Flashing Light Crossing Signals 802
Typical Arrangements 802
Control Circuits 802
Frequency-Shift Overlay Track Circuit Control 806
Flashing Light Control Circuit 808
Motion Detector 810
General Description 810
Operation 814
No Train - System On Standby 817
Train Enters Approach Track Circuit 817
Train Motion Detected 817
Train Stops On Approach Track Circuit 817
Train Restarts Toward Crossing 817
Train Reaches Island Track Circuit 817
Train Leaves Island Track Circuit 817
Train Stops and Backs Toward Crossing 818
Train Departs From Approach Track 818
Thermal Timer Replaced By Motor Timer 818
Highway Crossing Gates 818

Typical Arrangements
Flashing light signals are available in various arrangements,
usually with 4-inch diameter aluminum
masts, 13 feet high (5-inch and 13 or 16 feet
high when used with electric gate mechanisms).
Light units are available to display in one direction
only, in either direction 1800 apart, or at 900 (as for
a road paralleling the railroad before right angling
across the tracks). Bells may be added, as well as
signs indicating the number of tracks to be crossed.
Figure 801 shows a basic flashing light signal.
The signal mast is usually fifteen feet from the
nearest rail, and six feet from the edge of the highway.
It is general practice to locate two four-unit
signals per crossing with lights on each to flash in
both directions at 40 to 45 flashes per minute.
Modern LEX-C® flashing light units are constructed
of high-impact plastic (including the
lenses) resistant to weathering, accidental damage,
and snow blocking. The optical system is highly
efficient and projects a powerful warning beam
with horizontal spread and downward deflection to
meet a wide range of conditions. Lamps commonly
used range from 17 to 25 watts. Special windows
in each light unit allow personnel on passing trains
to check whether the lamps are lighting.
Housings, as a rule, are located about 25 feet
from the highway and 15 to 20 feet from the nearest
rail or where room is available. Underground cables
feed the signals and track circuits, and a 3-inch
pipe is typically run under the street to protect
Advance warning signs are installed off the
railroad right of way and are usually considered as
the responsibility of the highway department. All
installations should be in accordance with pertinent
railroad and government requirements.
Control Circuits
Control circuits for crossing warning systems are,
in general, similar but vary widely in detail for individual
locations because of switching, station
stops, multiple tracks, availability of commercial
power, and other factors.
On tracks where trains are operated in both
directions, the track circuit control system must be
arranged to clear the crossing signals as soon as
the train is off the highway intersection
- regardless Figure 801. Basic flashing light signal for rail-highway
of the direction of train operation. crossing warning.



Figure 802A. Simplified circuit showing use of three d-c track circuits to control flashing light, rail-highway crossing signals - shown with no train, signals dark.

Figures 802 A, B, C, D, E show a control circuit using three battery-relay d-c track circuits. When relay XR is energized, the warning devices - flashing light signals with or without gates - are controlled to clear the highway. Figure 802A shows the circuit with no train on any of the three track circuits. Note that any one of the track relays, 1TR, 2TR, or 3TR, down, will de-energize XR.

Thus, in Figure 802B, when an eastbound train shunts iT, relay 1TR drops to open the circuit to XR, and the warning devices are actuated. In Figure 802C, the train is across iT and 2T. As soon as 2TR dropped, 3XSR picked up - one of the two directional stick relays. (‘Stick” means that a relay is circuited so that it can be kept energized - once picked up - through its own front contact). Thus




Figure 802B. Simplified circuit showing use of three d.c track circuits to control flashing light, rail-highway crossing signals - shown with train on IT, signals flashing.

3XSR was picked up through 3TR front and backs of 1TR, 1XSR, and 2TR. It now has a stick circuit through a back of 2TR and one of its own fronts.
As the train continues to move eastward, Figure 802D, it clears iT and occupies 2T and 3T. When 3TR drops, a second stick circuit to 3XSR is established, through 1TR front, 3TR back, and 3XSR front.

Figure 802E shows the train on 3T, just clear of 2T. As soon as 2TR picked up, a circuit was closed to pick up XR. This circuit checks 1TR up, 3XSR up, and 2TR up, thus establishing that the train has passed west to east and is clear of “island” circuit 2T across the highway. With XR up, the warning devices are cleared to permit highway traffic to cross the tracks.






