GB2178878A - Alarm system with priority override - Google Patents

Abstract

An alarm system suitable for fire detection, includes a plurality of transceiving units which each include a sensor for monitoring a parameter pertaining to, e.g. fire detection. A control unit transmits a sequence of polling signals to the transceiving units whereby the state of each transceiving unit is sequentially determined. Many transceiving unit detects a change in the monitored parameter which represents a priority condition, it communicates a priority signal to the control unit during the polling sequence. The priority signal is included in the response of the last-polled transceiving unit in such a way as to avoid interfering with the usual information contained in its response. A transceiving unit in a priority condition thereby interrupts the polling sequence and thereby reduces the response time of the system in detecting a priority condition. The priority signal is preferably a current pulse which occurs in a given position during a sequence of synchronizing pulses. The address of a priority transceiving unit may also be included without interfering with the usual response of the last-polled transceiving unit. <IMAGE>

Description

SPECIFICATION Alarm system with priority override This invention relates to an alarm system in which a plurality oftransceiving units are each provided for monitoring at least one parameter, and in which a control unit transmits a sequence of polling signals to the transceiving units whereby the state of the par ametermonitored by each ofthetransceiving units is determined.

Alarm systems of the above type are known for use in fire detection, i.e. wherein each transceiving unit includes a sensorfor monitoring a parameter pertaining to fire detection. The transceiving units are connected to the control unit by means of a two wire cable. The cable supplies power and carries messages between the control unit and each of the transceiving units. The control unit polls each of the transceiving units in turn in order to receive messages concerning the status or magnitude of the parameter monitored by each of the sensors. Each transceiving unit responds to a polling signal containing its unique address so that the respective messages are re ceived seq uential Iy by the control unit. The control unit can also send data to thetransceiving units to initiate action, such as a test sequence.

If there were 100 transceiving units connected to the two wire cable, it could take several seconds two poll all ofthetransceiving units. Whilstthis is gener- ally acceptable in a fire alarm system comprising smoke and heatsensors, it is unacceptable in systemps which include manual call points (break-glass switches) where alarms should be recognised in less than one second in accordance with BS 5839, Part 1.

A response time of less than one second can be achieved by increasing the data rate. However, this requirestheuseofhigherfrequenciesandthewaveforms which are received are likely to be attenuated and/or distorted unless correctly terminated screened cable is used. Therefore, an increase in data rate can prove expensive, unreliable and difficultto implement in practice.

Oneway of avoiding delay in signalling a priority condition is to arrange for any one of a plurality of manual call points to cause a significant and constant increase in line current to occurwhen the call point is operated. This increase in line current can be recognised by the control unitasa priority or ""interrupt" signal. The control unit can then poll only the manual call points to determine which point has been activated. If the number of call points is less than about 20, then the activated call point can be identified in less than one second. However, this method is not always practical because the multiple operation of call points may produce a line voltage drop which impairs the address and response messages.

The present invention seeks to overcome these problemsofthepriorart.In particulartheinvention seeks to reduce delay in responding to a priority condition without affecting the usual transmission of messages in the polling sequence.

In accordance with the invention, when any ofthe transceiving units detects a change in the monitored parameterwhich represents a priority condition (herebytermed the ""priority transceiving unit"),the priority transceiving unit communicates a priority signal to the control unit during the polling sequence. The priority signal is included in the response ofthe transceiving unit which received the last polling signal (herebytermedthe ""last-polled transceiving unit"). The priority signal is included in the response ofthe last polledtransceiving unit inwuch a way as to avoid interfering with any other information contained in its response.

The polling signal suitably includes address data and synchronizing data in the form of voltage pulses.

In this case, the response preferably includes current pulses which occurbetweenthevoltagesynchroniz- ing pulses. The presence of at least one current pulse in a given position indicates a priority condition. For example, under normal operating conditions (where none bf the transceiving units has detected a priority condition), no current pulse is generated at the start of the response message of the last-polled transceiv- ing unit.However, if any one ofthetransceiving units responds to a priority condition (i.e. detected by its respective sensor), it generates a current pulse at the start of the response message ofthe last-polled transceiving unit and this current pulse is rapidly detected by the control unitthereby overriding or inter ruptingthe polling sequence.

