EP1062845A1

EP1062845A1 - Apparatus for and method of transmitting and receiving data over a low voltage power distribution system

Info

Publication number: EP1062845A1
Application number: EP99910486A
Authority: EP
Inventors: Timothy John Thorn
Original assignee: AND Software Ltd
Current assignee: AND Software Ltd
Priority date: 1998-03-13
Filing date: 1999-03-15
Publication date: 2000-12-27
Also published as: GB2335334B; WO1999048340A1; GB2335334A; GB9805431D0

Abstract

An apparatus for and a method of transmitting and receiving data over a low voltage power distribution system are described. During signalling periods, the power supply and the loads are disconnected from the power line to allow data to be transmitted. In certain embodiments, the track units may store programs enabling operation without a central controller. The system is particularly suited to low voltage halogen lighting systems.

Description

- 1 -

APPARATUS FOR AND METHOD OF TRANSMITTING AND RECEIVING DATA OVER A LOW VOLTAGE POWER DISTRIBUTION SYSTEM

The invention relates to a method of and apparatus for transmitting and receiving data over a low voltage power track. The invention is particularly suitable for use with Halogen-type low voltage lighting systems.

Low voltage power distribution systems are known in which a single power supply feeds a power track for distributing power to one or more electrical loads on the track. A typical example of such a low voltage power distribution system is a lighting system, in which a single power supply is used to supply power at about 12V AC via the power track to a plurality of spatially separated lighting units containing Halogen-type lamps.

In may cases it would be desirable to control the power supply to the electrical loads individually; for instance, in the case of the lighting system, it may be desirable to adjust the brightness of each lamp individually. In known low voltage power distribution systems such individual control is not possible since a common power track is used to supply power to all of the electrical loads.

It is known from US 5072216 to provide a transmitter for transmitting data over a mains power line via a track to lighting units. The precise method of signalling employed is not disclosed, but any system must rely on mains-borne signals, and mains power may interfere with the data transmission.

EP-A-0038877 discloses a system for transmitting one of a selection of frequencies over a mains power distribution line for the purpose of controlling devices such as lights, in which the mains alternating current is interrupted for a certain section of each cycle in order to allow a frequency signal to be transmitted.

Both of the above disclosures are concerned with mains-powered systems. In such systems, signalling is less problematic, as the devices connected to the track have a relatively high impedance, and a low voltage signal can be transmitted without difficulty. For example, a 240V 60W bulb has a nominal impedance of about 1000Ω. In contrast, in a 12V system with four 60W bulbs, the nominal impedance of the line is only about 0.6 Ω which would place considerable demands on any data transmitter and would result - 2 - in significant attenuation of any signal transmitted. In addition, with a low voltage line, any signalling voltage is significant in comparison to the track voltage and so may affect devices connected to the track.

US-A-4300126 discloses a system for communicating over high voltage power lines and in particular for reducing attenuation of signalling frequencies of the order of 10kHz in a 13.2kN to 120N step-down transformer. US-A-4408185 discloses a mains signalling system in which information is transmitted at high frequencies.

It is an object of the present invention to provide an apparatus for transmitting data over a low voltage power track. By low voltage is meant a voltage which can be safely touched by a human, for example, less than about 80 volts and preferably around 12 or 24 volts .RMS. The invention further aims to provide a system for controlling lights in a low voltage track system.

According to a first aspect of the present invention there is provided a control unit for use with a low voltage power distribution system which comprises an alternating current power supply for supplying low voltage power and a power track for supplying power from the power supply to at least one electrical load, the control unit comprising control means for determining a signalling time in the voltage waveform for transmitting data, switch means for disconnecting the power supply from the power track at said signalling time, and transmitting means for transmitting a plurality of data bits over the power track during a signalling period while the power supply is disconnected. The power distribution system may be, for example, a low voltage track lighting system. This arrangement can allow data to be transmitted without requiring a separate data line and with minimal interference from the power supply.

The control unit may be adapted to transmit at least four bits, and preferably at least eight bits, in any one signalling period. The control unit may be adapted to transmit messages which span at least two signalling periods. This provides the advantage of allowing complex messages to be transmitted. For example, three bytes each of eight bits, plus start and stop bits, may be transmitted in each signalling period. A message may be, for example, six or more bytes long and therefore span two or more signalling periods.

The control means may be adapted to determine a signalling time which is close to, but - 3 - offset from, a zero crossing instant of the voltage waveform. For instance, the signalling time may be offset from a zero crossing instant by less than 4 milliseconds or by between 10 microseconds and 3 milliseconds or by at least about 0J millisecond and at most about 2 millisecond or by about 1 millisecond. The signalling period may begin on one side and end on the other side of a zero crossing instant of the voltage waveform. In this way data transmission can take place with minimum disruption of power transfer from the power supply to the electrical loads. This feature may be provided independently.

