СЦБИСТ - железнодорожный форум, блоги, фотогалерея, социальная сеть
Вернуться   СЦБИСТ - железнодорожный форум, блоги, фотогалерея, социальная сеть > Уголок СЦБИСТа > Книги и журналы > Article-RW
Закладки ДневникиПоддержка Сообщество Комментарии к фото Сообщения за день
Ответить в этой теме   Перейти в раздел этой темы   Translate to English    
 
Translate to English В мои закладки Подписка на тему по электронной почте Отправить другу по электронной почте Опции темы Поиск в этой теме
Старый 09.12.2023, 07:58   #1 (ссылка)
Crow indian
 
Аватар для Admin

Регистрация: 21.02.2009
Возраст: 40
Сообщений: 29,987
Поблагодарил: 398 раз(а)
Поблагодарили 5988 раз(а)
Фотоальбомы: 2576 фото
Записей в дневнике: 698
Репутация: 126089

Тема: ELEMENTS OF AUTOMATION AND TELEMECHANICS


ELEMENTS OF AUTOMATION AND TELEMECHANICS


Sensors

Sensors used in railway automation and telemechanics systems, depending on their functional purpose, can be divided into two groups: information and pulse sensors.

Information sensors, depending on the nature (type) of information represented by the output signals, can be divided into state sensors and parametric sensors.

Status sensors generate information about the state of control and monitoring objects in the form of “object on - object off”, “area occupied - area free”, “signal present - no signal”, “parameter OK - parameter not OK”, “object OK” - the object is faulty”, etc. This class includes track circuits, wheel passage sensors and gauge control sensors for rolling stock, sensors for the presence of vehicles in the control zone, various threshold sensors that respond to the achievement of certain (threshold) values by controlled parameters, etc. The simplest information sensor is an electromagnetic relay.

Parametric (measuring) sensors generate information about the values of parameters of control and monitoring objects. This class includes meters for speed, weight, temperature, various electrical parameters, etc.

Pulse sensors form (produce) signals necessary for the operation of various automation and telemechanics devices. This class includes pendulum transmitters, code path transmitters, pulse generators, etc.

The output information of sensors used in railway automation and telemechanics systems is presented in the form of electrical signals. Depending on the method of transforming input information, in other words, depending on the type (physical nature) of input signals, sensors can be classified.

Among magnetic sensors, inductive and induction can be distinguished. The principle of operation of inductive sensors is to change the inductance (self-induction coefficient) of a coil with a core due to a change in the magnetic resistance of its magnetic circuit. Magnetic resistance changes either when exposed to the ferromagnetic mass of the wheel (axle), or as a result of a change in the strength of the electric current that creates the magnetic field. The principle of operation of induction sensors is based on the phenomenon of electromagnetic induction: when the ferromagnetic mass of the wheel acts on the connecting magnetic field, the value of the magnetic flux changes, as a result of which an EMF is induced in the coil with the core. Induction sensors, which are most widely used in gas transportation systems, have two main types - magnetic induction and induction electromagnetic. Magnetic induction sensors have a constant binding magnetic field, its source is a permanent magnet. Induction electromagnetic sensors have a variable coupling magnetic field, its source is an alternating voltage (current) source. Based on the principle of constructing circuits for processing output signals, induction electromagnetic sensors are called differential-transformer type sensors. Depending on the method of switching electrical circuits, sensors can be divided into contact and non-contact.

Sensors can also be divided into passive (the sensor requires an external power source to operate) and active (the sensor operates without an external power source).

Let's consider the design and principles of operation of some sensors widely used in signaling systems and devices.

Wheel passage sensors are used in systems where it is necessary to count the number of axles of rolling stock - in devices for monitoring rolling stock while the train is moving (PONAB, DISK, KTSM), automation devices for hump humps, as well as in systems for monitoring the free state of track sections using the axle counting method (UCP) SO, ESSO). The sensor is installed inside the track on the base of the rail and generates an electrical signal when the wheel pair passes through a control point - the sensor installation point (when the wheel passes over the sensor). In systems for monitoring rolling stock while a train is moving [44], magnetic induction sensors of the PBM-56, DM-88, DM-95M, DM-99, ShMP-93 type are mainly used, less often - self-generating sensors of the D50 type and electronic sensors DAS; on hump humps [56,57] - PBM-56 and DP50-80 sensors, as well as inductive-wire IPD sensors; in the UKP SO system - DPEP sensors [59]; in the ESSO system - DPV-02 sensors [23] (the last three sensors are induction electromagnetic).

