FR2565379A1 - DISTRIBUTED INPUT / OUTPUT SYSTEM FOR A PROGRAMMABLE AUTOMATE - Google Patents

Abstract

THE INVENTION RELATES TO PROGRAMMABLE AUTOMATO INPUT CIRCUITS. AN INPUT SYSTEM FOR A PROGRAMMABLE PLC INCLUDES IN PARTICULAR A SET OF 24-26 INPUT MODULES WHICH CAN BE PLACED CLOSE TO THE PROCESS 30 WHICH IS CONTROLLED. EACH MODULE IS CONNECTED TO A CENTRAL UNIT 20 BY AN INPUT CONTROLLER 22. EACH MODULE INCLUDES A SET OF OUTPUT CIRCUITS 37, 38, 39 WHICH CAN OPERATE SELECTIVELY AS INPUT CIRCUIT OR OUTPUT CIRCUIT, SELECTION S 'PREFERABLY PERFORMING UNDER CONTROL OF THE CENTRAL UNIT. APPLICATION TO THE CONDUCT OF INDUSTRIAL PROCESSES.

Description

The present invention generally relates to methods and

  devices intended for use with "programmable logic controllers", and it relates in particular to an intelligent input / output system for such controllers. Process control with a programmable logic controller involves inputting input signals from various sensors associated with the process, and

  the application of output signals to control elements

  dice of the process. The process is thus controlled according to

  of a registered program and conditions of the pro-

  which are indicated by the sensors. Such a com-

  This is obviously applicable to various and numerous processes, and programmable controllers make it possible to advantageously control the sequential operation of

  industrial processes, conveyor systems and

  chemical, petroleum and metallurgical

example.

  The development of PLCs

  is relatively recent. A modern programmable controller

  does not include a central processing unit (CPU) consisting of

  Typically, a processor for executing the recorded program, a memory unit of sufficient size to store the program and the data relating to the state of the inputs and outputs, and one or more power supplies. In addition, an input / output (I / O) system provides the interface between the central processing unit

  processing and the input devices and the elements

  the process that is ordered.

  Input / output systems remained relatively

  have not changed since the advent of PLCs

  grammars, and it is in this area that an improvement is most needed. Although some progress has been made in the I / O systems, the improvements have

  have generally been made in the extension of the following

  lives in the past. For example, the patent

  U.S. 4,293,924 discloses an I / O system in which

  interface is increased. Another technique,

  illustrated by U.S. Patent 4,247,882, is

  to improve the structure in which the

  input / output system. However, because of the comple-

  increasing demand for processes, and because of the need for greater information exchange between the process and the central processor, it has become

  necessary to provide better solutions to the problems

my input / output.

  The conventional I / O system includes a number of individual I / O points, and each of them is assigned a specific function of either receiving the signal from an input device (eg a contact limit switch, pressure-sensitive contact, etc.), either to apply a control signal to an output device (eg an electromagnet, an engine start circuit, etc.), depending on the

  configuration of the circuit relative to the particular I / O point

  link. An I / O point is so specialized as to be a point of entry or an exit point, and you can not

  convert it easily from one use to another.

  A problem that affects modern I / O systems (especially when used with a

  complex process) consists of the high cost of

  tallation. Modules, or circuit boards, I / O are

  typically housed in bays or casings

  sis of cards. The control of an important or

  plex requires a large number of I / O points in each bay or chassis. This necessarily leads to high cabling costs (both for the main-

  than materials), because it is necessary to

  to bring wire from the I / O bay

  all input and output devices.

  The use of a large I / O bay then raises additional problems, as it is frequently difficult to bring all the wires into the bay to make the connections. Although a good technique

  known is to incorporate at least one part of a system

  I / O system in a cabinet or bay remote from the uni-

  treatment center (to try to bring the

  I / O circuits of the process that is controlled), these problems

  they are still not resolved because there is a

  input / output wiring center at a location that

  although distant, is nevertheless unique. Complications

  Additional information is emerging regarding heat dissipation in a concentrated I / O system and for this reason it is frequently necessary to

  operating an I / O system below its possibili-

optimal locations.

  Another problem that affects current I / O systems is that diagnostic and troubleshooting operations are difficult, both in the event of a fault in the PLC itself, and in the event of a fault in the PLC. process that is controlled. Experience has shown that most PLC-related faults that occur during process control occur in the I / O system. The part corresponding to the central unit is now very elaborate because it has necessarily benefited from advances in technology

  microprocessors and computing, for example.

  However, when an electrical fault actually occurs

  it is often extremely important to be able to

  detect and diagnose early on the nature of the

  problem. It is of course desirable to detect a defective item by means of a warning given in advance rather than after the loss of control

part of the process.

  With current I / O systems, it is difficult to detect faults early, and even when a fault is reported, its precise nature and location may not be apparent. In many cases, it is even difficult to distinguish between PLC input / output faults and faulty elements (eg motors, push buttons,

  etc.) involved in the process. We can easily see

  the absence of diagnostic elements, especially for the I / O systems of the PLC. It is therefore highly desirable to achieve improvements

  to diagnose and prevent faults in systems

I / O systems.

  The problem of fault diagnosis is some-

  complicated by the fact that each I / O point is usually protected by a fuse. Although the fuse

  protects the particular I / O module against overcurrent

  it often complicates the problem. As an example

  a simple transient current can blow the fuse, thus leaving the I / O point completely off, until you can locate the point

  defective and replace the fuse.

  A somewhat related problem is the exchange of diagnostic and control information between a control part and a controlled part of an I / O system. For example, distributed I / O modules may be used to define the configuration of an I / O system. In such a case, it is desirable to provide simple and reliable means and methods for

the exchange of such information.

  Conventional I / O systems still present

  Another disadvantage is that, as

  born above, each I / O point operates strictly in

  point of entry or exit point. We can not

  easily turn the same point from one use to another.

  The user of a PLC must therefore select

  separate the input and output functions, based on an initial estimate of requirements. There is a marked lack of flexibility with respect to unforeseen future needs. In addition, because I / O points are typically available in groups (for example, six or eight points per circuit board), there are frequently a large number of unused I / O points.

in a control system.

  The main object of the invention is therefore to

  an entry / exit system that eliminates these

  from traditional I / O systems. More particularly, however, it is sought to provide an I / O system in which one can select for each I / O point a

  operating mode corresponding to either a point of

at an exit point.

  It is also sought to provide a system of

  in / out where each I / O point is auto-

  protected against overcurrent and overload conditions

  voltage, without using fuses or circuit breakers,

  where each I / O point is diagnosed

  continuous and automatic tic to search for faults, both in the I / O system and in the process that is controlled, and in which the detected faults are

  identified and reported automatically. An additional goal

  specific purpose of the invention is to provide a system

  I / O system that is simple and economical to wire and operate, and provides individual I / O points in distributed groups or modules, intended to be placed in

  immediate proximity of the process, or the particular element

  link the process to order. A further object of the invention is to provide an I / O system which includes means for monitoring, controlling and troubleshooting each I / O point independently of the conventional central processing unit. Other goals, features

  and advantages of the invention will emerge from the description

detailed below.

  The present invention provides a system of

  intelligent output / output for use with a programmable logic controller, and includes a set of input / output (I / O) modules, each of which may be placed near the process being controlled, or part of this process. Each module is interconnected, via a transmission link, to a central processing unit (CPU) via a controller

  each module itself consists of a set of

  of input / output circuits, each of which can be used selectively in an input circuit (for accepting an input signal from the process), or in an output circuit (for applying an output control signal to the input circuit). process). The selection of the operation according to one mode or the other is preferably carried out under the control

  of the UCT and its operating program registered

  tré. Each I / O module further comprises a unit of

  operation control (microcontroller) which

  directly each I / O circuit and provides an orderly exchange of diagnostic and control signals between each I / O circuit and the I / O controller and the CPU. The communication between the control unit of the operation of each module and each associated I / O circuit is preferably effected by means of a pair of

  drivers, one of whom is driving a set of

  A plurality of recurrent control signals (eg in signal frames), while the other carries coded diagnostic signals. A characteristic of each

  The I / O module consists of the means

  seen to connect conductors that route input and output signals between I / O circuits and

the process.