The switch means may comprise a pair of transistors with opposite polarities. For example, in the case of bipolar transistors, the transistors may be connected in parallel with commonly connected collectors or commonly connected emitters and protective diodes such that each transistor conducts in a half cycle of the voltage waveform when the switch is on. In the case of field effect transistors, the transistors may be connected in series with commonly connected drains or commonly connected sources. This allows the power supply to be effectively connected or disconnected from the power track under electronic control at all points in the voltage waveform. The transistors may advantageously be MOSFETs connected in series. Gate drive means may be provided for supplying to the gates of the transistors a voltage sufficient to turn off at least one of the transistors during the signalling period, and a voltage which follows the supply voltage with a substantially fixed offset, such that both transistors remain on, at other times in the voltage waveform. In this way the drive voltage will always be high enough to keep the transistors in the on condition when no data is to be transmitted, without exceeding the safe operating voltages of the transistors. The gates of the MOSFETs may be commonly connected, in which case a single gate voltage may be derived, or alternatively the gates of the MOSFETs may be driven independently.

The control unit may be for use with track units which are arranged so that, during the signalling period, the track has a nominal impedance within a predetermined range. By nominal impedance is meant the impedance to which the track reverts during the signalling periods when no track units are signalling. The control unit may comprise an impedance element such as a load resistor which substantially determines the nominal impedance. The transmitting means may be adapted to transmit data by connecting and disconnecting a current source to the track. The track units may signal by switching a load to and from the track, and the control unit may further comprise a reception circuit for detecting data voltage levels on the track. Since the track presents a known impedance - 4 - during the signalling periods, the data voltage levels will not alter significantly, resulting in more reliable data transfer.

According to a second aspect of the present invention there is provided a power supply unit for a low voltage power distribution system, comprising a low voltage transformer and the above mentioned control unit.

According to a third aspect of the invention there is provided a track unit for use with a low voltage power distribution system which comprises an alternating current power supply and a power track for supplying power from the power supply to at least one electrical load, the track unit comprising control means for determining a signalling period in the voltage waveform, the signalling period being a period during which the power supply has been disconnected from the track for data transmission, and transmitting means and/or receiving means for transmitting and/or receiving data over the power track during said signalling period. The power distribution system may be for example a low voltage track lighting system.

The track unit may be for connection between the power track and an electrical load, in which case it may further comprise switch means for disconnecting the electrical load from the power track during said signalling period. The switch means may be adapted to connect and disconnect the electrical load during the signalling period to transmit data. In this way the track unit may transmit data without requiring a separate transmitter circuit. The switch means may comprise a pair of transistors connected in series with opposite polarities. The transistors may advantageously be field effect transistors, preferably MOSFETs, with commonly connected gates. Gate drive means may then be provided for supplying to the gates of the transistors a voltage sufficient to turn off at least one of the transistors during the signalling period, and a voltage which follows the power track voltage with a substantially fixed offset, such that both transistors remain on, at other times in the voltage waveform.

Alternatively, the track unit may be for connection to a sensor for sensing an ambient condition and may further comprise transmitting means for transmitting data from the sensor over the track, for instance in response to a request for data. The sensor may be selected from one of a heat sensor, a movement detector, an infrared detector, a sound detector, a light detector, a temperature detector or a voice recognition device, although other types of sensors could be used.

The track unit advantageously includes a memory for storing parameters for a program. For example, if the track units are required to dim sequentially or to flash, parameters such as program type, rate of dimming, maximum and minimum brightness etc. or rate of flashing and mark-space ratio may be stored. In this way, the track unit may be programmed and the commands may then be executed without requiring a continuous stream of instructions from the controller. This may significantly reduce the amount of information to be transferred in a system with several track units and complex lighting sequences. The program is preferably stored in a non- volatile memory, preferably integrated in a micro controller which controls the lights. Another advantage is that, once the track units are programmed, the controller is not required and the track units may be operated on a "dumb" track. Preferably, where the track units are arranged to execute a stored program, timing is based on counting of track power cycles. In this way, synchronisation may be maintained over a long period of time between separate devices on the same track.

In an independent further aspect, the invention provides a track unit for use with a low voltage power distribution system comprising an alternating current power supply and a power track for supplying power from the power supply to at least one electrical load, preferably a light, the track unit comprising means for storing a program and means for controlling the electrical load based on the stored program.

The track unit is preferably programmable and preferably includes non-volatile memory means, for example flash memory, for storing a program, or program parameters, so that the unit can be programmed after manufacture. The unit may be programmed by means of a direct connection or by data transfer to a component when the unit is only partially assembled. More preferably, the unit may be arranged to receive a program over the track, using a signalling protocol according to other aspects of the invention. The track unit is preferably arranged to disconnect the electrical load from the track so that the track unit can be installed on a track in accordance with other aspects of the invention without causing intereference. The track unit may include control means for determining a signalling period in the voltage waveform, the signalling period corresponding to a period during which the power supply is temporarily disconnected from the track for data transmission; and at least one of means for disconnecting the electrical load from the track and transmitting means and/or receiving means for transmitting and/or receiving data over the power track during said signalling period.

In a development of the above aspect, the invention provides a kit comprising a plurality of track units according to the above aspect, each unit having a respective unit program stored therein, the unit programs not all being identical, so that, when the units are installed on a track, the units together execute a desired ensemble program. For example, the units may be arranged to light, flash or dim in sequence. The units may include means for signalling between units, for example to establish synchronisation, preferably using a protocol similar to any one described elsewhere in this specification. In this way, a lighting ensemble can be provided in which an overall controller is not required.