Embodiments ofthe invention will now be described in more detail with respect to the accompanying schematic drawings in which: Figure lisa block diagram of an alarm system, Figure2is a block diagram of a transceiving unit in accordance with an embodiment of the invention, Figure3is a block diagram ofa logic unit used in the transceiving unit, and Figure4is a pulse diagram illustrating a message transfer sequence in accordance with an embodiment of the invention.

Referring to the drawings, Figure 1 schematically illustrates a plurality oftransceiving units 1a,1 b,1 c, etc, connected to a control unit 2 by means of a twowire cable 3. Cable 3 supplies powerto each ofthe transceiving units and also conveys messages between the control unit and the transceiving units. As shown in Figure 2, each transceiving unit (1) includes at least one sensor4(notshown in detail,butof known construction) which monitors a given parameter or parameters pertaining to fire detection and which is capable of responding to a priority condition (e.g. in the case of a fire). The term ""transceiving unit" is used generally herein to denote a unitwhich includes, for example, a sensor, a transceiver and other circuitry and which is provided for location at a given point (e.g. in a building) to detect a priority condition (such as a fire) at that point. Whilstthe preferred embodiment of the invention will be described with reference to fire detection, it will be understood that the invention could be applied elsewhere, e.g. in a burglar alarm system. In the latter case, a priority condition is one where a transceiving unit detects an intrusion.

As shown in figure 2 each transceiving unitembodying the invention includes a sensor 4; a line interface unit5which is connected to sensor4and to a linear circuitry unit6 having an LED drive 7; and a logic unit8 connectedtothe linear circuitry unit 6 and to a bank of switches or links 9. The linear circuitry is also connected to the sensor4 and it provides an analogue-to-digital interface with the logic unit8 besides supplying messagessignalsto and from the line in terfaceunit5which is connectedto cable 3.

Each transceiving unit (1) includes means for decoding its unique address from a sequence of polling signalstransmitted bythe control unit 2. On decoding its address, a transceiving unit causes data to be sent back to the control unitconcerning,e.g. a para meterwhich is monitored by the sensor. Each transceiving unit thereby sequentially returns data to the control unit aboutthe parameterbeing monitored at the locations ofthe respective sensors.

As shown in Figure 1,thecontrol unit2 includes a logic unit (e.g. a microprocessor) 10 with a serial data input 11 and a serial data output 12; a linevoltage regulator 13; a voltage pulse modulator 14which is connected to data output 11; and an input current pulse monitor 15 connected to data input 10. Units 13,14 and 15areconnected, in series, to cable3 as shown. The logic unit 10 is provided with software suitable for controlling all aspects ofthe system. The line voltage regulator 13 supplies powerto all ofthe transceiving units. The voltage pulse modulator 14 receives output serial data from the control unit 10 and modulates the supplyvoltage accordingly.The input current pulse monitor 15 detects the current pulsestransmittedbythetransceiving unitsand transfers the input data to the logic unit 10 in serial form. Figure4illustrates a messagetransfersequence between the control unit 2 and one ofthe polled transceiving units (1). The message sequence is initiatedfromthecontrol unitwithalong duration voltage pulse 16 that serves to reset the transceiving system. Pulse 16 is followed by ten bits in the form of positive-going pulses whose mark-to-space ratio is varied according to the bit being transmitted. Seven of these bits represent the address ofthetransceiving unit to which the polling signal is directed. Three of the bits control an output port on thetransceiving unit.Aseries of synchronizing voltage pulses 19, generated by the control unit and having a constant mark-to-space ratio, follow the ten bit sequence.

After receiving the ten bits, the respective transceiving unit decodes the seven bit address and then transmits a response message 17, consisting of twenty-one bits, to the control unit. These twentyone bits are in form ofthe current pulses 19. The current pulses 18 cause corresponding voltage drops 20 to occur and these voltage drops are detected across a monitoring resistor by the input current pulse monitor 15 in the control unit2to providethe data inputto logic unit 10. Voltage drops 20 are interleaved with the synchronizing pulses 19 as shown in Figure 4. A slight delay preferably occurs between the trailing edge of a synchronizing pulse and the leading edge of a current pulse so as to avoid any corruption ofthe synchronizing pulse.Likewise,thetrailing edge of a current pulse does not occurtoo close to the leading edge of the next synchronizing pulse.