  The rest of the description refers to the drawings

  attached which represent respectively: Figure 1: a simplified block diagram of a programmable controller system comprising an intelligent input / output (I / O) system according to the invention; Figure 2: a perspective representation

  of a possible physical form for an I / O module indi-

  viduel and a hand-held verification device,

  both having a configuration intended for use

  in the I / O system of Figure 1; Figure 3 is a block diagram showing in more detail one of the I / O modules of Figure 1; Figure 4: a simplified block diagram of a transmission section and a control and detection section for an I / O point of the type shown in Figure 3; Figures 5 and 6: representations of signals

  showing the relationship between certain signals that intervene

  in the circuits of Figure 4; Figures 7A, 7B and 7C are diagrams showing various usable input / output switching circuits

  with the I / O circuit of Figure 4; Figure 7A shows

  trant a DC source circuit, Fig. 7B showing a DC absorption circuit and Fig. 7C showing an AC circuit;

  Figure 8: a diagram showing in detail a section

  control and detection for the I / O point of FIG. 4; Figures 9A, 9B and 9C are diagrams showing in detail a transmission section for the I / O point of Figure 4; and

  Figure 10: A truth table showing the correlation

  correspondence between diagnostic and state data and a

  4-bit coded signal, for the realization of a common logic

  in a state encoder for the transmission section.

  FIG. 4. The programmable logic controller of FIG. 1 comprises a central processing unit (CPU) 20, an input / output (I / O) controller 22, a set of input / output modules 24. 26 and a data link 28 which interconnects each I / O module 24-26 to the controller

  I / O 22. These elements, excluding UCT 20, constitute

  generally kill the input / output system of the con-

  troller. The CPU 20 has a substantially conventional structure and may include one or more microprocessors for data manipulation and control, plus a

  memory for the recording of operating programs

  data, input / output data and other

  abutments, which are used in the execution of the registered program and for the implementation

  of the control function. In addition, other

  conventional devices, such as power supplies, are incor-

  as necessary for the CPU 20 to perform all of its functions. The I / O controller 22 controls the information exchanged between the

  various I / O modules 24-26 and CPU 20.

  Each 24-26 I / O module can be placed separately

  remotely from the CPU 20 and the I / O controller 22,

  and in the immediate vicinity of the process that is controlled.

  Although only three I / O modules are shown in Figure 1, their actual number can obviously be significantly higher. As an example, the system described here can easily accept sixteen separate I / O modules. Each I / O module is independent of the others and

  each of them can be reserved for the control of a proces-

  distinct from the one that is commanded by all the others

I / O modules.

  By way of example, in FIG. 1, the I / O module 26 controls a process 30 represented in a generalized form. The input and output signals that are associated with the process 30 are routed through leads 32 that extend between the process 30 and the I / O module 26. The process 30 can obviously take virtually any form . However, he understands in all

  various cases sensors, switches, etc. (not shown

  specifically), which are intended to detect the state and conditions of the process. 30 The information that comes from the process is in the form of

  input signals to the I / O module.

  also includes controlled elements (eg pumps, motors, etc., which are also

  not shown), which receive the output signals

  from the I / O module 26 and thereby perform control of the process 30. In a similar manner, each of the other I / O modules 24, 25 is interconnected with

  input and output devices and associated apparatus

to a process.

  The data link 28 is preferably a serial link, although parallel signal transmission between the CPU 20 and the I / O modules 24, 26 can easily be realized. In all cases, the modules The transmission link 28 may consist of a twisted pair of conductors, a coaxial cable or a fiber-optic cable 24-26 I / O are connected to the transmission link 28 to communicate with the device 20. The transmission link 28 may consist of a twisted pair of conductors, a coaxial cable or a fiber cable. optical; all these structures are acceptable, depending on considerations such

that cost and availability.

  In FIG. 1, the I / O module 24 shows in block diagram form the electronic structure

  general overview of each I / O module.

  This structure thus includes a micro-contr8-

  their 36 having interface access for the exchange of in-

  training with the CPU 20 and including an associated memory

  (not shown) for the implementation of a

  registered operating gram according to which the various elements of the I / O modules are controlled and diagnosed with regard to appearing faults;

  a set of I / O points (or "I / O circuits") indi-

  37-39, each of which can be used selec-

  as an entry point or exit point, and each of them being individually

  and directly by drivers to elements of

  or out of the ordered process; and a lead bus 40 for interconnecting the I / O points 37-39 to the micro-controller 36. The number of I / O points 37-39 in any particular I / O module 24-26

  depends on practical considerations such as dissi-

  However, by way of example, it has been found to be quite practical and convenient to provide sixteen points.

I / O per I / O module.

  A verification device 42 is provided for

  check the integrity and proper functioning of the

  entry and exit and to ensure the maintenance

  this and troubleshooting. The verification device 42 is preferably sized to be able to be held at the

  hand, so that they can be moved easily and

  moderately from one I / O module to another. It is designed to be connected to each I / O module by means of a cable

  which includes a connector that adapts to another connection

  attached to the I / O module. The cable and the adapted connectors are shown schematically in FIG. 1, which shows the verification device 42 connected to the

  I / O module 24 via an interface access

this of the microcontroller 36.

  When the hand-held verification device 42 is connected to an I / O module, it verifies and controls the I / O points of that module and displays diagnostic information about the module. . The hand-held verification device replaces

  advantageously performs these functions independently of the

  20, and even in the absence of the CPU 20. The verification device 42 is for example able to enable and disable output points and read the state of entry points. In case a defect

  appeared, the verification device 42 can also

  provide an indication of the nature and location of

  fault. It may be noted that the verification device

  manually, 42, includes a data display

  44 which displays alphanumeric characters, and a

  set of keys 46 which make it possible to perform a

  address grammar and operate the modules

I / O 24-26.

  We will now consider Figure 2, which shows

  be preferred physical forms for a

  hand-held verification and for an individual I / O module

  vidual. The I / O module 51 shown is substantially in the form of a terminal strip which

  includes a row of conductor terminals, 53,

  to establishing connections with the conductors that are connected to the input and output devices of the controlled process. Terminals 53 may be in the form of screw connections in which the screws are tightened on a lead wire or a terminal lug. Each I / O circuit is assigned to a connection

  corresponding terminal. In addition, terminals are

  for connecting an external power supply

  (alternative or continuous), and for establishing

  connections with the data link as shown in Fig. 1. Visual indicators are provided in the form of light-emitting diodes (LEDs) to indicate the status of each I / O point. Additional LEDs 57 and 58 indicate the operating state of the module 51. For example, the LED 57 indicates the existence of a fault condition (internal or external to the module) and

  LED 58 indicates normal operating conditions.

  A connector 59 is mounted on the module 51 so as to

  adapt to a cable connector 60 and ensure

  nexion with the hand-held verification device,

  49, via the cable 61.

  As described above and in connection with Figure 1; the hand-held verification device 49, which is shown, is capable of stimulating the I / O module to which it is connected. In other words, the hand-held verification device makes it possible to operate

  an I / O module and to control it in a very com-

  plete, even if this module is not connected to a unit

  central processing unit, as shown in FIG.

  The block diagram of Figure 3 shows in more detail an I / O module 80 (substantially identical to any one of the modules 2426 of Figure 1). The I / O module 80 thus comprises a group of 8 separate I / O points, 81-88, each of which exchanges information, control and diagnostic signals with the microcontroller 90. alternating or continuous electrical energy is applied to the terminals H and N. The energy source which is connected to the terminals H and N feeds

  both an internal continuous feed 94 and

  any external output (eg

  ordered) which are ordered by the controller

  programmable program to which module 80 belongs.

  94 is simply the continuous supply for all the elements contained in the I / O module 80 whose

  operation requires continuous electrical energy.