According to a fourth aspect of the invention there is provided a device for connecting and disconnecting a low voltage alternating current power supply from a load comprising a first field effect transistor connected in series with a second field effect transistor in a back to back configuration, said first and second field effect transistors being connectable between the power supply and the load, said first and second field effect transistors having commonly connected gates, and a drive circuit which is adapted to supply to the gates of the transistors a drive voltage which is sufficient to turn off at least one of the transistors when the power supply is to be disconnected from the load, and a voltage sufficient to turn on both of the transistors when the power supply is to be connected to the load, whereby said drive voltage follows the supply voltage with a substantially constant voltage offset when the power supply is to be connected to the load. In this way the drive voltage will always be high enough to keep the transistors in the on condition when the power supply is to be connected to the load without exceeding the safe operating voltages of the transistors. By back to back is meant that the transistors have commonly connected drains or commonly connected sources.

According to a fifth aspect of the present invention there is provided apparatus for transmitting data in a low voltage power distribution system which comprises .an alternating current power supply, a plurality of electrical loads, and a power track for supplying power from the power supply to the plurality of electrical loads, the apparatus comprising means for disconnecting the power supply from the power track during a portion of the voltage wavefoπn, means for disconnecting the electrical loads from the power track during a corresponding portion of the voltage waveform, and means for - 7 - transmitting data over the power track while the power and loads are disconnected.

According to a sixth aspect of the present invention there is provided a track lighting system comprising an alternating current power supply, a power track connected to the power supply, and a plurality of lamps connected to the power track, and further comprising the above described apparatus. The track lighting system may further comprise means for receiving data and means for controlling the lamps.

According to a seventh aspect of the present invention there is provided a method of transmitting data in a low voltage power distribution system which comprises an alternating current power supply, a plurality of electrical loads, and a power track for supplying power from the power supply to the plurality of electrical loads, the method comprising disconnecting the power supply from the power track during a portion of the alternating current cycle, disconnecting at least one of the electrical loads from the power track during a corresponding portion of the alternating current cycle, and transmitting data over the power track while the power and loads are disconnected.

Prefeired features of the present invention will now be described, purely by way of example, with reference to the accompanying drawings, in which:-

Figure 1 shows a low voltage power track system with which the present invention may be used;

Figure 2 shows the principles of the signalling protocol used in a preferred embodiment of the present invention; Figure 3 shows an example of the frame format of transmitted data;

Figure 4 shows a block diagram of a control unit according to a prefeired embodiment of the present invention;

Figure 5 shows a circuit diagram of a control unit according to a preferred embodiment of the present invention; Figure 6 shows a voltage waveform generated within the control unit of a preferred embodiment of the present invention;

Figure 7 shows an example of a modified voltage waveform for dimming a lamp; Figure 8 shows a block diagram of a track switching unit according to a prefeired embodiment of the present invention; Figure 9 shows a circuit diagram of a track switching unit according to a preferred - 8 - embodiment of the present invention;

Figure 10 shows a block diagram of a track interface unit according to a preferred embodiment of the present invention; and

Figure 11 shows a schematic diagram of a modified signalling sequence allowing rapid response to interrupts.

Referring to Figure 1, a low voltage power distribution system comprises a transformer 10, for converting mains voltage to a low voltage such as 12N AC, a track control unit 14, a power track 16, track adapters 18,20, track switching units 22, and track interface units 30. Low voltage high current feeds 12,13 connect the transformer 10 to the control unit 14, and the control unit 14 to the power track 16, respectively. The track adapters 18,20 connect Halogen lamps 32 and sensors 40,42 to the track 16. In this example, the sensor 40 is a passive infrared detector and sensor 42 is a light level detector. The track switching units 22 may be mounted in lamp holders (not shown), or in track adapters 18. Similarly, track interface units may be mounted in sensor housings (not shown) or in track adapters 18. The control unit 14 is further connected to input means 62, which may be, for example, a dimmer, a switch, a keypad or a remote computer. Track switching units 22 and track interface units 30 will be referred to generally as track units.

In operation, the control unit 14 modifies the power waveform fed to the track to allow data to be transmitted along the track. During a defined signalling period, the control unit disconnects the transformer from the track. At the same time, the track switching units 22 disconnect their respective lamps from the track. This allows the interchange of information and commands between the various elements of the system along the track. For instance, the track interface units 30 may each transmit information from their respective sensors 40, 42 to the control unit 14, and the control unit can use this information to decide how the individual lamps in the track should be driven. This information is then transmitted back to the appropriate track switching units 22 which then set the desired lamp brightness. In this embodiment, the track switching units vary the light levels of their associated lamps using phase angle switching techniques, although other methods, such as pulse width modulation, may be employed.

Signalling Protocol

In a preferred embodiment the signalling period occurs in a defined period around the zero cross points of the power voltage, to give minimum power loss. In one example, the - 9 - disconnections are nominally 2 milliseconds long, but this period may be altered and need not be precisely defined nor constant from one cycle to another. The principles are illustrated in Figure 2.

During disconnection, the control unit 14 provides a single defined load, for example, a resistance of 2.2 kΩ, although other values may be used. This allows data transfer to occur by current switching. If the lamps were not disconnected from the track, the track would present an impedance which varied with the number and location of track units, and hence the data signal levels would vary. By providing a predefined and relatively constant load during the signalling periods, the data signal levels will be substantially constant which allows accurate data transfer.