Asufficient numberofsynchronizing pulses 19are are transmitted in orderto contain an initial bitwhich represents the presence or absence of a priority signal 21 andtwentyfollowing bits relating to the status of the polled transceiving unit. Each transceiving unit switches the current on the supply cable 3 in orderto providethetwenty-one bit sequence wherein the presence of a current pulse corresponds with a logic ""1" and the absence of a current pulse corresponds with a logic ""0".

The presence of a current pulse or logic ""1" in the priority signal position (i.e. between the first two syn- chronizing pulses 19) represents a priorityorinter ruptcondition in one ofthetransceiving units. The next three bits indicate the type of sensor connected to the transceiving unit, the following three bits indi catethe status of an output port on the transceiving unit, and the next seven bits are the digitised value of the signal at an analogue input ofthetransceiving unit. The messagetransfersequence isthen concluded with another long pulse 22 which is identical to pulse4 which preceded the messagetransfersequence.

Transmission is half-duplex and is arranged in a "burst" type of format. In normal operation, messages follow directly, one afterthe other.

Under usual operating conditions, none ofthe transceiving units transmits a logic ""1" in the priority signal position. As each transceiving unit is polled in turn, it decodes its respective address and it is then enabled to transmit its response message.

Although sequential polling signals aresimu- Itaneouslysupplied in parallel (see Figure 1 )to all of thetransceiving units, onlythetransceiving unit which decoded the currently transmitted address is normally enabled to return its response message to the control unit (such a transceiving unit is hereinafter referred to as the ""last-polled" transceiving unit"). The control unittherebymonitorsthecondi- tion of each ofthetransceiving units as the polling sequence proceeds.

If any one ofthetransceiving untis is in a priority condition (hereinafter termed a " " priority transceiv- ing unit"), it transmits a single current pulse 21 inthe priority signal position followed by a sequence of logic ""0"s. A prioritytransceiving unit does not need to decode its address in order to interruptthe polling sequence, because itcan insertacurrent pulse in the priority signal position beforethetwenty bit sequence is transmitted by the last polled transceiving unit.As the priority transceiving unit transmits logic ""0"s after the priority signal, the twenty bit sequence ofthe last polled transceiving unitis un- affected by the superimposed response messages of the last polled transceiving unit and the priority transceiving unit.

The control unit is capable of locating the priority transceiving unitwhilst the priority signal is still being applied. For example, asthe polling sequence continues, it will eventually reach the priority transceiving unit. The priority transceiving unit does not need to decode its address since it will insert a sequence of logic ""0" s instead ofthe normal twenty bit sequence. When the control unit reads the first seven bits ofthe twenty bit sequence, it will detect the sequence of logic ""0" s instead ofthe address ofthe transceiving unit. The control unit thereby re cognizesthattransceiving unit being polled is in a priority condition and hence it will detect its location.

If more than one transceiving unit is in a priority condition, more than one current pulse will be simultaneously present in the priority signal position.

However, when the priority signal is detected by the control unit, it will instantly knowthatthere is a priority condition in the alarm system and it can thereby cause an initial alarm to be given without waiting forthe polling sequence to be completed. As the control unit follows the normal polling sequence, it will note which transceiving units send back a sequence of "O"sinstead ofthe respective addresses and the control unit will thereby sequentially determine the locations of the priority transceiving units.

In orderto reduce thetimetaken to determinethe locations of several priority transceiving units, the control unit can be conditioned to poll onlythose transceiving units at which a priority condition is likelyto be present. Hence, instead of polling, e.g.

100transceiving units in a normal sequence of operation, the control unitwill poll only 10transceiving units in a zone where a priority condition is likely to exist. Such zones may correspond with respective floors, orsections of a building which is protected by a fire alarm system which embodies this invention.

The above technique may be modified in orderac- celerate detecting the location of priority transceiving units.