The magnetic induction sensor (Fig. 3.1) consists of a magnetic head 4w of a fastening device 5. Inside the magnetic head on a steel base there is a coil 2 with a permanent magnet 3 located inside it. When installing, the distance from the rail head 7 to the sensor is selected so that when the wheel passes the air the gap between the ridge and the magnet was minimal, but sufficient to exclude mechanical contact even with the maximum possible rolling of the wheel. The design of the sensor eliminates the movement of the magnet inside the coil due to vibration when a train passes, which ensures a low level of sensor interference.

The sensor works as follows. If there is no wheel in the sensor installation area (Fig. 3.1, a), the magnetic flux F of the permanent magnet is closed through the fastening device, the rail and the air gap between the rail head and one of the poles. When the wheel flange 6 passes through the air gap (Fig. 3.1, b), the magnetic flux changes as follows: first, when the size of the air gap decreases, the magnetic flux increases and reaches its maximum value when the center of the wheel is above the center of the sensor; then, when the size of the air gap increases, the magnetic flux decreases and reaches its initial value at the moment the wheel leaves the air gap (from the sensor’s coverage area).

As the magnetic flux increases in the coil, an emf is induced, creating a bell-shaped voltage pulse of positive polarity; when the magnetic flux decreases, the EMF induced in the coil creates a voltage pulse of negative polarity. The amplitude and duration of the sensor output signals are determined by the rate of change of the magnetic flux (i.e., the speed of the wheel): the amplitude of the pulse is directly proportional, and the duration of the pulse is inversely proportional to the speed of the wheel. The output signal of the sensor is sent to the actuator (indicated by IE in Fig. 3.1), which, when exposed to a pulse of positive polarity, generates a signal for the passage of the wheelset, and when exposed to a pulse of negative polarity, returns to its original state.

The design of the sensors DM-88, DM-95M, DM-99 (magnetic sensor) and ShMP-93 (Stanke magnetic noise-resistant sensor) is shown in Fig. 3.2. The ShMP-93 sensor differs from other sensors in the presence of two coils with magnets in the magnetic head.

The principle of operation of self-generating sensors of the D50 type is based on the disruption of self-generated oscillations when the wheel enters the sensitivity zone of the sensor (250-300 mm above the center of the sensor).

Electronic sensors DAS (adaptive reading sensor), DAS-A (adaptive reading sensor, analog) and DAS-A+ (adaptive reading sensor, analog, positive polarity) have a number of distinctive features.

Inside the magnetic head of the sensor there is an electronic circuit filled with a special compound. The main elements of the electronic circuit are a highly stable power generator with a frequency of 65 kHz, an inductive sensitive bridge, a threshold device and a system

adaptation to environmental conditions. The connection of the supply generator to the inductive bridge and the connection of the sensitive diagonal of the bridge to the threshold device are carried out through matching transformers.

When a wheel passes within the sensor's coverage area, the bridge becomes unbalanced, caused by a change in the inductance of its active coils and losses due to eddy currents arising in the wheel. The threshold device is triggered and generates a pulse of negative (rectangular for DAS and bell-shaped for DAS-A) or positive (bell-shaped for DAS-A+) polarity at the sensor output.

The induction electromagnetic (differential transformer) sensor DP50-80 (Fig. 3.3) consists of a magnetic head 7, shims 2, platform 4 and hook bolt 6.

Inside the magnetic head there are two rod magnetic wires - signal 2 and compensating 5 with windings wlc, w2c and w1k, w2k, respectively.

The principle of operation of the sensor is as follows. The voltage of the power source IP (20±2 V, 50 Hz) is supplied to the windings Wjc and WjK, creating two magnetic fluxes - regulating Fk, inducing EMF in the windings w2c and w2k. In the absence of a wheel in the sensor installation area (Fig. 3.3, a), the magnetic flux Fs is closed through the rail head and air gaps, and the magnetic flux Fk is closed through the rail base, fastening device and air gaps. The magnetic fluxes Fs and Fk differ in amplitude, phase and frequency, therefore, the EMF induced by them in the signal and compensating windings is also different. At the sensor output there is a detuning signal, which is compensated in the PS signal converter - at the PS output the signal is zero.

When the flange of wheel 8 passes through the air gap (Fig. 3.3, b), the magnetic flux Fs increases. As a result, the EMF induced in the winding w^ increases, and a signal of the presence of a wheel in the sensor installation area appears at the output of the signal converter.