The data signal levels may advantageously be at the same voltage levels as the zero cross detection thresholds. In the present example, any voltage above 2.5 volts is considered as a data 1 and any voltage below as a data 0. However, other signalling protocols may be used, for example a positive voltage may be considered as data 1 and negative voltage as data 0, or any other threshold voltage could be used.

In the present example data bits are 100 microseconds wide corresponding to an instantaneous bit rate of 10 kbits per second. Preferably at least 8 bits are transmitted per half cycle, although this may be higher. For example, transmitting 40 bits per half cycle would give a sustained data rate of 4kbit/s. Shorter bit durations (for example 50 microseconds) may improve data rates, but noise or radio-frequency interference suppresion may be more problematic.

In the present example, a multi-bit self clocking scheme, such as is used in known standard asynchronous communications (for example RS 232), is used, in which each byte is preceded by a start bit and ends with a stop bit. This allows low cost clocks to be used. Other schemes, such as Manchester coding, may also be used, although there is no specific need to preserve a 50 % mark-space ratio, as the track is fully DC coupled. Some calibration of the clocks may be undertaken by using the mains frequency.

Figure 3 shows the frame format of transmitted data. The frame format is kept relatively simple, although the flexibility to send variable types and volumes of data is maintained. There is also a need to prevent false decodes, so a considerable amount of redundancy is - 10 - included. A CRC is an option, but this is not fully effective for short messages. Instead, a checksum is used with deliberate additional redundancy provided in other forms, such as long device identities or data repeated in true and inverse forms. For simplicity the basic format is a fixed frame in any disconnection period of 3 bytes.

There is always a period between the disconnection commencing and the signals being superimposed on the track. It follows that at the start of every signalling period the track voltage is zero (i.e. a data 0), and a start bit is therefore required for all units to synchronise to the bit timing. This is a data 1. Immediately following this start bit there is a 'new message start' indicating bit, and after this the data follows 8 bits at a time with start and stop bits.

Note that each disconnection period may or may not contain any data, and that individual messages can and in general will span several disconnection periods. The consecutive elements of a message are shown in Table 1.

Table 1

field/element length in value remarks bits expansion bit 1 0 allows different formats in the future short id bit 1 0/1 1 = short device id, 0 = foil device id spare 2 - message length 4 0-15 number of bytes in message device id 8 or 24 derived from physical identity command 8 0-255 what to do parameters 0-n 0-n parameters for command; length depends on command 11

check sum 8 0-255 sum of all fields from expansion bit to end of parameters

The protocol operates on a master/slave basis with the control unit always the master and the track units always the slaves. The slaves do not respond until commanded to do so by the control unit. When track units are commanded to respond they must do so in the immediately following transmission period. They will transmit data addressed to the controller, as device 0.

In certain arrangements, it may be preferable to transmit only a single byte per disconnection period as the reduced data rate may simplify switching and interference suppression.

In the aπangements described above, the controller sends commands to output devices such as lights and interrogates sensors. The controller typically operates a control loop in which commands are sent and sensors are polled in sequence. If there are a large number of devices in the system, or if a complex control sequence is employed, the delay before a particular sensor is polled may become longer than is desired. To facilitate rapid response to sensors, it is possible to modify the protocol so that a single "interrupt" bit is provided. This may be achieved by the control unit applying a positive signalling voltage to the track for one bit during which any of the sensors may pull the voltage down to zero (in the same manner as the sensors transfer data back to the control unit in response to a request to do so) to signal an interrupt. Following detection of an interrupt, the control unit may be programmed to interrupt any pending output sequence and switch immediately to a polling sequence to determine the cause of the interrupt.

More than one interrupt bit may be provided so that the sensor causing the interrupt may be directly identified. However, provision of several interrupt bits will normally reduce the efficiency of data transmission. Another alternative is for the control unit to switch immediately to a data reception mode following detection of an interrupt (but not normally otherwise) and to continue to supply the positive signalling voltage to enable the interrupting sensor(s) to identify itself. For example, in a system with four sensors, the control unit may be arranged, following detection of an interrupt, to maintain the positive signalling voltage for four bit cycles and each sensor may be arranged to pull the track - 12 - voltage down to zero during a respective assigned bit position (these bit positions may be soft-configured by the controller during initial configuration). This method has the advantage that multiple interrupts can be instantly identified. For example, if the sensor assigned to the first bit is a Passive Infra-Red (PIR) sensor, the second is a switch, the third is an ambient light detector and the fourth is an infra-red receiver, detection of the sequence "0110" following an inteπαipt bit indicates both that there has been movement detected by the PIR sensor and a command sent to the infra-red receiver. This scenario is illustrated schematically in Fig. 11. The above or a similar mechanism may provide an efficient means of responding rapidly to one or more sensors.

Whilst the above method may give a rapid response to interrupts, it requires additional logic in the track units to detect an interrupt and to supply interrupt status. Another potential problem with having the interrupt bit following the start bit is that certain sensors (for example an IR receiver) may produce frequent interrupts and this may interfere with normal operation of the controller. To prevent "hi-jacking" of the track by a device which causes frequent interrupts it is therefore preferred to position the interrupt bit after the stop bit (see Fig. 3). Another advantage of having the interrupt bit after the stop bit is that certain track devices which do not need to respond to the interrupt bit (for example output only devices) can be arranged simply to ignore data after the final stop bit of the frame of Fig. 3; this may simplify programming (although such devices must ensure that the track is still available for data transmission, by keeping any loads unconnected, for a sufficient time for interrupt signalling to complete).