  Each I / O point 81-88 is connected to the micro-

  controller 90 by a respective pair of leads -102. A driver of each pair, called line D, carries control data to the associated I / O point; the other line, called the M line, carries status and diagnostic information from the I / O point to the microcontroller 90. Each I / O point 81-88 is also connected from way to receive

  continuous electrical energy (for example 15 volts)

  from the power supply 94, and each of them is connected to the terminals H and N of the energy source. If the external energy source which is connected to the terminals H and N is, for example, an alternating network at 115 or 230 volts, the terminals H and N correspond simply respectively to the

  phase conductor and the neutral conductor of the network.

  However, if the external power source is a DC source, the H terminal may be the positive side

  from the source and the N terminal the negative side. In addition, each

  I / O module 81-88 includes an IN / OUT terminal that performs a dual function. If the I / O point is to work in the output point, the INPUT / OUTPUT terminal for that point is connected to the controlled element (or the

  load) in the process, that this point must control.

  If, on the other hand, the I / O point is to function as an input, the INPUT / OUTPUT line relating to this point receives

  the input signal that comes from the input device.

  The same INPUT / OUTPUT line thus fulfills both functions.

  tions, under the control of the order from the micro-

  controller 90 and the second connection (or reference connection) of the input or output device. As

  for example, the I / O point 82 is represented in

  in which it operates at the point of exit, by turning on or off a charging device 89. The load 89 is connected between the IN / OUT line of the I / O point 82 and the line N towards the source energy. On the contrary, the I / O point 84 is

  represented in conditions in which it functions

  is an input point, with an input switching device 91 connected between the INPUT / OUTPUT line and the line H of the energy source. Any 81-88 I / O point can be operated in the output mode.

  DC power source, in a DC absorption circuit

  DC or AC source, depending to some extent on the internal I / O point circuitry. We will describe more completely hereinafter this

aspect of the circuits.

  The information that is transmitted to the micro-.

  controller 90 from each I / O point 81-88, through the connection via line M, includes data indicating the state of the charging current (high or low), the power level which is applied at this I / O point, the temperature condition of the I / O point, the state of any input device, and further information, all of which will be presented below.

more detailed way.

  The control of each I / O point 81-88 is finally determined by a central processing unit, as indicated generally in connection with FIG. 1. In FIG. 3, the communication with such a CPU is performed via an interface port (preferably a serial aspect) of the microcontroller 90, and a data link 106 (28 in Fig. 1). Other substantially similar I / O modules

  module 80 of Figure 3 can also be connected to

  to the data link 106. Although

the microcontroller 90 reacts to the orders of the central unit

  processing controller, it also provides localized and distributed control of each I / O point within the I / O module 80. The microcontroller 90 is an operation control unit and it works in accordance with a program registered and under the control of orders from the central processing unit and signals

  received on line M, from each I / O point 81-88.

  Although this is not shown specifically in FIG. 3, the microcontroller 90 also includes a memory for recording the program and for recording other data necessary for program execution and execution. of the desired control function. The simplified block diagram of FIG. 4 shows a preferred embodiment of an I / O circuit, excluding the output switching device. The I / O point thus comprises a transmission section 111 and a control and detection section 113. The transmission section 111 (which will be considered first) comprises a timer 117, an output data filter 119, an output selector 120, a two-bit counter 121,

  a latch of conservation of the last state, 123, a basement

  By default, the state circuit 124, a state encoder 125, a state latch network 127, and a data selector 129. The transmission section 111 receives, on the line D, a GIS signal from the computer. operation control unit (e.g. microcontroller 90 of Fig. 3) and a set of status indication signals (or

  diagnosis) on a six-wire bus 115. The

  transmission system 111 applies a command signal CON-

  DUCTEUR / BLOCK at the control and detection section 113, and it transmits a diagnostic signal (STATE) to the

  micro-controller, on line M. The command signal CON-

  DUCTEUR / BLOQUE finally orders a communication device

  mutation (preferably an insulated gate transistor, or

  IMT, which will be considered later), whose function

  This is not the same depending on whether the I / O point should work as input or output. Figures 5 and 6

  show the relationship between certain signals that intervene

  in the operation of the transmission section 111, and reference will be made to these figures in relation to the

figure 4.

  The GIS control signal is a pulse train.

  The present invention includes coded information including conductive / blocked status information, last state conservation information (HLS), deferred status information (DEF), and synchronization information. It consists of a series of "frames", each containing two or four

  impulses followed by the omission of an impulse, that is

  ie a "missing impulse". The missing pulse functions to resynchronize the operation of the transmission section 111. Each of the two or four pulses has a duty ratio of 25% or 75%. The duration T between pulses within a frame

  is fixed and this duration is also that of "the impulse

  The GIS control signal is initially applied to a timer 117, and under the effect of

  the rising edge of this signal, timer 117 rests on

  heights and triggers his delay cycle. The temporality

  The sender 117 thus emits a rising edge of the clock signal CLK approximately 0.5T after each rising edge of SIG. The signal CLK is used to drive the clock input of a two-bit counter 121, an output data filter 119 and flip-flops 123 and 124. Unless it has been restored before, the timer 117 also issues

  a rising edge of the SYNC timing signal approxi-

  1.5T after a rising edge of GIS, and it transmits

  a falling edge of the LOS signal after a noticeable time

  longer after a rising edge of GIS (eg

  2.5T). Normally, the rising edges of GIS appear

  at intervals of T, so that the tempo

  The riser 117 is restored before the SYNC or LOS transitions can occur. However, at the appearance

  of a "missing pulse" (synchronization interval

  2T appears between rising edges of SIG, causing SYNC to go high for about 0.5T. The SYNC pulse restores the transmission section 111 and therefore signals that a new frame is about to begin. If a time greater than 2, 5T appears between rising edges of GIS, LOS goes low, which signals to the transmission section 111

  that a signal loss has occurred.

  The conductive / blocked state information that is transmitted by line D to the I / O point is contained in the first two pulses of each frame of the

  control signal. An impulse with a cyclic ratio

  75% corresponds to a logic "1"

  conductive mutation), and a cyclical ratio of 25% cor-

  responds to a logical "0" (switching device

  than). As will appear later, the clock pulse that appears at 0.5T after the rising edge of a SIG pulse, actually causes sampling of the SIG pulse at this time. Thus, if a pulse having a duty cycle of 25% (0.25T) has been emitted, a low level or "zero" at 0.5T is obtained. If on the other hand a pulse having a duty cycle of 75% (0.75T) has been emitted, a high level or "one" at 0.5T is obtained. The first two pulses are also emitted

  redundant; that is, the first two impulses

  should be matched (both to 1 or both to 0) for transmission section 111 to react to

  the order DRIVER / BLOCK. For this purpose, the communication signal

  SIG command is applied to the output data filter 119

  which actually samples and compares the two first

  first pulses of the control signal. If the two pulses are different (for example due to noise interference), the output data filter 119 maintains the last valid DRIVER / BLOCK sequence

who was received.

  If one frame of the control signal contains four pulses instead of two, the third and fourth pulses are used to respectively update the last state of the conservation latch, 123, and the state latch taken by default, 125. contents of these flip-flops 123

  and 124 is changed only upon receipt of the third

  sth and fourth pulses. A logical "1" in the posi-

  the third pulse sets the high pulse

  last state conservation signal, HLS; a logical "O"

  that in the position of the third pulse causes the HLS signal to go low. The HLS signal appears at the

  output of the HLS 123 flip-flop and is applied to the selection

  output 120 and the state encoder 125. Similarly,

  a fourth impulse sets the default state signal at the high or low level

  (high level = conductive state, low level = blocked state).

  The default status signal DEF and its complement DEF

  appear as output signals from the

  124. The default status signal DEF is applied to the state encoder 125 and its

  DEF complement is applied to the output selector 120.

  In the case of a plague of transmission from the micro-

  controller (that is, a loss of the LOS control signal to the low state), the HLS signal instructs the

  output selector 120 is to maintain the conductive state

  previously blocked, or to take the state taken by

  default. If the HLS signal is a logical "1", the previous state

  tooth will be maintained; if HLS is equal to "0", the transition to the default state will take place as soon as LOS goes low. We see the advantage of this operation: in case

  loss of communication between the I / O point and the

  positive control (that is, the microcontroller of FIGS. 1 and 3), the conductive / blocked condition is forced

  in a preferred, preselected state.