According to one modification, the control unit transmits a dummy address after it has detected the priority signal. The priority transceiving unitthen re sponds to the dummy address (i.e. in the case where only onetransceiving unit is in a priority condition at any one time) by transmitting its address instead of a sequence of ""0"s. The normal polling sequence is interrupted whilst the control unit sends the dummy address and receives the address ofthe priority transceiving unit. The control unit can thereby read the address of the priority transceiving unit directly and determines its location more rapidly.Atransceiving unitwould be the only one in a priority condition if it were eitherthefirstto respond to a priority condition (e.g. on detecting the outbreak of a fire), or if itwere the onlytransceiving unit enabled to transmit a priority signal (e.g. due to adopting the time-outtechniquedescribed below). Only the priority transceiving unit would respond to the dummy address because the other (non-priority) transceiving units would still respond only to their respective addresses. Thus, it would not be necessaryto complete the polling sequence in orderto lo cat a priority transceiving unit.

According to another modification, instead of transmitting a single current pulse as a priority signal, a prioritytransceiving unit transmits a sequence of pulses, representing its address, before the time slot normally occupied bythe twenty bit sequence. In otherwords, the first eight synchronising pulses 19 plus the address bits are transmitted within a time slot which is reserved for the transmission of a priority signal. The control unit therefore immediately recognises the address of a priority transceiving unit. As before, the twenty bit sequence, is unaffected by the priority signal since it occurs in a separate time slot.Alternatively, a single current pulse represents a priority signal as before and seven additional bits are added to the twenty one bitsequence to give the address of a priority transceiving unit. In this case, afterthe control unit detects a priority signal, it is conditioned to receive a twentyeight bit sequence containing the address ofthe prioritytransceiving unit.

According to further modification, instead oftransmitting a priority signal in the form of a single current pulse 21, a prioritytransceiverwould be enabled to transmit its address to the control unit during the phasenormally reservedforsignalling from the control unit to a transceiving unit. For example, current pulses designating the address of a prioritytransceiving unit would be transmitted in the spaces between the voltage pulses in the initial ten bits trans mitted bythecontrol unit (see Figure3).

As the magnitude ofthe current pulse in the priority signal position will change according to the numberoftransceiving units in a priority condition, the control unit can be conditioned to determined howmanytransceivingunitsareina prioritycondition (e.g. 20 mA represents one priority unit, 40 mA represents two priority units, 60 mA represents three priority units, etc).

At least in the preferred embodiments of the invention, a time-outtechnique is employed to restore communication with a prioritytransceiving unit after the control unit has detected a prioritysignal.Ac- cording to thetime-outtechnique, the or each priority transceiving unit is reset after a pred eterminedtime has elapsed whilstthe priority or interruptsignal is present. This predetermined time interval may be controlled bya timing circuit or by a counting circuitof known construction.For example, each transceiving unit may be provided with a counter which is indexed by one for each message trans fersequence. Such a counter could countthesingle current pulse which represents a priority condition in each message transfer sequence. When a predetermined count has been exceeded, the priority transceiving unit is reset so that it will transmit its response message 17(see Figure 3) after decoding its address in the normal polling sequence.

One advantage of restoring communication is that each transceiving unit can be remotely tested, from the control unit, by causing a priority signal to be present during a test sequence. However, in the case where a transceiving unit is in a true priority condition, a bit on an input port remains set to indicate then an interrupt has occurred at that address. This bit can be erased by the control unit in a preprogrammed test sequence and/orthis bit can be erased by resetting the transceiving unit (e.g. by resetting a switch at a manual call point).

The system may include a latching arrangement so asto capture transient priority signals. This would preventthe loss of priority signals during a time-out sequence.

Thetime-outtechnique can also be helpful when locating each of two or more prioritytransceiving units. For example, as a fire spreads, the transceiving units in thevicinitywill go sequentially into priority conditions. These priority units are then sequentially timed-outto enablethe control unitto determine their respective locations.

Advantages of at least the preferred embodiments ofthe invention are as follows: (a) A priority or interruptsignal can betransmitted to the control unit in typically less than 100 ms.

(b) The priority or interrupt signal does not cause any corruption or deterioration in the normala transmission of data from each ofthetransceiving units pertaining to their status, or normal monitoring of a particular parameter.