Figure 802F is the same circuit as Figures 802A-802E except that half-wave rectified circuits are used on the approach track circuits. Track relays 1TR and 3TR are connected to their track circuits in the usual manner. They are also each connected to a.c. Separate secondaries are used. The Type SVA rectifiers are, in effect, connected directly across their respective relay coils. One

half of each a-c sine wave thus passes through the relatively low resistance rectifier. The other half of each sine wave, blocked by the valve action of the rectifier, energizes the relay coil. During the half cycle blocked by the rectifier, the current induced by the collapsing coil flux also flows through the rectifier thus tending to keep the relay energized.




Figure 802C. Simplified circuit showing use of three d-c track circuits to control flashing light, signals, train across iT and 2T.

rail-highway crossing




Figure 802D. Simplified circuit showing use of three d-c track circuits to control flashing light, rail-highway crossing signals, train across 2T and 3T.

A train shunting either rectifier removes its effect from the circuit so that the d-c relay is fed a.c., and the armature promptly releases.
This circuit features high shunting sensitivity, and it is economical to install and convenient to maintain. Relay cases, line drops, batteries, etc., are not required at the far ends of the track circuits.

Frequency- Shift Overlay Track Circuit Control
Figure 803 is the same circuit as shown in Figures 802A-802F except that FSO (frequencyshift overlay) track circuits are used. These electronic track circuits, which are described in more detail in Section 400, are self-limiting and require



I ___________ I LINE I
1 U L I WTh __




no insulated joints. Modern rail-highway crossing protection installations commonly use FSO circu its.
The circuits shown thus far require three track circuits, an approach circuit for each direction and an island circuit across the highway. Figures 804A- 8040 show a circuit where control of the warning system is accomplished by two FSO electronic track circuits, which need no insulated joints. Figure 804A shows the circuit with both approach-

es unoccupied and XR up - the crossing warning cleared.
When an eastbound train approaches, Figure 804B, transmission from f1 transmitter is shunted away from f1 receiver, and ft OTR drops, opening the circuit to XR and activating the warning devices. Also, a pickup circuit is now closed to f2XSR.
Figure 804C shows the train across the highway, in the overlap area of the two overlay circuits. With the train shunt here, both transmitters are

Figure 802E. Simplified circuit showing use of three d-c track circuits to control flashing light, rail-highway crossing signals, shown as train clears 2T, signals cease flashing.




blocked from their receivers, both track relays are down, and f2XSR is held up by a stick circuit. Note that a circuit to pick up XR is complete except for the break at a front off, OTR.
In Figure 804D, the train has just cleared the highway, allowing ft energy to get to the f1 receiver,

thus picking up f OTR and completing the circuit to pick up XR, which clears the warning devices.
Flashing Light Control Circuit
So far we have shown control for XR. Figures 805A-805C show how XR controls the flasher relay EOR and energy to the lamps.


Figure 802F. Simplified circuit showing use of half-wave rectified approach track circuits to control flashing light, rail-highway crossing signals.






Figure 803. Same circuit as Figures 802A - .802F but with FSO (frequency shift overlay) track circuits, which require no insulated joints.




Figure 804A. Rail-highway crossing warning system control circuit with overlapped FSO track circuits, shown unoccupied.

General Description
The GRS solid-state train motion detector, Figure 806, for use with approach and island track circuits,

automatically clears highway crossing warning devices when a train stops and/or reverses direction while on the approach to a single or multiple track grade crossing. Thus, delays of highway traffic are reduced during switching operations on the crossing approaches. A ring sustain timer keeps the crossing warning devices operating during momentary loss of track shunt or mo















Figure 804B. Rail-highway crossing warning system control circuit with overlapped FSO track circuits, shown with eastbound train on west approach.

mentary train speed reductions below the detection threshold.
The motion detector operates in conjunction with island and approach track circuits to provide train detection and crossing device activation even in the event a failure occurs which prevents normal

functioning of the motion detector. GRS approach frequency shift overlay (FSO) track circuits are recommended although d-c approach track circuits may be used. Optional island occupancy and relay driver boards eliminate the need for separate island track circuit equipment. These boards simply plug into the motion detector module.











Figure 804C. Rail-highway crossing warning system control circuit with overlapped FSO track circuits, shown with train in overlap area.