The device of the induction electromagnetic sensor DPEP is shown in Fig. 3.4. Inside the magnetic head of the sensor (the magnetic head and elements of fastening the sensor to the rail are not shown in Fig. 3.4) there is an inductor 5 located parallel to rail b, and four coils 1-4c with the same number of turns. The coils are located above the inductor, with coils / and 2 located further from the rail than coils 3 and 4.

The IP power supply (voltage 24-42 V, frequency 71.4 kHz) generates a current that, flowing through the inductor, creates an alternating magnetic field. Magnetic fluxes, closed through the rail head, air gaps and coil cores, induce an emf in the coil windings. The output signals of the sensor are supplied to the circuits for separating useful signals PS1 (from coils 1 and 3) and PS2 (from coils 2 and 4). Since coils / (2) and 3 (4) are located at different distances from the rail, the emfs induced in them are different, but the PS1 and PS2 circuits are configured in such a way that if there is no wheel in the sensor installation area, the signals at their outputs are zero.

When the wheel flange passes over the coil, the air gap decreases and the magnetic flux increases. As a result, the EMF induced in the coil increases, and a wheel presence signal appears at the output of the corresponding circuit (PS1 or PS2). The presence of two pairs of coils in the sensor allows you to determine the direction of movement of the train.

Inside the magnetic head of the induction electromagnetic sensor DP V-02 of the ESSO system there are two pairs of coils. The windings of the coils are turned on counter to compensate for the EMF induced by the traction current. The alternating current of the power source induces an emf in the windings, which changes as the wheel flange passes over the coil. To determine the direction of movement of the train, the frequencies of the currents inducing EMF in each pair of coils are different (50 and 150 kHz).

The device of differential-transformer type sensors makes it possible to control their position relative to the rail: displacement of the sensor will cause a change in the values of magnetic fluxes, and, consequently, a change in the level of the detuning signal, which will lead to the appearance of a non-zero signal at the output of the conversion (selection) circuit.

The inductive type sensor is a coil with a steel core placed in a housing made of non-magnetic material. The sensor is attached close to the rail - under the sole or to the neck. The sensor works as follows. The current flowing along the rail creates a magnetic field, the magnetic flux of which induces an emf in the coil. The value of the EMF, and therefore the level of the output signal of the sensor, is proportional to the value of the current flowing along the rail. The output signal of the sensor is fed to the actuator, which is turned on or off depending on the signal level. In Fig. Figure 3.5 shows the installation of the inductive track sensor DIPZ-800 [42], used in devices for monitoring the filling of tracks at hump humps.

An inductive-wire IPD sensor is used on marshalling humps to determine the state (vacant or occupied) of the control pre-switch sections and to detect cuts in the system for monitoring the filling of the tracks of the train formation park (under-hill park) [57]. The components of the IPD (Fig. 3.6) are the electronic unit 7, installed in a transformer box of type TYA-2, and cable 2, located inside the track. The cable is connected to the input of the electronic unit, and the IP control relay located in the relay room is connected to the output of the electronic unit.

IPD works as follows. The harmonic oscillation generator, which is part of the electronic unit, produces a signal at a certain frequency. When the controlled area is free, the electronic unit generates and supplies a control signal (voltage sufficient to attract the armature) to the winding of the IP relay, which is energized. The loop connected to the input of the harmonic oscillation generator is the sensitive element of the sensor. When the moving unit enters the controlled area, the wheel pairs and rails form a short-circuited loop, which is the load of the generator. As a result, the frequency and amplitude of the signal produced by the generator changes, and the electronic unit stops supplying voltage to the winding of the IP relay, which releases the armature.


Sorting humps also use optical and radio sensors for detecting mobile units, which record the fact that a mobile unit is in a certain area of the controlled area [56,57].

The block diagram of an optical sensor - photoelectric device (PMT) is shown in Fig. 3.7. The main elements of the photomultiplier are: radiation source 1 - illuminator, consisting of a power transformer, a lighting (traffic light) lamp and a lens; photosensor 4 based on a photoresistor; control circuit 5; actuator 6.

In the absence of a movable unit in control zone 3 (Fig. 3.7, a), the luminous flux 2 (directional beam of light) of the illuminator is perceived by photosensor 4. Current flows through the photoresistor, and the control circuit generates a signal to turn on the actuator. When the movable unit 7 blocks the light beam (Fig. 3.7, b), the illumination of the photoresistor sharply decreases, the flow of current through it stops and the control circuit 5 turns off the actuator 6, recording the occupancy of the controlled area.