  The two-bit counter 121 counts the CLK pulses to produce an output count, SO and S1, which takes binary values between zero and three. This count value indicates the pulse of a frame that is being received, and is applied (in the form of SO and S1) to the output data filter 119, to the last state conservation latch 123, to the state latch taken by default 124 and the data selector 129, so that each circuit responds only to the pulses

appropriate frames.

  The signals in Figure 5 show the relationships

  between the signals SIG, CLK, SYNC, LOS, and the signal

  DRIVER / BLOCK for various conditions. For the first

  first frame (for convenience, arbitrary

  frames by frame numbers), redundant pulses

  25% duty cycles are output, which corresponds to "0" or a blocked state of the switching device. Clock pulses are produced at 0.5T after each rising edge of a GIS signal. A synchronization interval or "missing pulse" follows the two redundant pulses. The missing pulse shows a SYNC pulse, which indicates the end of a frame. Because the two SIG pulses have a duty cycle of 25%, the CONDUCTIVE / BLOCK signal remains low and the LOS signal remains at the low level.

high level.

  For the second frame, the first SIG pulse has a duty cycle of 25% and the second has a

  cyclical ratio of 75%. The lack of identity can result

  for example, interference caused by noise. In such a case, the CLK and SYNC pulses are again produced as in the first frame and the LOS signal remains high. However, because the SIG pulses are different, the CONDUCTIVE / BLOCK signal retains its previous value, which in this case is the low level. In the third frame, the SIG pulses both have a duration corresponding to a duty cycle of 75%, which indicates that the signal CONDUCTIVE / BLOCK representing the state

  of the switching device must rise to the CONDUC-

  TOR. This occurs at the moment of the rising edge of the clock pulse following the second SIG pulse. For the fourth frame, the control pulses are no longer identical, so that the CONDUCTIVE / BLOCK line remains at the high level. The fifth frame lowers the CONDUCTIVE / BLOCK line to a low level, with the appearance of redundant pulses having both

  cyclical 25%. The sixth GIS pulse frame

  takes four pulses with a duty cycle of 75%.

  The duration of the sixth frame is somewhat prolonged

  to accept the four pulses and "the impulse

  The first and second SIG pulses bring the CONDUCTIVE / BLOCK signal high.While this is not shown, the third pulse of the frame causes HLS to go high, simultaneously to the front.

  amount of the resuiting clock pulse, and the

  third pulse of the frame makes DEF high.

  In addition to the conductive /

  As a default state and last state conservation, the GIS control signal provides synchronization for the return of status or diagnostic data to the microcontroller. The state encoder 125 accepts, as a

  that input signals, six states of switching devices

  present on the driver bus 115 and coming from

  of the control and detection section 113, in comparison

  bit pattern CONDUCTOR / BLOCK, DEF and HLS. The state encoder 125 combines these input signals to form a four-bit coded state s which is applied to the state latch network 127. The data selector 129 is a one of four type selectors that accepts the four bits of data from the state latch network 127 and which then sequentially outputs this information

  four-bit status to the microcontroller, through the

  The output signal of the two-bit counter i21 indicates the count of the SIG pulses and it controls the data selector 129 so that the latter emits a bit for each received SIG pulse. The four bits are encoded so that the first bit (XO) indicates whether or not there is a fault condition, and the second bit (X1) indicates whether a voltage appears on the output load or not. If a fault is present (X0 = 0), the third and fourth bits (X2 and X3) indicate the nature of the fault. If no fault has occurred (X0 = 1), the third bit indicates the value corresponding to the last state conservation and the fourth bit indicates

  the value corresponding to the state taken by default.

  The microcontroller 90 (FIG. 3) determines the amount of information to be received from the transmission section 111, based on the number of pulses per frame contained in the control signal, GIS, which

  is transmitted to the transmission section 111. The micro-

  controller reads the status signal on line M, immediately

  After sending a rising edge of the SIG signal on line D, the number of pulses per frame in the control signal and the number of status bits read back per frame are the same. Normally, the microcontroller emits two pulses per frame and it reads in return X0 and X1. If X0 indicates a fault, the microcontroller switches to four pulses mode.

  frame, so that it can read a fault message

  held in bits X2 and X3. In the absence of a fault, one can also use the four-pulse mode to read and write in the HLS 123 and in the rocker

  by default, 124. In such a case, the third

  second and fourth GIS pulses position or

  respectively, latch HLS 123 and the state latch taken by default, 124, and bits X2 and X3 of the signal

  indicate the state of these two flip-flops.

  The control and detection section 113 of FIG. 4 comprises a switching logic circuit 133, a comparator circuit 135 and a gate drive circuit 137. The switching logic circuit 133 receives the DRIVER / BLOCK signal generated by the switch section. 111 and, depending on the state of other input signals, apply a gate signal corresponding to the gate terminal of a power switching device, by

  through the gate driver circuit 137. The

  positive power switch is preferably an insulated gate transistor, which will be described hereinafter

more complete way.

  Among the other signals that are applied to the switching logic circuit 133 are signals representative of the level of the supply voltage and

  the temperature of the power switching device

  this. Signals representing the tension of sour food

  tion and charge as well as the charging current are

  applied as input signals to the comparator circuit

  135. The comparator circuit 135 generates a set of signals that indicate the level of the charging current.

  compared to a lower limit, an intermediate limit

  preselected upper limit and limit. The cir-

  cooked comparator 135 also produces a signal

  the level of the voltage of the load compared to the level of the voltage of the supply line and, in the case of the ac supply, a signal which

  indicates the zero crossing of the AC voltage.

  All these signals are applied as

  to the switching logic circuit 133, through the

  A further input of the switching logic circuit 133, labeled ALTERNATIVE / CONTINUOUS, is provided to pre-select the operation in an alternative mode.

or in continuous mode.

  The switching logic 133 generates the set of diagnostic signals applied to the state encoder 125 via the six-wire bus 115. This set of diagnostic signals is constructed from the voltage level and current supplied by the comparator circuit 135, and from the temperature and supply voltage signals. The six diagnostic signals can be used to indicate,

  for example: (1) a charge is open or disconnected

  ted; (2) the load exceeds a first upper limit value, which requires immediate response for protection purposes; (3) that a charging current exceeds a second upper limit value, which requires

  reaction for the purpose of protection only if the

  rant remains above the limit for a certain preselected period; (4) the voltage of the load has been or has not been applied; (5) the relative level of the supply voltage; and (6) the relative temperature of the

  power switching device.

  The grid signal from the section of

  command and detection 113 can control various cir-

  cooked input / output switching. For example, it is possible to use for the input switching circuit /

  output of the switching means constituted by

  field effect sistors or thyristors. In any case, a preferred switching circuit comprises a shunt current branch which comprises means for producing a signal representative of the current applied to the current.

  a connected load. However, switching circuits

  most preferable use a gate transistor

isolated, or abbreviated TGI.

  IMT is generally a semi-

  power conductor whose conductor can be controlled

  and blocking. It is thus possible to unlock the TGI and block it by means of its gate terminal. Some versions

  of the TGI include a so-called simulation section of

  which is a section of the IMT intended to convey

  a proportional fraction of the total current of TGI. The simulation section is advantageous in that it can be used to control the total current without resorting to high power dissipating shunt resistors for current sensing. A single gate signal controls the flow of current in both

  the main section of an IMT and in its simulation section

  lation. The insulated gate transistor is described (although under a different name) in an article by BJ Baliga et al., Titled "The Insulated Gate Rectifier (IGR): A New Power Switching Device", IEDM 82 (December 1982), pages 264 - 267. An IMT with a simulation section

  is the subject of the US patent application.

  No. 529,240, assigned to the plaintiff. Figs. 7A-7C show various input / output switching circuits using TGIs that can be used in the system.

I / O that is described here.