The motion detector is available in eight frequency channels ranging from 164 to 760 Hz. The lower frequency channels provide train motion detection farther in advance of the crossing than the higher frequency channels. Depending upon bal las

conditions and the frequency channel used, train speed in excess of 80 mph may be detected as far as 4000 feet in advance of the crossing. The higher frequency channels--are more sensitive to motion closer to the crossing. Depending on ballast



EAST ——+







Figure 804D. Rail-highway crossing warning system control circuit with overlapped FSO track circuits, shown after train clears highway.

conditions and the frequency channels used, train motion as slow as 2 mph may be detected from 400 to 600 feet from the crossing.
Motion detector frequency channels are compatible with GRS frequency shift overlay channels.

No insulated joints are required. The motion detector’s highly selective transmit and receive track filters minimize harmonic frequency spillover and pickup of other channels - an important consideration in areas where multiple crossings are closely spaced.













flashing light
The approach track circuit activates and controls the highway crossing warning devices until motion of the approaching train is detected by the motion detector; control is then transferred” to the motion detector. Should the train stop or slow


Figure 805 B. Lamp lighting Circuit for flashing light signals, XR de-energized and flasher relay EOR with back contacts closed.
to a speed below the detection threshold before entering the island track circuit, the motion detector will clear the warning devices after the external ring sustain timer times out. The devices are reactivated when the train moves toward the crossing again and its motion is detected.

When the train leaves the island track circuit, the warning devices are immediately cleared. Should the train then stop and reverse direction (move back toward the crossing), the motion detector reactivates the warning devices when the train motion is detected.
Figure 805C. Lamp lighting circuit for flashing light signals, XR de-energized and flasher relay EOR with front contacts closed.
Figure 807 is a block diagram of a grade crossing protected by a motion detector, FSO approach overlay track circuit, and separate FSO island track circuit. Note that there are no approach track terminations shown for the motion detector because it is powered only when the approach track relay (AT) releases and, hence, motion detection definition or range is that of the FSO approach overlay.

Figure 805 A. Lamp lighting circuit for signals, XR energized.


‘ ‘\

Figure 806. GRS solid-state motion detector.






With this arrangement, system voltage transients, such as the track voltage changes when car wheels roll over an end termination, are over before the motion detector is turned on. Hence, no entry timer is required to allow time for the motion detector to recover (eliminates additional relays) and the motion detector can respond to trains immediately over its entire approach track length instead of waiting until the entry timer times out. The motion detector cannot be used with its internal island option when power is turned on and off by the approach track circuit relay, since the crossing would continuously ring whenever power is off and the island relay is down.
The relay logic for the crossing is shown in Figure 807A. The functions of each relay are as follows:
1. Approach Track Relay AT
This relay is normally energized when both
approach tracks are unoccupied and electrically

intact. The release of AT initiates the ringing or flashing of the crossing warning devices and applies power to the motion detector. To distinguish between the two approaches to the crossing (for example east approach and west approach), the AT relay function may be implemented with two relays, such as an east track (ET) and a west track (WT) relay.
2. Motion Relay MD
Energized when no approaching train motion is detected by the detector and the motion detector is energized (via AT or continuously); de-energized when train motion is detected.
3. Island Track Relay IT
Energized when the island track is unoccupied; de-energized when the island track is occupied. IT cancels the ring sustain time delay when a train leaves the island track, thus minimizing excessive ringing of the crossing.

Figure 807. Arrangement of a typical crossing using a motion detector, separate FSO island track circuit, and FSO approach track circuits.



4. Crossing Relay XR
Energized when the approach is unoccupied or occupied by a motionless train previously detected, and the ring sustain time delay has expired. Also energized when a train moves away from the crossing (either forward or backward) if motion was previously detected.
5. Motion Detector Stick Relay MDS
Energized when motion is initially detected and remains energized until the train leaves the approach track. Enables the MD relay to terminate ringing of the crossing after the train stops and the ring sustain time delay has expired.
6. Thermal Timer Relay TE
TE and TEPS form a timer. The timing cycle for each relay is the period required for a nichrome heating coil to both heat and cool a bimetallic strip which operates a contact. TEPS removes energy from TE when the TE front contact closes. In some applications, TEPS in conjunction with the IT relay cancels the ring sustain time delay when a train leaves the island track circuits. Also