For photoelectric devices, wave emitters in various ranges, for example, low-power lasers or infrared radiation generators, can be used as a radiation source.


Radio sensors for detecting mobile units operate in the ultra-high frequency (microwave) range of electromagnetic waves. The main elements of the RTD-S type sensor (Fig. 3.8) are: a transmitting module consisting of a modulating signal generator 1, a microwave oscillation generator 2 (frequency 9.8 GHz) and a transmitting antenna 3; a receiving module consisting of a receiving antenna 6, a limiting amplifier 7 and a fixation device 8.

Installation of RTD-S can be carried out using single-channel (Fig. 3.8, a-d) or two-channel (Fig. 3.8, e-h) options as follows. In Fig. Figure 3.8, i, b shows the installation of transmitting and receiving modules on different sides of the controlled section of path 5. In this case, with a free section (Fig. 3.8, a), signal 4 emitted by the transmitter is perceived by the receiver. The fixation device produces an output signal 9 corresponding to the free state of the area. When the mobile unit 10 enters the sensor coverage area, the signal from the transmitter to the receiver does not pass, as a result of which the fixation device generates an output signal corresponding to the occupied state of the area.

In Fig. 3.8, c, d shows the installation of the transmitting and receiving modules on one side of the controlled section of the track. In this case, the occupied state of the area (Fig. 3.8, d) is recorded when the receiver perceives the signal reflected from the moving unit. When there is a free section, the signal emitted by the transmitter does not arrive at the input of the receiver.

In Fig. 3.8, e, f shows a two-channel option using one transmitting and two receiving modules installed on opposite sides of the controlled section of the track. In Fig. 3.8, g, h shows a two-channel option using one transmitting and two receiving modules installed on one side of the controlled section of the track. In both cases, the state of the section (output signal 12) is determined by the decision device 11 based on the results of processing the signals generated by the fixation devices of both receiving modules.

At hump humps, radar speed meters (RIS) RIS-V2 [42], RIS-VZ, RIS-VZM [57] are used as speed sensors for the movement of cuts along the braking positions. The principle of operation of the RIS is based on the use of the Doppler effect, the essence of which is that the signal (electromagnetic oscillations) reflected from a moving object changes its frequency to a value proportional to the speed of movement.

RIS works as follows. The transceiver module generates electromagnetic oscillations in the microwave range (frequency 37.5 GHz) and, using an antenna, emits them in the direction of a moving object. The same antenna perceives the signal reflected from the object, which is selected by the transceiver module and processed by the processing module. The difference in frequencies of the transmitted and received signals determines the speed of the object.

Radar sensors are also used as vehicle detection sensors at crossings equipped with UZP barriers [27]. The sensors perform ultrasonic location of the barrier device cover areas with subsequent processing of the reflected signals. The main elements of the sensor are an ultrasonic piezoceramic transducer, a generator of probing pulses, an amplifier of reflected signals and a circuit for temporary processing of these signals, which selects only signals reflected from objects located in the controlled area.

UKSPS rolling stock derailment monitoring devices are designed to detect derailed wheel pairs or hanging parts on a train that extend beyond the lower clearance and can damage track elements or floor equipment (electric switches, traffic lights, track boxes).

The UKPSS floor sensor (Fig. 3.9, a) is a mechanical type sensor and structurally consists of five metal brackets with shelves (corners) located outside (1.5) and inside (2, 3, 4) of the rail track on a wooden sleeper or on special platform b installed in the space between the sleepers.

The executive element that records the state of the sensor is the control relay KS, which receives power from source 7 via a circuit passing through all five corners. When mechanically impacted by elements of the rolling stock (impact), the bracket breaks, which leads to a break in the power supply circuit of the control relay. The relay releases the armature, which records the activation of the UKPSS sensor.

A significant operational disadvantage of the UKPSS sensor is the need to restore it after each operation (mechanical impact), which requires significant time costs associated with the arrival of maintenance personnel to the installation site of the distillation equipment.

In order to eliminate this drawback, devices for monitoring violations of the lower clearance of reusable rolling stock SKVP-2 have been developed and are being implemented on the railway network [28].