Rather than having devices supply interrupt status immediately following an interrupt, it may be preferable for the devices to return a status word, in a similar manner to that depicted in Fig. 11 in response to an explicit command from the controller. This has the benefit that the controller remains in command and is able to complete any output sequence and then request interrupt status at a convenient point in the control loop.

In a system in which there are a large number of output devices such as lights, it may be desirable to switch several or all of these on or off together rapidly or to dim several or all lights together. Rather than sending individual commands to each device, commands may be broadcast to all devices or a group of devices. This may be achieved either by allocating certain device identities to groups of devices so that a command addressed to such a device identity is acted upon by several devices or by including a dedicated - 13 - "broadcast" bit in the signalling or protocol.

Control Unit

The control unit converts raw low AC voltage power feed from the main transformer to a composite signal which is fed to the track. This composite signal consists of three signal conditions, which are time-multiplexed onto the track:-

• The low voltage, 'high current capable' AC power signal from the transformer.

• A load resistance with zero voltage. • A load resistance with a positive signalling voltage.

In order to achieve this situation the following main functional elements are used:-

A power switching element. ^Φ A voltage threshold detector for the AC power signal.

A microcontroller to provide overall control and signal timing. A track interface transmission circuit A track interface reception circuit

Figure 4 shows a block diagram of the control unit 14. A power switching element 50 is operable under control of microcontroller 52 to disconnect the power from the track during signalling periods. The transmission circuit 54 and reception circuit 56 are operable during the signalling periods to transmit or receive information to or from the power track. The microcontroller 52 receives as inputs a signal from voltage threshold detector 60, and commands from an external device 62, which may be for example a keypad or an external computer.

Figure 5 shows a circuit diagram of one implementation of the control unit. Referring to Figure 5, the power switching element consists of power MOSFETs Q1.Q2. In this case a single pair of MOSFETs is used, connected in series but "back-to-back" so that current flowing from drain to source in one passes from source to drain in the other. This back- to-back connection configuration is necessary to allow the combined switching element to block voltages of either polarity. This is because of the "parasitic diode" inherent in the construction of a power MOSFET which means that even when no gate drive is present to turn the devices ON they will still conduct if subjected to a reverse polarity - 14 - voltage between drain and source. Connecting two MOSFETs in series in opposite directions means that in the OFF condition one or other will always be operating with coπect polarity regardless of the applied polarity and there will therefore always be an effective barrier to current flow.

In the ON condition both MOSFETs are driven to conduct heavily by an appropriate gate drive signal. This gives the series combination the desired low resistance. Note that even the MOSFET which is subject to reverse polarity can still be turned folly ON by the gate drive signal.

The characteristics of currently readily available MOSFETs allow currents of 2 to 3 amps or more to be switched with minimal losses. Higher currents can be handled either by using devices having increased capacity, or by simply connecting further pairs of MOSFETs in parallel with the first pair. The thermal characteristics of parallel connected MOSFETS allow them to share current safely. As more devices are connected in parallel the gate drive circuit must be capable of providing increasing current drive at switching points to charge/discharge the increased effective gate capacitance.

The gate drive circuitry consists of Q5, Q7, Q8, Q6, Q9 and their associated components. .An important aspect of the gate drive circuitry is the generation of a signal, known as

N_GATE_ON, by Cl and Dl. The principle is that this signal follows the AC power waveform which the MOSFETs must switch but is displaced from it by a relatively fixed

DC offset voltage equal approximately to the peak voltage of the AC power waveform.

This gives a high gate to source voltage for the MOSFETs when in the ON condition, to keep RDS_ON (drain to source resistance in the ON state) to a minimum, but keeps the gate to source voltages always within the safe operating region in both ON and OFF conditions and at all parts of the AC power waveform.

The signal V_GATE_ON is shown in Figure 6.

The behaviour of the circuit may be analyzed by considering in turn the two cases ON and OFF for both Ql and Q2.

ON CASE - 15 -

Here the signal LINE CONTROL is low, turning off Q5 and in turn Q7 and Q8. In this case both bases of transistors Q6 and Q9 are pulled close to V_GATE_ON by R7. It follows that the voltages at the gates of Q2 and Ql will also both be close to V_GATE_ON.

In this ON case both MOSFETs are conducting so the midpoint will follow closely the voltage at the "A" leg of the power feed (the leg to which the MOSFETs are connected). Importantly N_GATE_ON also follows this but displaced by a DC offset. The effective gate to source voltage is therefore this DC offset, and doesn't vary significantly over the half cycle. This in turn means that it can be high without ever exceeding the allowable maximum.

To highlight the value of this scheme it could be contrasted with an alternative case where the gate voltage was held at a steady voltage level whilst the sources varied in voltage. In order to hold the devices ON, this steady level would have to be high enough to exceed significantly the maximum positive voltage at the A leg power feed. For a nominal 12 volt supply the A leg of the power feed will vary between plus and minus approximately 17 volts, so to give say 5 volts of excess gate drive on the positive half cycle, the gate voltage would need to be 22 volts. This would imply an effective gate to source voltage of 17 plus 22 or 39 volts, very considerably in excess of the maximum safe gate to source voltage allowed for typical devices of this type (approximately 20 volts).