  In the DC source circuit of FIG. 7A, the gate signal is applied to the gate terminal 140 of a P-channel TGI 141, which includes a transmitter 142 for a main current section and a

  transmitter 143 for a simulation current section.

  The positive terminal of the DC power source is directly connected to the main transmitter 142, and is connected through a current detection resistor 145 to the emitter 143 of the simulation section. The collector of the TGI device is externally connected to one end of the combination in parallel of a surge suppression diode, 147, and a pre-charge resistor 148. The opposite end of the diode combination 147 and the pre-charge resistor 148 is returned to the negative terminal of the DC power source. The connection point of TGI 141

  and the diode / pre-charge resistance combination, constituting

  kills the INPUT / OUTPUT terminal 149. Although in actual use, an input device and a load would not be connected simultaneously, a load 150 is represented between the INPUT / OUTPUT terminal 149 and the load return terminal. (That is, the output) 152, and an input device 153 is shown between the INPUT / OUTPUT terminal 149 and the input return terminal 155. The terminals of

  155 and 152 are respectively connected

  in common with the positive and negative lines of

  the power source continues. The resistance of

  charge 148 has a relatively high ohmic value and the

  current sensing resistor 145 has an ohm value

  relatively low, as well as the pre-charge and current detection resistors that are used in the circuits of FIGS. 7B and 7C. For example, for a source of 120 volts, the pre-charge resistance 148 may be of the order of 20 k-L and the resistance of

  current detection 145 may be of the order of 10-.

  When operating the circuit of Fig. 7A as an output circuit, charge current is controlled by unblocking and blocking the TGI 141 at appropriate times. The load current flows from the power source, passes through the TGI 141 and the load, and returns to the source. The control of the charge current is facilitated by the TGI simulation section, which provides a signal representative of the current of the

  load at the connection point between the detection resistor

  current 145 and the transmitter 148. A signal of

  load, confirming that the voltage of the load

  is actually applied, is taken at the point of con-

  connection of the pre-charge resistor 148 and the collector of the TGI 141. A supply voltage signal is taken

  on the opposite end of the pre-charge resistor 148.

  Surge suppression diode 147 provides a shunt for reverse currents produced by inductive loads. When the circuit of Fig. 7A is operated in the input circuit, the TGI is kept in a locked state. The state of the input device 153 is then detected.

  (open or closed) by controlling the voltage that is

  located at the terminals of the pre-charge resistor 148. This status signal is controlled via the line

load voltage.

  The input / output circuit of FIG. 7B, operating in DC absorption circuit, comprises the same functional elements as the source circuit of FIG. 7A, but in a configuration

  somewhat different. When running this cir-

  In the output circuit, the load 147 is connected between the INPUT / OUTPUT terminal 158 and the load return terminal 159. The TGI 161 are switched to the conductive or blocked state to control the load current. It should be noted, however, that TGI 161 is an N-channel TGI. The collector terminal-is connected to one end of the parallel combination of a suppressor diode.

  overvoltages 165 and a pre-charge resistor 167.

  This combination is connected in parallel to the terminals 158 and 159 to which the load 167 is connected. A current detection resistor 168 is connected in series between the emitter of the simulation section and the

  negative terminal of the DC power supply.

  The transmitter of the main section is directly

  connected to the negative terminal of the power supply con-

  tinues. A current signal of the TGI, representative of the

  In this case, the charge current is taken at the connection point of the current detection resistor 168 and the emitter 163 of the simulation section. The load voltage signal is taken from the INPUT / OUTPUT terminal 158, and the supply voltage signal is taken from the positive terminal of the DC power source, which is

  also connected to the input return terminal 160.

  As with the DC source circuit, contemplated above, when the input / output circuit is used as an input, the TGI 161 is held in the off state and the state is detected. of the input device 170 by means of the voltage which is developed across the pre-charge resistor 167. This status signal is transmitted by the

voltage line of the load.

  In Fig. 7C, which shows an alternative input / output circuit, P-channel and channel TGIs are used.

  N, connected in parallel, respectively bearing the

  175 and 176. The gate signal of the TGIs is applied to a gate control circuit 178 which produces two

  simultaneous gate control signals (of opposite polarity)

  to control (ie block and release)

  quer) the TGIs 175 and 176. A series-connected current detection resistor 180 is associated with the simulation section of the TGI 175, and a detection resistor

  connection connected in series is associated with the sec-

  Simulation simulation of the TGI 176. A current signal of the TGIs, representative of the charge current flowing in the TGI, is obtained by comparing the signals developed across the two current detection resistors.

  and 181 in a differential amplifier 183. A circuit

  firing suppression of transient voltages 185 is con-

  connected in parallel with the main section of the TGI and between the INPUT / OUTPUT terminal 186 and the return terminal 187

  of the input device. The return terminal 187 is also

  electrically connected in common with one side of the

  AC power line. A resistance of

  load 189 is connected between the INPUT / OUTPUT terminal 186 and the load return terminal, 190. The latter

  190, is connected to the other side of the power line.

alternative.

  When the circuit of FIG. 7C operates in an output circuit, the gate control circuit 178, operating under the control of the gate signal of the TGIs,

  command the IMTs 175 and 176 so that they are simultaneous

  conductive or blocked, which circulates or blocks

  than the current of the load. The load 191 is connected between the INPUT / OUTPUT terminal 186 and the return terminal of

  O10 the load, 190. During the circuit operation of

  the load 191 is not connected, and an input switching device 192 is connected between the INPUT / OUTPUT terminal 186 and the return terminal 187. The TGIs 175 and 176 are kept in the off state and the state of the input switching device 192 is determined by the presence or absence of a voltage on the voltage line of the

  charge; the presence of a voltage indicating an interruption

closed input driver.

  Referring to FIG. 8, which shows in more detail the control and detection section, it can be seen that the DRIVER / BLOCK signal from the sec-

  transmission is applied to an entrance to the door

  NAND 195, to the inverter 196 and to the restaurant inputs

  The other input of the NAND gate 195 receives the output signal of the NAND gate 201. The first input of the NAND gate 201 receives a signal which is high or low. at the low level, depending on whether the output circuit should operate in an output circuit

  AC or DC-current output circuit

  tinu. Note that this signal can be generated by a switch or jumper suitably connecting the AC / DC selection line to a high or low reference value. The remaining input of the NAND gate 201 receives a signal from the zero crossing detector 202, through the inverter 201a, to indicate the cases in which the voltage of the AC power line ( for AC output circuits) is within a certain range in the vicinity of the zero voltage. Thus, in the case of an AC output, the NAND gate 195 transmits the CONDUCTIVE / BLOCK signal only during a zero crossing of the AC line voltage. The zero crossing detector 202 may be any one of a number of conventional circuitry producing a signal which indicates that the AC input signal is within a certain range.

  corresponding to a zero crossing. For a ride out

  continuously, the state of the NAND gate 201 allows the

  signal transmission CONDUCTIVE / BLOCKED by the door NO-

  AND 195. The DRIVER / BLOCK signal from the NAND gate 195 is applied to the positioning input of the flip-flop 203. The signal of the Q output of the flip-flop 203 is applied to one of the three inputs of the ET 205,

  whose output provides the grid signal of the TGIs.

  The two remaining entries of the ET 201

  are attacked by the outputs Q flip-flops 198 and 199.

  Both flip-flops 198 and 199 are restored when the CONDUCTIVE / BLOCK signal goes to the off state. The rocker

  198 receives a positioning signal from the compa-

  controller 207 whenever the current of the TGI exceeds a preselected value. Indeed, a signal representative of the TGi current is applied to the inverting input of the

  comparator 207, while a reference voltage

  An excessive level of TGI current is applied to its non-inverting input. For example, the tension

  reference value may have a value corresponding to a

  In a similar way, the flip-flop 199 receives on its positioning terminal a signal which comes from the

  power control circuit 209. The control circuit

  feed control 209 may be any of the

  many well-known means that produce an indi-

  as to whether the DC supply voltage is greater or less than a certain preselected value. By

  therefore, from a functional point of view, a voltage of

  Low power or an excessively high IMT current will invalidate the AND gate 205. This forces the TGI (which is connected to the output of the AND gate 205) to a locked state, and the TGI remains in this state until it is closed. 'to this

  that the default is deleted.

