originally they were the type that had separate red, yellow and green lights, of the type that is generally used today on other than Italian railways. These signals are subject to phantom aspects when an external light (usually a low sun) enters the optical systems. It is a property of optical systems that light rays are reversible, so this light is reflected at the mirror behind the lamp filament, and again exits to the front. The result is that all the lights appear brightly lit, and the intended aspect cannot be determined. A type of signal that did not suffer from this defect was invented in 1920 by the Hall Signal Company in the United States. This was the "searchlight" signal that was essentially a polar relay that rotated a screen with coloured filters, with a single optical system that projected light of the selected colour through a single aperture. Any external light entering the signal would be coloured by the filters in the same way as the internal light, so the aspects could not be confused.

The searchlight signal was adopted by the FS and became the standard light signal in 1947. It was not only easy to use electrically, but the moving parts were enclosed and protected, so that maintenance was inexpensive. In the 1980's, it was decided to minimize maintenance by devising a static signal, with no moving parts, that would retain the external appearance of the searchlight signal, with the colours displayed through a single aperture. There are at least three ways to do this. Optical fibres from three independent coloured sources can be merged randomly into a single bundle. Or, interference filters can reflect lights of the required colours, combining them into a single beam. Finally, with the development of high-intensity light-emitting diodes (LED's), a display consisting of diodes of different colours can be used.

Fibre-optic signals were developed and proved practical. However, the optical losses are high, requiring high-intensity bulbs. LED signals (of a single colour) are already in use for traffic signals, but I have not heard of a railway application. The control of the beam would be rather difficult in any case. It may be noted that LED signals would not be subject to phantom aspects, since there is nothing coloured in them.

Interference filter (dichroic) signals have proved practical. The light from any three filaments is focussed by ellipsoidal mirrors to a single point that is the focal point of the main aspheric lens. One interference filter reflects yellow light, but allows red light to pass unobstructed. The second interference filter reflects green light, but allows yellow and red light to pass unobstructed. The interference filters are transparent to light of a wavelength greater than a certain cutoff, but act as a very good mirror for shorter wavelengths. They are made by depositing multiple thin layers on a plane substrate, and are commercially available.

Separate red, yellow and green sources are used, but it would really only be necessary to use red and green. The combination of the two lights would give an excellent yellow. Similarly, only red and green LED's are necessary, and the same applies for fibre optic signals. One wonders why advantage was not taken of this property, which would make the signals cheaper and simpler.

The standard Italian signal colours are sensations corresponding to the spectral wavelengths 495 nm for green, 593 nm for yellow, and 637 nm for red. If the illumination is by an incandescent lamp, the coloured filters must not be too narrow in wavelength, since this reduces the brightness greatly.

References

C. Zenato, Evoluzione Storica e Tecnica del Segnalamento Ferroviario Italiano (Salò: Editrice Trasporti su Rotaie, 2006).

M. Antonio, La Segnalectica Ferroviaria Italiana (Pozzuolo del Friuli: Martino Antonio, 2001).

R. F. I., Regolamento Sui Segnali, Edizione 1947 as amended, Reprinted 2002.

P. M. Kalla-Bishop, Italian Railways (Newton Abbot: David and Charles, 1971).


Return to Railways Index

Composed by J. B. Calvert
Created 18 June 2008
Last revised 3 April 2009