OFF CASE

Here the signal LINE CONTROL is high, turning on Q5 and in turn Q7 and Q8. In this case both bases of transistors Q6 and Q9 are pulled close to ground or the A leg of the power AC waveform, whichever is the more negative, (through D9 or D10). It follows that the voltages at the gates of Q2 and Ql will also both be close to ground or the A leg of the power AC waveform whichever is the more negative.

When the A leg of the power feed is positive, the gate drive signal will be close to ground because D9 will be reverse biased. Q2 will be folly off but the parasitic diode in Ql will mean that the mid point junction will be close to the A leg of the track, which will be close to ground (because there is no supply voltage to the track). It follows that the sources of Ql and Q2 are close to ground. As the gate voltage in this case is also close - 16 - to ground the gate to source voltage will be low. Ql and Q2 will therefore be off and will be operating within the safe area of gate to source voltage.

When the A leg is negative, the gate drive voltage will be close to the A leg of the power feed because D9 will be forward biased. The parasitic diode in Q2 will be active so that both sources will also be close to the A leg of the power feed, again giving a low effective safe gate to source voltage for both Ql and Q2.

The zero cross voltage threshold detector circuit 60 senses the AC power waveform and derives from it a logic signal which has a low going pulse around the zero cross point. The microcontroller 58 uses this to derive the timing of the exact zero cross point.

The voltage threshold detector is centred around Q3 and Q4. When the AC power voltage is low, Q3 and Q4 are off and the *ZERO_CROSS output is therefore pulled high by Rl. As the power feed voltage rises at the start of a positive half cycle, current is fed to the base of Q3 via R2 and R16 and this turns Q3 on, causing the *ZERO_CROSS output to go low. On the negative half cycles the current through R2 and R16 passes through the emitter-base junction of Q4 which is operating in common-base configuration and therefore pulls *ZERO_CROSS low.

In summary Q3 conducts on positive half cycles, Q4 conducts on negative half cycles and neither conducts for a small period close to the zero cross point. The *ZERO_CROSS signal consequently is high around the zero cross point and low at all other times.

C4 smooths out switching and other transients to prevent these from disturbing the switching points.

D3 and D5 add two diode drops to the output voltage in the negative half cycles to ensure that *ZERO_CROSS does not drop below zero (which would be outside the allowable minimum input voltage for the microcontroller).

The microcontroller controls track disconnection activity, and possibly other system functions as well. In one example, the microcontroller generates the necessary control signals for the track disconnection activity, and handles data interchanges on the track. In this example the microcontroller also acts as the host for the lighting control application - 17 - and therefore runs the logic which will decide what level of illumination to set for each lamp.

In another example, microcontroller generates the necessary control signals for the track disconnection activity and handles the data interchanges on the track, but does not act as the host for the lighting control application itself. In this alternative the microcontroller therefore only does signalling and switching, and is linked via a suitable bus (e.g. I2C or RS232) to another, main application microprocessor. This approach (especially using RS232) has the following advantages:-

• Accommodates, readily, splitting user interface i.e. control box from high current wiring.

• Natural support for PC front end.

In a third example, a number of fonctions such as time of day maintenance and non- volatile storage are added to the dedicated microcontroller of the second example, so that it can re-install settings and get going on power up, with the main application microcontroller only being required to alter program settings.

The requirements of the microprocessor may be summarised as follows:-

• Relatively fast CPU for rapid response to data etc.

• EEPROM (external or on-board) or battery backed RAM option

• UART if possible. • Secure one time programmable mechanism available (i.e. not a ROMless based solution)

• Flash programmable (for in-situ or on-site uploads).

• Analogue to digital converters might be useful for parametric measurement (e.g. light level, temperature etc.) local to the controller box.

• Time of day clock running from battery-backed power supply unit etc.

Around the zero cross point, Ql and Q2 disconnect the power feed from the track conductors. In the track units, similar switching arrangements also disconnect all the loads. With power and loads disconnected in this way the impedance of the track is largely determined by resistor Rl 1. To allow data transmission a constant biasing current is generated through Rl 1, causing a voltage to be developed between the track conductors. - 18 -

Transmission from control unit to track units is achieved by switching this bias current on and off. Track units can detect the presence or absence of the bias current by simply monitoring the track voltage.

When the controller is expecting communications from a track unit it activates the track bias current through Rl 1 and monitors the resultant track voltage. The track units signal back to the controller by shunting the bias current by connecting a suitable low impedance across the track. This causes the track voltage to drop to a low value which the controller can then detect.

The transmission circuit consists of Q12 and Q13 and their associated components. Its purpose is to superimpose a switchable bias current through Rl l to introduce a track signalling voltage during power and load disconnections. This is controlled by the signal DATA_ENABLE. When high this turns Q12 on, driving current through ZD1 which applies the corresponding Zener voltage (nominally 5.6V) to the base of Q13. Q13 then operates as an emitter follower and effectively applies a constant voltage of around 5 volts across R12. This configuration causes Q13 to operate as a constant current source, passing cuπent out on to the combination of Rl 1 in parallel with the track. D4 protects Q13 from the reverse collector voltages which otherwise would arise during the periods for which the foil AC power wave is applied to the track.

When DATA_ENABLE is low Q12 is off, and this keeps Q13 off, so no current is generated and Rl 1 is simply connected across the track with no biasing voltage.