(c) The location of any prioritytransceiving unit can be determined whilstthe priority signal is still present.

(d) Where the time-out technique is employed, the design of the transceiving units can be such that, in a priority condition, a failure in time-out operation will only prevent communication being restored with a priority receiving unit without affecting the transmission of normal response messages from othertransceiving units in the system.

(e) By using a suitable output port on each transceiving unit, a remote test (initiated by the control unit) can be carried outto ensure that each transceiving unit is capable of providing a priority or interrupt signal.

Figure 3 is a block diagram showing the construction of a logic unit used in a transceiving unit in accordance with the preferred embodiment of the invention. Details of conventional sensors and general circuitry of known construction will not be given since these will be apparentto those skilled in the art.

As shown in Figure 3, a pulse width decoder 30 supplies clock signal output30c. It also supplies a decoded signal to an input shift register31 which stores the first ten bits in the message transfer sequence (Figure 3). An address comparison unit32 compares the seven address bits31bstored in register31 with seven address bits entered into an address input32a of unit31. The latterseven bits can be set by switches 9 (Figure 2) on the transceiving unit, or otherwise hard-wired into the circuitry ofthe transceiving unit.

Ifthe addresses match, the address comparison unit 32 supplies a signal to a transmit response enable unit 34. The signal is transmitted underthe control of an address comparison enable unit 35 which is actuated by the output of an input data counter 36 having ""reset" and ""shiftregisterclockpulse" inputs 36a and 36b respectively. The ""reset" signal is receivedfrom an initialisation and resetdetector37 having a " "power-up reset" signal input37a and an input 37b for the demodulated line signal applied to input30a of decoder 30.The initialisation and reset detector37 monitorsthedemodulated linesignalfor a combination that indicates the start or finish of a messagetransfer sequence (i.e. pulses 16 and 22).

The reset detector 37 prepares the logic circuits for the reception of a message sequence. The transmit response enable unit 34 is actuated by the clockinput so asto cause an output response synchronisation unit38to supply shift register clock pulses to an output shift register 39. Shift register 39 is loaded with data from output shift register load control 33.

This data includes the seven address bits, three bits indicating the type of sensor connected to thetransceiver, three bits indicating the status of the input port and seven bits representing the digitised value of the signal at the transceiver analogue input. This ditigised value is derived from an analogue-to-digital converter controller40 that provides a count signal to an analogue-to-digital converter counter41 connected to the appropriate shift register input of unit 33. Output data representing the address etc of the polled transceiving unit is thereby clocked out of register 39 into an output current modulator 42 which causes line current to be switched to provide output current pulses to the linear circuitry 6 (of Figure 2).

The circuit is controlled by the clock pulses so that the current pulses (or voltage drops) are interleaved with the voltage synchronisation pulses as illustrated in Figure 4.

An interrupt request detector 43 responds to input bits on ""sensor type" inputs 43a and ""sensorcon dition" input43bso asto providean interruptrequest signal on its output. The " "type" and ""condition" inputs are suitableforthe particular form of sensor in the transceiving unit. The control unit is programmed to recognised these signals when associated with a particular location and the control unitthereby confirms that a correct unit has been installed at that location. This signal causes the firstbitto be set to logic " " 1 " in the interrupt signal position and the remaining bits (in the twenty-one sequence) to be set to ""0" in the register 33.The output signal also causes the address comparison unit 32 to respond as if it had decoded the correct address in the message transfer sequence, whereby unit 34 triggers unit 38 to cause the output of register 39 to be clocked into the otuput current modulator 42. The input bits supplied to the interrupt request detector 43 are derived from circuitry in the sensor4 which responds to the parameter being monitored by the transceiving unit and which provides " "type" outputs.

The correct combination of input bits to the interrupt request detector 43 causes the interrupt request signal to be generated. The transition on the request signal initiates a time-out unit44to starta time-outas described above. Completion of the time-out period inhibits the generation ofthe interrupt request signal untilthetime-outhas been reset by a change in the state of i n put 43e to detector 43, or a change in the state of one of the " "type" inputs 43a. Input43cis one of the three inputs 43b supplied to input port 45.