in some applications, TE is replaced by a GRS motor-driven timer relay.
The circuit shown in Figure 807A, the simplest application of the motion detector, uses one approach track relay AT. (This circuit could include separate approach circuits). The approach wraparound track circuit, Figure 807, extends from one approach entrance to the other. The length of the approach entrance from the crossing is a function of the desired crossing warning time and the highest train speed expected through the crossing, plus a few seconds of additional time to allow for relay release, gate motor activation, etc. Generally, 20 seconds of crossing warning is desired and 5 seconds of additional time is allowed for relay and/ or gate motor operation. For example, if the highest expected train speed through the crossing is 90 mph (132 ft/sec), the approach entrance must be placed at least 132 ft/sec x 25 seconds, or 3300 feet from the crossing.
The circuit shown in Figure 807A is also simplified because powerto the motion detector is turned on by an approach track relay contact, thus elimi IT



Figure 807A. Relay logic for motion detector with wraparound applied motion detector energy, ring sustain timer, and separate FSO island track circuit.



nating the need for entry timers and termination shunts, but necessitating the use of a separate island track circuit (not the use of the internal motion detector island option). Terminating shunts may still be required if it is not desirable that movement beyond the approaches have any effect on the track circuits. For example, on continuous track if an approach is occupied by a motionless train, the motion detector may still detect movement on the opposite approach (even beyond the approach circuit) unless terminating shunts are used. The separate island is required because island relay IT must be up even if the motion detector is turned off (unless a train is in the island).
The resistor-capacitor network through contact 14 of IT eliminates a relay that would otherwise be required to shorten the crossing ringing (ringby) when a train leaves the island track circuit. There are other application circuits which are variations of the basic circuit shown in Figure 807A. The RC network is a printed circuit board which can be mounted on the back of a Type B relay plug board.
No Train - System On Standby
With no train on either approach track or island track, the relays are in the state shown in Figure 807A. The XR relay is held up by energy through contact 13 of AT and contact 15 of IT. Energy is withheld from the motion detector, keeping MD released. The MDS, TE, and TEPS relays are released.
Train Enters Approach Track Circuit
A train entering the approach track circuit shunts the track, thus reducing the signal level that is normally present at the wraparound track circuit receiver, releasing AT. The energy path through contact 13 of AT to XR is broken, releasing XR and activating the crossing warning devices. Simultaneously, energy is applied through contact 1 of AT to the positive energy terminal on the motion detector. When energy is first applied to the motion detector, the rising potential (charging of the filter capacitors) picks up MD. Subsequently, if motion is detected, MD releases. The closure of front contact 3 charges the 2700-mfd capacitor through the 10-ohm resistor to the system supply voltage (9 to 15 volts).
Train Motion Detected
When train motion is detected, MD releases. The 2700-mfd capacitor discharges through contact 3 to pick up MDS. Front contact 5 of MDS supplies energy to maintain the pickup of MDS after the 2700-mfd capacitor discharges. The pickup of MDS indicates that the motion detector has seen motion at least once during the passage of the current train. Thus, the motion detector gains

“control” over the state of the XR relay. MDS ensures that MD is capable of indicating motion and then indicating no motion before MD can control the crossing. Should the motion detector fail, MD remains constantly energized or constantly deenergized, thus MDS cannot be energized, and control of the crossing remains with the approach track circuits.
Train Stops On Approach Track Circuit
If a train for which approaching motion has been detected stops on the approach short of the island, the crossing warning devices are turned off in the following manner. The absence of approaching train motion allows MD to pick up and apply energy through contacts 2, 4, 6, 7, and 21 to the TE relay. When TE warms up, its contact 11 opens and contact 8 closes to apply energy through contacts 2, 4, 6, and 8 to the coil of the TEPS relay, causing TEPS to pick up. The opening of TEPS contact 7 removes energy from TE; closing of contact 9 allows energy to remain on TEPS.
After a time delay, TE contact 8 opens and contact 11 closes. An energy path is now completed through contacts 2, 4, 6, 10, 11, 13 and 15 to XR which picks up and turns off the crossing warning devices. The period of time between the pickup of MD (due to the absence of approaching train motion) and the pickup of XR is referred to as the ring sustain timer delay. This time delay is critical to adequate protection of the crossing.
Train Restarts Toward Crossing
When the stopped train restarts toward the crossing, motion is detected and MD releases which, in turn, releases XR to initiate the crossing warning devices. In addition, stick” energy is removed from the coil of TEPS, which releases.
Train Reaches Island Track Circuit
When the train reaches the island track circuit, island relay IT releases. The 2700-mfd capacitor, associated with IT contact 14, charges to the supply potential through resistor R. With IT contact 15 open, no energy can reach XR, thus the crossing warning devices remain in operation to ensure crossing protection independent of train motion. Island relay front contact 21 ensures that the ring sustain timer will not be activated by momentary picking up of the MD while the train is occupying the island, thus avoiding excessive ringby while the timer times out.
Train Leaves Island Track Circuit
When the train leaves the island track circuit, the IT relay again picks up. After the motion detector detects motion away from the crossing and the train leaves the island, MD picks up and applies