  The Q output signal of the flip-flop 198 is used as an over-current interrupt signal, and this signal is one of six switching device status signals that are applied to the bus driver. 115 (Figure 4). In addition to being directed to the AND gate 205, the output signal Q of the flip-flop 199 is also applied to an input of the

* logic gate 210. The signal coming from the con-

  supply gate 209 is applied to the remaining input of the logic gate 210, so that the output signal

  this door indicates the status of the continuous power supply.

  This output signal is also one of six signals

  switching device status.

  The flip-flop 203 receives a recovery signal from the output of the NAND gate 212. Of the two input signals of the NAND gate 212, the first

  is the inverted CONDUCTIVE / BLOCK signal from the

  196, and the second comes from the NAND gate 213.

  The AC / DC selection signal is applied to an input of the NAND gate 213, and the output signal

  comparator 214 is applied to the other input by the

  intermediate of the inverter 201b. The comparator 214 is a control comparator for the current of the TGI, and the

  The current signal of the TGI is applied to its inverse input.

  Seuse. A reference voltage corresponding to a relatively small minimum value of the TGI current (e.g., 0.05 A) is applied to the non-inverting input of the comparator 214. This combination, including the NAND gate 212, inverter 196, the NAND gate 213 and the comparator 214, acts via the flip-flop 203 to prevent switching of the TGI (in a

  alternative mode of operation) unless the

  load of the TGI is less than the reference value. The current signal of the TGI is also applied to the non-inverting input of comparator 215, where it is compared with a current value of

  intermediate reference. The reference current value

  this intermediate (corresponding for example to 2 A), is

  applied to the inverting input of the comparator 215.

  However, a delay network including resistance

  216 and the capacitor 220 is also connected to the

  non-inverting comparator 215. The combination of

  the resistor 216 and the capacitor 220 delay the voltage

  on the non-inverting input of the comparator 215, with respect to the current of the TGI. Thus, the output of the comparator 215 will be affected only if the current of the TGI exceeds the reference value for a prolonged period. If the

  overcurrent is simply short-lived, no change

  the state of comparator 215 does not occur. The output of comparator 215 and the output of comparator 214 are provided as status signals of the switching device. These signals are used

  as diagnostic signals and they respectively indicate

  if the current of the TGI is higher or lower than the intermediate reference value, and if this current is higher or lower than the lower reference value, so that the micro-controller can trigger a

corrective action, if necessary.

  In case the current of the TGI exceeds the intermediate reference value, corrective action

  undertaken only if the value and duration of the surin-

  The current of the load may exceed the intermediate reference value for a period of time before corrective action is taken. In some cases, it is preferable to suppress the delay network (i.e. resistor 216 and capacitor 220) and implement the delay function by

  software in the micro-controller. The comparison of the

  of the IGT or load, with the inferior reference

  re, or minimum value, allows to generate a signal of

  diagnosis (corresponding for example to 0.05 A) which indicates

  only if a load is connected, or if it is open, in the case where it is connected. The signal of

  the Q output of the flip-flop 217 is a status signal of the dis-

  positive switching, having a diagnostic function, which indicates whether a voltage is present or not on the connected load. The positioning input terminal of

  the flip-flop 217 is connected to the output of the NOR gate

  AND 218. The NAND gate 218 receives on its first input terminal the inverted AC zero crossing signal from the inverter 219 and receives on its remaining input terminal the output signal. The comparator 221 compares the voltages of the supply line and the load to produce a logic signal which indicates whether the voltage of the load

  is greater or less than a pre-selected percentage

  tion of the supply line voltage. For example, the output signal can indicate whether the voltage of the load is greater than or less than 70% of the voltage of the supply line. The supply line and load voltages are applied by respective input resistors 223 and 224 to the input terminals of the comparator 221. Functionally, the NAND gate 218 prevents a change of state of the output of the

  switches every time the voltage of the power supply line

  alternative is within a certain range

  around zero volts. Thus, in fact, the decisions concerning

  the state of the load voltage are not taken each time the supply line voltage

  alternative is close to a zero crossing.

  Flip-flop 217 is restored by the output signal of NAND gate 226. The first input of NAND gate 226 receives the inverted zero crossing signal from inverter 219, and the second input receives the output signal of the comparator 221,

  after it has been inverted by the inverter 227.

  The signal remaining among the state signals of the switching device is provided by the temperature control circuit 229 and indicates the relative temperature of the switching device constituted by a

  TGI (or several TGIs in the case of an alternative output

  tif). The temperature control circuit 229 preferably consists of a simple temperature detector.

  Junction P-N, 229, which is in good thermal communication

  only with the TGI. For example, the temperature detector 229 may be selected to indicate the

  fact that the temperature of the TGI has exceeded 150 C.

  Fig. 9, formed by Figs. 9A-9C, shows in more detail one embodiment of the transmission section (111 in Fig. 4). The output signals that timer 117 produces are

  elaborated by an RC delay network which is

  killed by a resistance 300 and a time capacitor

  The resistor 300 and the capacitor 301 are connected in series between a positive voltage source + V and a common circuit point. The connection point

  between the resistor 300 and the capacitor 301 is connected

  at the inverting input of the comparator of the signal LOS, 303, and at the non-inverting inputs of the comparators of the signals SYNC and CLK, respectively bearing the references 304 and 305. The resistors 308-312 form a network

  voltage divider in which the resistors are

  connected in series between + V and the common circuit point.

  Each connection point between resistors 308-312

  The divider network thus provides a voltage reference.

  The highest reference voltage, taken from the point

  connection between the resistors 308 and 309, is applied

  at the non-inverting input of the comparator 303. The other voltage reference values, in descending voltage level order, are correspondingly applied to the inverting inputs of the synchronization comparator 304 and the clock comparator 305, and

  at the non-inverting input of the control comparator 314.

  The collector terminal of transistor 315 is

  connected by the collector resistor 316 to the capacitor

  timer 301, whose other terminal is

  connected to the emitter of transistor 315. The conductive / off state of transistor 315 controls the charging / discharging cycle of capacitor 301 and is itself controlled by

  the output Q of the flip-flop 317. A resistor 318 is con-

  connected between the base terminal of transistor 315 and the

  tie Q of the flip-flop 317. The catering terminal of the

  flip-flop 317 receives the output signal from the com-

  control paratrooper 314. The control comparator 314

  continually compares the voltage across the

  timer 301 (which is applied to the input

  comparator 314), with the reference voltage

  from the point of connection of the resistors

311 and 312.

  When considering the functioning of the

  117, it can be assumed initially that the Q output of the flip-flop 317 is at a low level, which keeps the transistor 315 off, so that the capacitor 301 loads at a certain voltage level, and the comparator output control 314 is thus in the low state. Under these conditions, a rising edge of a pulse applied to the clock input of the flip-flop 317 via the splitter amplifier 320 causes a high level to appear on the output Q. This unlocks the transistor 315. which discharges the delay capacitor 301. Under the effect of the discharge of the capacitor 301, the signal of

  CLK output of comparator 305 is forced to a low level.

  The output of the comparator 304 is also forced to the low level, if it is not already low, and the output of the comparator of the signal LOS, 303, is forced to the high level.

  if she is not already at this level.

  The discharge of the capacitor 301 is detected by the comparator 314 whose output goes high, which restores the flip-flop 317. The Q output of the flip-flop 317 then goes to the low level, which blocks the transistor 315 and thus allows the capacitor 301 to start to

  recharge. When the charging voltage becomes

  Highly high, the clock comparator 305 is triggered, which produces a high level CLK signal. If the capacitor 301 can continue to charge, it reaches a certain voltage level which triggers first the comparator of the signal SYNC 304, then the comparator of the signal LOS, 303. The comparator of the signal

  SYNC, 304 is thus triggered by a "pulse failure"

  te ", and the comparator of the LOS signal is triggered by a loss of the SIG signal lasting for about 2.5 T, as

previously described.