The ""type" inputs 43a are supplied to detector 43c and to the load control 33 as shown in Figure 2.

An output port control 46 switches on three outputs 46a in response to the last three bits in the ten bit sequence transmitted by the control unit (see Figure 4) and stored in the input shift register31. These outputs can be turned on or offseparately.

The signals at the input port 45 are loaded into the output shift register 39 via the output control 35.

Input43c is used to initiate the interrupt request and time-out but would remain set after a time-out as an indication that an interrupt has occurred.

The output port 46b may be used to initiate a remote test, in which case the input port 45 enables the successful application of the test signal to be ded uced by the control unit.

It will be understood that the above description relates generally to a preferred embodimentofthe invention and that modifications and/or changes are possible, as will be apparent to those skilled in the art, without departing from the scope ofthe invention as defined by the follwoing claims.

Claims (13)

1. An alarm system in which a plurality oftranceiving units are each provided for monitoring at least one parameter, and in which a control unit (2) trans- mitsasequenceofpolling signalstothetransceiving units (1) whereby the state ofthe parameter monitored by each of the transceiving units (1) is detect mined, characterised inthatwhen any one ofthe transceiving units (e.g. 1 c) detects a change in the monitored parameter which represents a priority condition (the ""priority transceiving unit"), the prioritytransceiving unit (1 c) communicates a priority signal to the control unit (2) during the polling sequence, the priority signal being included in the response (17) ofthetransceiving unit (e.g. la) which received the last polling signal (the ""lastpolled transceiving unit") in such a way as to avoid interfering with any other information contained in its response (17).

2. Asystem according to claim 1, characterised in thatthe priority signal occurs in a given position in the response (17) of each transceiving unit (1).

3. Asystem according to claim 2wherein there- sponse contains bits of data characterised in thata ""0" is normally present in said given position and a priority signal is a ""1" in said given position.

4. A system according the any one ofthe preceding claims, characterised in thatthe priority signal includes at least one current pulse.

5. A system according to any one ofthe preceding claims wherein each polling signal includes address bits and synchronizing pulses in the form of voltage pulses, characterised in that the response of each transceiving unit (1) includes current pulses which occur in respective intervals between the synchronizing voltage pulses, the presence of at least one of the current pulses in a given position representing the priority signal.

6. Asystem according to claim 5, characterised in that the current pulse representing the priority signal of a prioritytransceiving unit (e.g. 1 c) precedes information pertaining to the last-polled transceiving unit (e.g. 1a).

7. Asystem according to anyone ofthe preceding claims wherein each polling signal includes address data, characterised in thatthe polling signal further includes data for effecting a remote test by the control unit (2), each transceiving unit (1) including an output port (46) responsive to said test data for communicating with the control unit (2).

8. A system according to any one of the preceding claims, characterised in that a prioritytransceiving unit (e.g. c) generates a priority signal for a predetermined period and/or during a period within which the prioritytransceiving unit is polled a given plurality of times, said period being long enough to enable detection of said priority signal bythecontrol unit (2).

9. A system according to any one ofthe preceding claims, characterised in thatthe control unit (2) transmits a dummy address after detecting a priority signal, the prioritytransceiving unit (e.g. 1c) being conditioned to respond to the dummy address in order to return its true address to the control unit (2), the polling sequence being interrupted to enablethe transmission ofthe dummy address and reception of the true address of the priority transceiving unit (1 c).

10. A system according to any one of claims 1-8, characterised in that the address of a priority trans ceiving unit(1c) is included in, or precedes, there- sponse ofthe last-polled transceiving unit (1 a) in such a way as to avoid interference with the information contained in the latter response.

11. Asystem accordingto anyoneofclaims 1-8, chararacterised in that the address of a priority transceiving unit (1c) is transmitted during a period in which the next polled transceiving unit is addressed by the control unit (2) without interfering with the address data.

12. A system according to any one ofthe preceding claims, characterised in that the magnitude ofthe priority signal is related to the number of priority transceiving units whereby the control unit (2) can determine how many transceiving units are in a priority condition.

13. An alarm system substantially as herein describbed with reference to the accompanying drawings.