energy through contacts 12 and 14 from the capacitor to the coil of the TEPS relay. TEPS is stuck up through its own front contact 9 and the energy path through contacts 2, 4 and 6. The pickup of TEPS opens contact 7 and closes contacts 9 and 10, thus TE is prevented from being heated and remains de-energized with contact 11 closed and contact 8 open. An energy path is now completed through contacts 2, 4, 6, 10, 11, 13 and 15 to pick up XR and turn off the crossing warning devices. Note that the pickup of XR immediately after the pickup of MD overrides any ring sustain ringby.
Train Stops and Backs Tou’ard Crossing
When a train departing from the crossing stops and reverses its direction so that it is again approaching the crossing, the crossing is protected as follows. The approaching train, once detected, again causes the release of MD. The opening of MD contact 4 releases XR, which activates the crossing warning devices, and removes “stick” energy from the TEPS relay. The relays are now in the same state as they were when train motion was initially detected on the approach. Any subsequent train operations (such as stopping or passing through the island again) are the same as those discussed above for a detected train.
Train Departs From Approach Track
When a train leaves the approach track circuit (Figure 807) sufficient energy from the FSO transmitter reaches the FSO receiver to pick up approach track relay AT. Energy is applied to XR through the front of contact 13 of AT and contact 15 of IT. Since XR is already energized through contacts 2,4,6,10,11,13 and 15, the pickup of AT simply transfers control of XR from the motion detector to the approach track circuit. The diode around the coil of XR holds the relay energized during the crossover time of AT contact 13. The opening of contact 2 removes all “stick” energy from the MDS and TEPS relays. Finally, power is removed from the motion detector by the opening of AT back contact 1, which releases MD. The relays and contacts are now in the “standby” state shown in Figure 807A.
Thermal Timer Replaced By Motor Timer
In Figure 807A, thermal timer TE may be replaced directly with a motor-driven timer with no change in circuit operation other than slight variations in timer operation. The timing accuracy and ease of adjustment of a motor-driven timer are superior to those of a thermal timer. Also, motor timers cannot be short cycled like thermal timers, which can occur if power is removed and reap- plied by the MD relay contact 4 during the heating cycle of the timer. When this occurs with a motor

timer, the timer resets and must go through its complete timing period when power is reapplied. When a thermal timer does not completely cool before power is reapplied, it does not take the full time period to close its front contact. The back check contact on a motor timer ensures that the motor timer resets after it completes its delay cycle.
At crossings where train and traffic density warrant, the installation of automatic gates, Figure 808, in combination with flashing lights is advisable. A favored practice is to install half gates, the gate on each side of the track extending only to the middle of the street and blocking approaching traffic so that a vehicle caught between the gates may get in the clear around the end. Automatic gates in combination with flashing lights are the most efficient crossing protection system short of actual grade separation.
The track circuit control circuits are the same for gate installations as for flashing lights. The principal differences are in the gate motor control.
The lights and bells should operate three to five seconds before the gates start to descend. The total descent time including the preliminary time lag is about fifteen seconds, which means that the gate may be down only about five seconds before the train arrives. The gates act as a physical barrier, while the lights and bells give advance warning.
Figure 809A shows a control circuit for one gate (the control for the other gate is the same) with the gate in its normal, clear position, the arm at 90°. Relay XGR is controlled the same as we have shown for XR in the flashing light circuits. Note that we have added XGPR, a slow-release repeater of XGR, and GPXR that repeats XGR when the gate arm is at 85 to 90°. The hold clear, when de-energized, lifts a finger out of a detent arrangement to allow the arm to descend. When energized, it holds the gate “hooked up” at 90°. The hold clear has two coils. P is a low resistance pickup coil, which must be de-energized to let the arm descend. H is a high resistance coil which, when energized, can hold the picked up detent mechanism to maintain the gate in its 90° position, as shown in Figure 809A. Note that both coils of the hold clear must be de-energized to allow the arm to descend.
The gate contacts are operated by the gear train in the gate and are used to coordinate circuit operations with specific positions of the gate arm.
Figure 8098 shows how we get the flashers to operate before the gate arm starts down. XGR is down, and GPXR has released to start the warning lights.