In summary, when the power feed and loads are all disconnected, DATA ENABLE switches the transmission bias current through Rl 1, which is always connected across the track.

The reception circuit is based around Q10. Q10 detects the presence or absence of significant track voltage. A track voltage will drive base current to Q 10 through R9. This will turn on Q10 and bring the RXD signal low. When the track voltage is too low to activate Q10, R-XD will be pulled high by R8. D7 shunts negative track voltages to prevent these from damaging the base emitter junction of Q10.

The RDX output could also be used to perform a diagnostic fonction on the track: R.XD - 19 - should be the inverse of the DATA_ENABLE input; if RXD remains high, this suggests that the track voltage is being shunted, for example by a faulty track unit.

Track Switching Unit Figure 8 shows a block diagram of a track switching unit. A track voltage detector 70 is connected to the power track 16. This circuit is used both to monitor the track voltage and to receive data. The output of the track voltage detector 70 is fed to microcontroller 72. The microcontroller generates the necessary control signals for disconnecting the load and for transmitting data onto the track. An optional transmitter circuit 74 may be provided. However, in this example, data transmission is achieved by switching the load on and off during transmission periods. This avoids the need for a separate transmitter circuit.

A drive circuit 76 generates a signal for control of a power switch 78, which disconnects the load 32 from the track 16 during data transmission periods (except when the track switching unit is itself transmitting data). Disconnecting the electrical loads from the track allows the track to present a known impedance during the data transmission periods. This helps to minimise noise and increase the reliability of data transmission.

A power supply (not shown) is provided to supply power to the microprocessor. The power supply may derive power from the track, or elsewhere. Preferably, the power is derived by rectifying the track voltage.

A detailed circuit diagram of one embodiment of the track switching unit is shown in Figure 9.

The microcontroller generates the necessary control signals for the load disconnection activity, and handles the data interchanges on the track. The microcontroller also provides some low level intelligence, such as generating power up and power down ramp fonctions for its local switching fonction, so as to relieve the control unit of the task of setting all the individual light levels in a typical dim-up or dim-down ramp; the control unit need then only send an "up" or "down" command.

In one example, an 8 pin PIC device is used (the PC12C5.XX). This can run from an internal clock, includes a built in watchdog, and has a suitable internal timer, along with - 20 - 6 general purpose .I/O lines.

The microcontroller provides the necessary conversion from logical (i.e. subjective) light levels to phase angle, so that, for example ramp-up sequences appear appropriately smooth and linear. The power supply for the microcontroller may consist of a bridge rectifier connected across the power track, a reservoir capacitor and a low power regulator scheme.

Back to back pairs of MOSFETs are used for the power switch in the same way as in the control unit. The gate drive circuitry is similar in principle to that used in the control unit but differs slightly since the power supplies in the track switching units are based on bridge rectifiers rather than on half wave rectifiers as in the control unit.

Dl and Cl generate a voltage which is equivalent to a DC offset added to the AC power waveform on the A leg of the track feed. This is similar to the N_GATE_ON waveform in the control unit. Again, the ON case and OFF case may be considered for both half cycles.

OFF CASE

The OFF case is selected by LAMP/TXD signal being high (that is, approximately 5 volts), so that Q3 is on, pulling its drain low. This means that the bases of Q4 and Q5 will both be close to ground and consequently the gate voltage of Ql and Q2 will be close to ground.

When the A leg is positive, the parasitic diode in Ql will conduct so that the mid point (that is, the junction of the sources of Ql and Q2) will be close to the B leg. Note that as a consequence of the action of the bridge rectifier, in the positive half cycle the B leg will be close in voltage to ground. It therefore follows that the effective gate to source voltages for both Ql and Q2 will be low and well within the safe operating region.

When the A leg is negative, the bridge rectifier holds the A leg close to ground and the B leg is be positive. In this case the lamp / load feed, that is, the A leg of the LAMP connector, will be positive because there is no effective voltage across the lamp. This implies that the parasitic diode in Ql will be reverse biased and that in Q2 will hold the mid point (that is, the junction of the sources) close to the A leg which is effectively close - 21 - to ground. In this half cycle therefore, the effective gate source voltages for both Ql and Q2 will again be low and well within the safe operating region.

ON CASE

In this situation LAMP/TXD will be low so that Q3 is off and the voltage at the bases of Q4 and Q5 (and therefore the effective voltage at the gates of Ql and Q2) will be determined by the voltage at the junction of Rl and R2. Cl, Dl, Rl and R2 will combine so that this effective voltage will be a fixed DC offset superimposed on the A leg track power waveform. As the sources of Ql and Q2 will be close to this A leg track power signal, the effective gate source voltage in the on case will be this DC offset. This offset will approximate to half the peak voltage of the power waveform, or around 8.5 volts. This will be sufficient to turn Ql and Q2 folly on but will not exceed the safe operating gate source drive voltage.

An advantage of using MOSFETs (or other transistors) for switching the load is that conventional phase angle switching need not be relied on to dim the lamps. Conventional phase angle switching relies on firing a triac at a point in the waveform, the triac remaining on until the next zero crossing period. The power waveform is therefore modified by removing the first part of a half cycle. This may lead to noticeable flicker, particularly at low light levels, and to the emission of radio frequency interference. Unlike triacs, MOSFETs can be switched off again and therefore multiple switching cycles may be employed in each half cycle, leading to reduced radio frequency interference and flickers. Figure 7 shows an example of how the power waveform may be modified by the MOSFETs to dim the lamps. In this example the power is switched on and off twice in each half cycle, but the switching could be done any appropriate number of times in a half cycle.