  In Fig. 9B, the signals SIG and CLK are applied to the output data filter 119 which includes flip-flops 325 and 326, a NONOU-EXCLUSIVE gate 329, a NAND gate 328, an inverter 330 and gateways.

  331 and 332. The GIS and CLK signals are

  respectively applied to the inputs D and C of the flip-flop 325 which retains, at its output Q, the high or low state of

  the immediately preceding GIS pulse, so that

  compare the values of the first two impulses

  sions of a frame. When the clock pulse appears, the value of the signal GIS is high or low depending on whether the value of the pulse corresponds to a duty cycle.

  % or 25%. For a pulse having a cyclic ratio

  than 25%, the Q output of flip-flop 325 is forced low; for a pulse having a duty cycle of 75%, the output Q is high. There is thus in

  makes a sampling of the GIS value at each

  clock pulse. The value of the Q output of the flip-flop 325 is applied to an input of the gate

  NO OR EXCLUSIVE 329, and the value of the GIS signal is applied

  at his other entrance. The value of the current pulse

  and the values of the previous pulses are thus

  adorned in the EXCLUSIVE NOR gate 329, whose output is at a high level each time the input signals

are the same.

  The output signal of the NOR-EXCLUSIVE gate 329 is applied as an input signal to the NAND gate 328 which respectively receives the count pulses SO and S1 on its other two inputs. The SO, SO, S1 and S1 values, taken together, indicate which

  is the pulse that is received, in a frame. As a result

  That is, if the first two pulse values of a frame are the same and if the second pulse is received, the output of the NAND gate 328 assumes a logical zero value. At all other times and under other conditions, the exit of NAND gate 328

corresponds to the logical value one.

  A logical zero at the exit of the NAND gate

  328 indicates the concordance between the first two

  res pulses of a frame, and a valid condition to update the Q output of the flip-flop 326. For this purpose, the output signal of the NAND gate 328 is applied.

  parallel to the input of inverter 330 and on terminals

  Controlling nons of the transmission gates 331 and 332. A logic zero on the output of the NAND gate 328 causes the transmission gate 332 to be blocked and the transmission gate 331 to be released, the latter transmitting the gating signal. GIS command at the input D of the flip-flop 326. The appearance of a clock pulse then transfers the new value to the output of the

flip 326.

  If, on the other hand, there is no redundancy in the first two pulses of a frame, the output of the NAND gate 328 is in logic state one, which keeps the transmission gate 331 in the off state. and the transmission gate 332 in the unlocked state. In these

  conditions, the output signal of the flip-flop 326 is

  by gate 332, so that the flip-flop 326

  serve the previous output state. The signal of the Q output of the flip-flop 326 represents a filtered version of the signal DRIVER / BLOCK which is then transmitted to the selector

exit 120.

  In addition to the filtered DRIVER / BLOCK signal, the output selector 120 receives the LOS signal, the last state HLS conservation signal, and the complemented default status signal, DEF. The output selector 120 (which includes the NOR gates 335-337 and the OR gate 338) has the function of selecting a desired value for the DRIVER / BLOCK output signal, in the event of a loss of communication between an I / O point and the microcontroller, i.e. a loss of the GIS control signal. If such a communication loss occurs, the output selector 120 produces a DRIVE / BLOCK output signal which is either the last transmitted value of SIG or a default value, depending on the HSL and DEF signals which are applied in

  as control input signals to the selector 120.

  The HLS and DEF signals are respectively generated by the last state latch, 123, and the state latch taken by default, 124. These latches are substantially identical but respond to different pulses in a control signal frame. . The HLS 123 signal toggle includes a NAND gate 340, transmission gates 342 and 343, an inverter 344 and a

  flip-flop 345; the state latch taken by default 124 (FIG.

  9C) comprises a NAND gate 348, transmission gates

  mission 349 and 350, an inverter 352 and a flip-flop 353.

  Because the circuit configuration and the operation

  of these two flip-flops are practically identical,

  only the flip-flop of the HLS signal, 123, requires an explanation

detailed description.

  The HLS signal flip-flop, 123, responds to the third pulse in a control signal frame (i.e., it responds to high level pulses SO and S1 from the two-bit counter 121), in a way that allows the update of the output of this flip-flop. The pulses S0 and S1 are applied as input signals to the NAND gate 340 whose output controls the transmission gates 342 and

  343. The output signal of the NAND gate 340 is applied

  to a first set of opposite control terminals of

  transmission gates 342 and 343, and the inverter 344.

  The output signal of inverter 344 is applied to a second set of opposite control terminals of the gates of

  transmission 342 and 343. Thus, during the

  transmission gate 343 is unlocked and the transmission gate 342 is blocked by the appearance

  of a third pulse in the frame of the signal of com-

  mande. Since the control signal is applied to the input of the transmission gate 343, this signal is transmitted to the input D of the flip-flop 345, which leads to

  day the HLS signal that is taken on the Q output of the bas-

  345. The output signal HLS is also returned to the input of the transmission gate 342, so that in the absence of a third pulse in a frame of the control signal, the HLS value remains stored. The clock signal is applied to the clock input of the flip-flop 345. The signal output signal of the HLS signal,

  123, is applied to the output selector 120.

  In comparison, the state latch taken by default 124 operates in substantially the same way,

  but responds to the fourth pulse in a frame.

  Thus, the default state latch reacts to the impulse

  SO and S1 of a control signal frame. It should however be noted that the output signal of the state latch taken by default, 124, is taken from the output

  of the flip-flop 353, so that the complementary signal

  re DEF is applied to the output selector 120.

  In normal operation, the output selector merely has the function of inverting and transmitting the control signal from flip-flop 326 and this signal then becomes the CONDUCTIVE / BLOCK output signal that is applied to the control and detection section.

  113 (Figure 4). However, in the event of loss of

  between the I / O point and the microcontroller (ie

  ie a loss of the GIS control signal), the signal of

  DRIVER / BLOCK output is forced to a desired state

  which is defined by the LOS and HLS signals. These latter two signals are applied as input signals to the output selector 120. In the case of a loss of communication, the output selector 120 maintains the last state or selects a state taken by default , according to what has been preselected. The preselection is performed so as to force the I / O point to a preferred state, or security, in case of loss of

Communication.

  The LOS and HLS signals are input signals of the NOR gate 335 whose output signal drives an input of the NOR gate 337. The second input signal of the NOR gate 337 is the signal which comes from the Q output of the flip-flop 326. The NOR gate 335 thus controls the NOR gate 337, in such a way that if either of the LOS or HLS signals is at a high level the NOR gate 337 simply inverts the control signal from the flip-flop 326. On the contrary, the LOS signal is low (loss of communication) and the HLS signal is also low, the output of the gate NO-OR 335 is high, which maintains a

  low level the output of the NOR gate 337.

  The LOS, HLS and DEF signals are applied to the NOR gate 336 whose output signal, as well as the output signal of the NOR gate 337, is applied as input signals to the OR gate 338 The output signal of the OR gate 338 is the control signal

  DRIVER / BLOCKED. Thus, in the presence of a loss of com-

  (LOS low) and in the absence of an order to keep the last state (HLS low), the output signal DRIVER / BLOCK of the OR gate 338 is selected to be the signal of default state DEF (that is, DEF is inverted by the NOR gate 336). The operation is thus such that if there is a loss of the communication and if the conservation of the last state is not selected, a condition taken by default is selected. Of course, by appropriately setting the flip-flop of the HLS signal 123 and the state latch taken by default, 124, it is possible to determine whether the last state is retained or whether the state condition

  default is selected.

  The foregoing describes in detail the direct path in the control and transmission section 111. As previously discussed, the return of the coded diagnostic information is performed by the state latch network 127 and by the data selector of one of four, 129. The coding of the information is described in detail in connection with FIG. 10; it is sufficient, however, to note here that the input signals X0-X3 applied to the state latch network 127 are coded so as to contain the diagnostic information and other information which must be returned to the microphone. -controller 90 of Figure 3. The network of state latches 127 may be a commercial device, such as the model MC14174 provided by Motorola Inc. The coded information, XO-X3, is stored in the network of latches of FIG. state 127 on the rising edge of the signal SYNC, which is also applied to the state encoder 125. A new data set is thus stored at each frame of the control signal. These data form a diagnostic signal that is representative of the parameters

I / O point operation.