Figure 808. Basic crossing gate with four LEX-C flashing light units.







Figure 809C shows the circuit just as XGPR drops, de-energizing the motor control relay and the hold clear. The gate is off the hook of the hold clear, and the motor is starting to drive it down.
Figure 809D shows the circuit at 500. Now current to the motor is cut off, and a dynamic snub circuit, shown in blue, helps to slow down the descending gate arm.
Figure 809E shows the circuit at 5°. Now the snub resistor is shunted out of the dynamic snub circuit so the gate arm will be further braked to come to a gentle stop at 0°.
Figure 809F shows the circuit condition when the train has cleared the crossing, XGR has picked up, and the gate arm is starting to rise. Note that the warning lights continue.
Figure 809G shows how the hold clear is energized at 76°. The hold clear is now picked up, ready to hold the arm when it stops.
Figure 809H shows how the warning lights are cut off at 85°. The hold clear, already picked up, is held up with its H winding energized.

Figure 8091 shows the arm at 88°. The motor is cut off to allow the arm to come to a gentle stop at 90°, at which time the circuit conditions will again be the same as shown in Figure 809A.
These simplified circuits do not show the bell control. This is taken through the gate-operated contacts. The bell is usually started with the lights and cut off at 100.
Indicator lights, if required, would be controlled through gate contacts 00 to 3° in series through all gates to indicate gates down to trainmen or remote operator. Lights may also be provided to indicate gates up by breaking another light circuit through contacts from 87° to 90°. It is sometimes required that an operator in a central tower control a number of gates at several streets during certain hours and then switch them to automatic operation at other times. Such controls and indicators are sometimes included on a regular all-relay interlocking control panel.
The only means of hand operation of the gates is to push them by hand, secure them with a hook

Figure 809A. Gate circuit, crossing clear, gate arm at 90°.



Figure 809B. Gate circuit, approach occupied, XGR down, XGPR holding on slow release to allow advance warning time for flashers and gate arm lights.

< L 85-90
r’ EBX


Figure 809C. Gate circuit. XGPR has dropped out, the hold clear is de-energized, and the gate is starting to drive down.









Figure 809D. Gate circuit. A dynamic snub circuit, shown in blue, is established at 500.

Figure 809E. Gate circuit. At 50, the resistor is shunted out of the dynamic snub circuit.




Figure 809F. Gate circuit. Train has just cleared the crossing. Gate is starting up.

Figure 809G. Gate circuit. At 76°, the hold clear is picked up through the P winding.





Figure 809H. Gate circuit. At 85°, the warning lights are cut off.

Figure 8091. Gate circuit. At 88°, the motor is cut off.








or wire, and pull an emergency switch to pick up the relay which cuts out the lights and bells until repairs are made.
The GRS Type D gate mechanism, Figure 810, has a d-c motor, which operates on 14 to 16 volts for all nominal length gates, and a circuit breaker or controller with 14 contact spaces, any of which can be varied as to angle of make or break independently of the others.
The motor drives the gate to the clear position, and a hold-clear device including coils, armature, ratchet and ratchet wheel, holds it in the clear

position until the energy to the holding coils is removed. This allows the ratchet to drop out of mesh with the ratchet wheel and the contacts on the hold-clear device to drop. With these contacts dropped, a circuit is made through the down field windings in the motor, and the gate is driven down to 45°. At 45° a contact on the controller opens and removes the energy from the motor, and the gates lower to the horizontal position by gravity. A “snub” circuit, using the motor as a generator, controls the final descent so that the gate arms come to a gentle stop. Some railroads, however, prefer a complete gravity gate, and the wiring is changed slightly to permit this action.

I., -—-,__