The sensor of the SAZH7-2 device is also a mechanical type sensor, but unlike the UKPSS sensor, it is not destroyed by mechanical impact on it. The sensor (Fig. 3.9, b) consists of sensitive elements 2 that generate electrical signals when subjected to mechanical action (impact) and protected from destruction by a metal plate 1.

Mechanical sensors also include weight meters [57], which are used on hump yards to determine the weight categories of cuts. The main elements of the weigher are a weighing element (hinge bridge), a lever system, a contact system containing six pairs of contacts, and a return spring. The operating principle of the weight meter is as follows. As the wheel passes over the weighing element, the latter bends and sets in motion the lever system, which acts on the contact system. Depending on the depth of deflection, determined by the pressing force of the wheel on the weighing element, from one to six pairs of contact springs are closed, fixing the weight category of the cut. The lever system of the weigher returns to its original position under the action of the return spring.

In addition to those considered, signaling systems and devices use other information sensors of various types and different functional purposes [42] - temperature, speed, voltage sensors, etc.

Pendulum and code path transmitters are widely used as pulse sensors in signaling systems and devices.

Pendulum transmitters (MT) produce uniform current pulses. The structure of the pendulum transmitter is shown in Fig. EVIL, huh. The main elements of the MT are a magnetic circuit 1 with two windings 2, an axis 11 with an armature 3 attached to it, a pendulum 6 and cam washers 8-10, contacts (control unit and workers 31-32,41-42).

In the initial (off) state of the transmitter, the armature is in position 3 (the armature axis is shifted relative to the magnetic axis Ml-M2), the pendulum is motionless (in the lower position 6), the washer £ closes the contact UK, contacts 31-32,41-42 are open. When a constant voltage of 12 or 24 V is supplied to the MT windings from source 72, the armature, under the influence of magnetic field forces, rotates counterclockwise so that its axis coincides with the magnetic axis Ml-M2 (position 4). Together with the armature, the axis of the transmitter rotates, as a result of which the pendulum begins to move to the right, and the washer £ opens the contact of the remote control, breaking the power supply circuit of the transmitter. The pendulum continues to move by inertia to the extreme right position (position 5). Then, under the influence of gravity, the pendulum begins to move in the opposite direction, turning the axis with the armature and washers clockwise. The armature returns to position 3. At the moment when the pendulum is in the lower position (position 6), the washer # closes the contact of the remote control, including the power supply circuit of the transmitter. The resulting magnetic field forces slow down the movement of the pendulum. The armature does not rotate to position 4, since the axis rotates in the opposite direction under the influence of the weight of the pendulum. After stopping in the extreme left position (position 7), the pendulum begins to move back. At the moment when the pendulum is in the lower position (position 6), the Popyat washer closes the contact of the remote control, turning on the power supply circuit of the transmitter, as a result of which the armature again rotates to position 4.

Thus, the pendulum, each time passing the lower position when moving counterclockwise, receives an accelerating force, i.e. When supply voltage is applied to the MT windings, undamped oscillations of the pendulum are established. When the axis rotates, the working contacts 31-32 and 41-42 close and open, through which current pulses enter the circuits of the signaling devices.

Currently, two types of pendulum transmitters are used [41]: MT-1 (MT-1M) - for powering pulsed DC rail circuits; MT-2 (MT-2M) - to turn on the flashing mode of traffic light lamps. The pulse diagram of MT-1 (MT-1M) transmitters is shown in Fig. 3.10, b, MT-2 (MT-2M) transmitters - in Fig. 3.10, c.

Code path transmitters (CTTs) are divided into contact and non-contact.

The following types of contact code path transmitters are used in signaling systems [42]: KPTSH-515, KPTSH-715, KPTSH-1115, KSh111-1315, operating from a voltage of 50 Hz; KPTSH-815, KPTSH-915, KPTSH-1015, operating from a voltage with a frequency of 75 Hz (until 1976, transmitters of types KPTSH-5, KPTSH-7, etc. were produced, in 1976-78 - transmitters of types KPTSH- 5M, KPTSH-7M, etc.). The letter Ш in the designation of the transmitter type indicates the plug version of the connector. Diagrams of code combinations of KPTSH transmitters are shown in Fig. 3.11, b. The most widely used transmitters are KPTSH-515 types.

(KGGGSH-5) and KPTSH-715 (KPTSH-7).