  The data from the state latch network 127 is transmitted bit by bit by the selector of

  data type one out of four, 129, towards the micro-

  controller 90 through the splitter amplifier 360. The data selector 129 responds to the current value from the 2-bit counter 121 by providing the values of XO-X3 in order. So, for example, when the first pulse of a frame is

  received, the XO bit of the diagnostic data is transmitted simul-

  ously. The data selector 129 may be a device

  like the MC14052 model from Motorola, Inc. The figure shows a truth table for a

  state encoder such as encoder 125 of FIG. 4.

  Those skilled in the art can easily make an encoder according to the truth table of FIG.

  elements of classical combinatorial logic.

  In Figure 10, the input conditions are indicated horizontally at the top of the left part of the table. The possible values that each entry can take are shown below, in columns. In the table, "1" indicates that a value is true (which corresponds for example to a high level signal), "O" indicates that a value is not true and X indicates states " indifferent "(that is, the value can be" 1 "or" O ", with no effect). The 4-bit output signal (XO-X3) of the state encoder 125 is represented in the part

  right of the table, with bits XO-X3 distributed horizontally at the top of the four columns. Each horizontal line extending

  The four columns thus represent a 4-bit word that unambiguously defines the state of the I / O point. This 4-bit word constitutes the diagnostic data that is sent back to the microcontroller 90 of FIG.

  finally to the controller's CPU (Figure 1).

  As an example, the first line in the truth table shows a high level in the column of

  voltage, while the remaining columns correspond to

  lay on "indifferent" or indeterminate conditions.

  In these circumstances, the 4-bit output signal is

  unambiguously determined to be a signal that does not include

  only zeros. This 4-bit word only includes

  zeros indicates a loss of power to the I / O point.

  As a further example, line six shows that the given command controls the transition from the output to the conductive state, but that the output is in a short-circuit condition. Thus, a "1" appears in the first column, under the heading DRIVER / BLOCK, indicating that the I / O point must be in the conductive state,

  while there is simultaneously an indication of over-

  in the overcurrent column (column 6). The 4-bit output word for this condition includes zeros for all bits, except for X3 which is at level 1. There is similarly a set of fifteen 4-bit words that define the various conditions of the

I / O point.

Claims (11)

  1. Intelligent I / O system with localized diagnostic and diagnostic capabilities   decision, which can be used in a project management system   industrial process of the type comprising a central processing unit (CPU) (20) operable in accordance with a stored program to accept various signals representative of the state of the process (30), and   produce signals to control the process according to   the program and the state of the process,   in that it includes: (a) an input / output controller   tie (22) placed near the CPU (20) for controlling the exchange of signals therewith; (b) at least one input / output module (2426) for being placed in relative proximity to the process (30) being controlled, for accepting input signals representative of process parameters, and for applying output to controlled elements of the process (30), this input / output module comprising: (i) a set of input / output (I / O) circuits (37-39, 81-88), each of they being designed to operate selectively either as an input circuit for accepting one of the input signals or as an output circuit for producing one of the output signals; (ii) means for connecting conductors carrying the input and output signals between the I / O circuits (37-39, 81-88) and   the process (30); (iii) a control unit of the   tion (36, 90) which can operate in accordance with a set of stored instructions, for the purpose of providing an ordered exchange of signals with the I / O controller (22), as well as controlling the circuitry I / O (37-39, 81-88) and to test the functional character   of these circuits; (iv) means (40) for connecting   ter the I / O circuitry (37-39, 81-88) to the unit   operating control (36, 90); and (c) a link   transmission sound (28) which interconnects the input / output controller (22) and the one or more input / output modules   (24-26), to route signals between them.   Intelligent input / output system according to claim 1, characterized in that the selection of the operation of each I / O circuit (37-39, 81-88) as an input circuit or as an output circuit, is done   according to instructions from the ITU (20).   An intelligent input / output system according to claim 2, characterized in that each I / O circuit comprises a transmission section (111) and a control and detection section (113); and in that the   transmission section (111) receives data from   command from the operation control unit   (36, 90) these data control the operation of the   fired I / O in an input circuit or an output circuit, and transmits to the operation control unit (36, 90) status and diagnostic data relating to   the I / O circuit (37-39, 81-88); and the section of   Mande et de detection (113) provides at least a portion of the status and diagnostic data and is switched on or off, or is maintained   in the off state, depending on whether the I / O circuit is respec-   selected to operate in an output circuit or in the input circuit.   Intelligent input / output system according to claim 3, characterized in that the means for   connect all the I / O circuits (81-88) to the uni-   Operation Control Unit (90) includes a   of conductors (95-102; M, D) for each I / O circuit (81-88), and one of the conductors (M) of the pair carries control signals to the E / I circuit (81-88). S, while the other (D) carries status and diagnostic information of the associated I / O circuit to   the operation control unit (90).   Intelligent input / output system according to claim 4, characterized in that the connection of   transmission (28) is a serial data link.   6. Input / output system intended for use   with an industrial programmable logic controller of the type comprising a central processing unit (CPU) (20) and an input / output control unit (22) which controls a signal exchange with the CPU (20), the CPU producing control signals for controlling various processes (3d)   according to the recorded program and the signals from   which are representative of the functional states of the processes, characterized in that it comprises: a set of input / output modules (24-26) for the exchange   signals between the PLC and the pro-   cessus (30), each module (24-26) in the set includes   an additional set of input / output circuits   (37-39, 81-88), a micro-controller (36, 90), means (40) for interconnecting the micro-controller.   each of the input / output circuits, and the means   seen for the connection of drivers who carry   input and output signals between the set of cir-   cooked in / out (37-39, 81-88) and the process   (30), and each input / output circuit can function   selectively in an input circuit or an output circuit under the control of the CPU (20); and a transmission link (28) interconnecting each input / output module (24-26) and the input control unit /   output (22) for carrying signals between each module   the input / output and the input / output control unit tie.   7. Input / output system according to the claim   6, characterized in that each input / output circuit (37-39, 8188) comprises a transmission section (111) and a control and detection section (113); and the transmission section (111) comprises means (117) for receiving a recurrent set of control signals (SIG) from the microcontroller (36, 90), means (119, 120) responsive thereto. of control signals by applying a DRIVER / BLOCK signal to the control and detecting section (113), means for receiving a first set of diagnostic signals from the control and detection section (113), means (125) for encoding the first set of diagnostic signals to produce an encoded set of diagnostic signals, and means (127, 129) for transmitting the encoded set of diagnostic signals to the microcontroller (36). , 90); and the control and sensing section (113) comprises means (133, 137) responsive to the DRIVER signal /   BLOCKED by determining the conductive or blocked state of the cir-   fired input / output, and means (135) responsive to operating conditions of the input / output circuit for producing the first set of diagnostic signals.   - 8. Input / output system according to the claim   7, characterized in that the means responsive to the CONDUCTIVE / BLOCK signal comprise a   insulated gate (TGI) (141, 161, 175, 176).   9. Input / output system according to the claim   8, characterized in that the TGI (141, 161, 175, 176) is of the type having a main stream section   (142) and a simulation current section (143).   10. Input / output system according to the claim   7, characterized in that the means provided for the   connection of conductors comprise a set of   (155, 152; 160, 159; 187, 190) having a terminal for each input / output circuit for connecting the   conductors going to   trea (153, 170, 192) and to output load devices (150, 157, 191); and a set of common terminals (149, 158, 186) for connecting conductors to input detection devices (153, 170, 192) or output load devices (150, 157, 191), according to that the input / output circuit operates as an input circuit or as an output circuit.   11. Input / output system according to the claim   10, characterized in that the means responsive to the CONDUCTIVE / BLOCK signal comprise an insulated gate transistor (141, 161, 175, 176) having a main current section (142) and a current section of simulation (143).