The main elements of the KPTSh (Fig. 3.11, a) are: an asynchronous single-phase electric motor, the rotor 3 of which has a short-circuited winding, and the stator has two windings 1 and 2, shifted at an angle of 90°; gearbox 4\ cam washers b, 7 and 8, mounted on axis 5; contact system R. The contact system consists of two groups of contacts - 31, Zh1, KZh1 and 32, Zh2, KZh2 (in Fig. 3.11, only one contact group is shown).

KPTSH works as follows. When a supply voltage of 220 V is applied, the rotor of the electric motor, under the influence of an alternating magnetic field created by the stator windings, begins to rotate at a speed of 982 revolutions per minute (rpm) at a supply voltage frequency of 50 Hz or at a speed of 1473 rpm at a frequency of 75 Hz. The gearbox reduces the rotation speed to 30.8 or 36.5 rpm, respectively, and at this speed rotates the axis with cam washers. The time required for a complete revolution of the code puck is called a code cycle.

Cam washers have a different number of projections around their circumference, corresponding to the number of code pulses. When the washer rotates, its protrusions close and open contacts. Each washer closes two pairs of contacts: washer 8—KZh1 and KZh2, washer 7—Zh1 and Zh2, washer 6—31 and 32. The closed state of the contacts corresponds to the pulse of the code combination, the open state corresponds to the interpulse interval. The washer serves to form the KZh code, washer 7 - code Zh, washer 6 - code 3. Thus, the KZh code combination consists of one pulse, the Zh code combination - of two pulses, the code combination 3 - of three pulses.

As can be seen from Fig. 3.11, b, during one code cycle, one code combination ZH and 3 and two KZH are generated. Based on the duration of the code cycle, contact CBTs are divided into two groups: with a normal cycle duration of 1.6 s and with an increased cycle duration of 1.86 s.

Non-contact code path transmitters have two modifications - BKPT-5 and BKPT-7, differing in the duration of the code cycle (1.6 and 1.92 s, respectively), the duration of pulses and intervals. The code combinations generated by the BCPT are shown in Fig. 3.12. Unlike KPTSH, BKPT does not have an electric motor and a contact system, and code pulses are generated by logic circuits made on the banks of semiconductor elements and integrated circuits. External power supply of the BCPT is provided by a voltage of 220 V with a frequency of 50 Hz, which is converted by the power supply into a constant voltage of ±12 V, necessary for the operation of the transmitter elements.

The equipment of power supply installations of electrical centralization posts includes non-contact pulse sensors DIB [18]. DIB is designed for pulse power supply of various circuits and elements and produces the following sequences of pulses: 1 s duration (at intervals of 0.5 s); duration 0.5 s (at intervals of 1 s); lasting 0.5 s (at intervals of 0.5 s). The DIB sensor is made on a semiconductor element base (transistors, diodes, zener diodes). In recent years, microelectronic pulse sensors have been developed and used [42] - DIM-1, designed to replace pendulum transmitters MT-1, MT-2, and DIM-2, designed to replace DIB sensors.
Admin вне форума   Цитировать 12
Похожие темы
Тема Автор Раздел Ответов Последнее сообщение
HollySys' Railway Automation Admin Wiki 0 18.02.2011 20:59
Feeding points of devices of automatics and telemechanics Толян Wiki 0 05.09.2010 14:40

Ответить в этой теме   Перейти в раздел этой темы   Translate to English


Здесь присутствуют: 1 (пользователей: 0 , гостей: 1)
 

Ваши права в разделе
Вы не можете создавать новые темы
Вы не можете отвечать в темах
Вы не можете прикреплять вложения
Вы не можете редактировать свои сообщения

BB коды Вкл.
Смайлы Вкл.
[IMG] код Вкл.
HTML код Выкл.
Trackbacks are Вкл.
Pingbacks are Вкл.
Refbacks are Выкл.



Часовой пояс GMT +3, время: 18:19.

Яндекс.Метрика Справочник 
сцбист.ру сцбист.рф

СЦБИСТ (ранее назывался: Форум СЦБистов - Railway Automation Forum) - крупнейший сайт работников локомотивного хозяйства, движенцев, эсцебистов, путейцев, контактников, вагонников, связистов, проводников, работников ЦФТО, ИВЦ железных дорог, дистанций погрузочно-разгрузочных работ и других железнодорожников.
Связь с администрацией сайта: admin@scbist.com
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 
Powered by vBulletin® Version 3.8.1
Copyright ©2000 - 2026, Jelsoft Enterprises Ltd.
Powered by NuWiki v1.3 RC1 Copyright ©2006-2007, NuHit, LLC Перевод: zCarot