The track voltage detector circuit is used for power region and data sensing. A 2 millisecond wide disconnection window results in a voltage step at the start of the power region of around 5 volts. The signal transmission voltage is also around this level.

The voltage detector circuit is formed by R4, R5 and ZD1. Full wave rectified track voltage is fed to this combination. When the track voltage is low R14 and R15 effectively - 22 - pull the ZERO_CROSS / RXD signal low. When the track voltage rises, the resultant voltage at the output of the bridge rectifier passes through R15, pulling ZERO_CROSS / RXD high. ZD1 clamps the voltage so that it remains within safe operating limits for the microcontroller input to which ZERO_CROSS / .RXD is connected.

In order to signal back to the control unit the track switching unit shunts the bias current fed to the track via a low impedance, reducing the track voltage to a low level in the process. An advantage of this signalling scheme is that no additional circuitry is required to achieve this in the track switching unit as the lamp/load power switch may be used to do this. Activating Ql and Q2 will connect the lamp across the track, shunting the track bias current and causing the track voltage to drop close to zero. Note that activating the lampΛoad switch during the transmission period does not significantly increase the energy transfer to the load because during the data transmission period it is only the biasing current which energises the track. This is in contrast to activating the switch during the power feed region where large currents can be sourced by the track power feed.

This aspect of the switching scheme helps significantly to reduce the complexity, and consequently the size and cost of the track switching units. It also minimises the power supply requirements in the units which again helps reduce size and cost.

Although the circuit schematics for both the control unit and the track switching units show the tracks as consisting of an A leg and a B leg, in practice there is no guarantee at all that the A leg leaving the control unit will be connected to the A leg of each track switching unit - wiring twists may mean that the control unit's A leg connects to the track switching unit's B leg and vice versa.

The signalling scheme copes readily with polarity inversions between the control unit and track switching units, as the signalling relies only on shunting a bias current using the existing power switch element. This element must in any event be capable of switching the normal AC power feed and can therefore also switch bias currents in either direction.

Track Interface Element

The track interface elements interchange data with the power/signal track and also typically measure or sense some parameter such as ambient light level, Passive Infra Red (PIR) movement detection, sound level and so on. Like the track switching units, these - 23 - units are physically small to allow discrete mounting.

The design of these units is similar to that of a track switching unit with the following exceptions:-

• They include circuitry to allow their microcontroller to sense the appropriate parameter.

• They use a similar MOSFET based switch as a backward signalling mechanism, but a defined low impedance load is used instead of the lamp and the MOSFETs therefore do not need to switch so much current.

Figure 10 shows a block diagram of a track interface unit. A sensor 80 detects ambient conditions such as light levels, sound, heat or movement and sends data representing the detected quantity to a microcontroller 82. The microcontroller also receives an input from track voltage detector 84, which enables it to determine signalling periods during which the power supply and the electrical loads have been disconnected from the track. The infoπnation received from the sensor is then sent over the track during signalling periods by connecting and disconnecting low impedance load 90 to and from the track. This is done by switch 88 and drive circuit 86, which may be of the same form as in the track switching units.

The track voltage detector 84 is also used for receiving information or commands from the power track. For example, the track interface unit may receive requests for data from the control unit.

A power supply (not shown) is provided to supply power to the microprocessor. The power supply may derive power from the track, or elsewhere, in the same way as the track switching units.

An infra-red detector (or ultrasound detector) may be provided to receive commands from a remote control. This has the benefit that the user may send commands to the system when looking generally in the same direction as the lighting on the track. The receiver is preferably arranged to transmit an encoded byte or other coded representation of a command or key press received from a remote control unit. Preferably, when such a sensor is included, it is preferable to employ a signalling protocol in which interrupts are provided to facilitate rapid transfer of information from the detector back to the controller. - 24 -

New Device Installation

Whenever a new system is installed the control unit is capable of detecting and registering all the devices on the track, either as a background activity or via a "register new device" command from the user.

In one example, all devices are allocated long ids at manufacture time of, for instance 32 bits. Using long addresses adds redundancy and security to the message transfer system, as noise will have to match a longer sequence to be confused with a genuine message. The idents may be algorithmically generated to have, say, a .known number of ones and zeroes, to maximise the value of this redundancy.

The wired OR nature of the track allows the control unit to detect the presence or absence of replies even when contention occurs, and consequently some initial logic may establish all the identities of all devices on the track at each power up, even if they are all newly connected simultaneously. This is a timeout based reply scheme where the timeouts are dependent on the identity details.

Individual track units include a unique device serial number (32 bits allows in excess of 4 billion units to be manufactured before identities wrap, 24 bits allows over 16 million before identities wrap). At power up in a new system the control unit establishes the long identities of each track unit by interrogation. It then establishes whether or not shortened identities (for example the eight least significant bits of the foil identities) will be sufficient to identify all the units in the system uniquely. In cases where these shorter identities can be used the message lengths can be reduced giving faster transmission times. This approach may be expanded to include a coded bit length field to allow very short identities to be used in small systems.

An alternative would be for the controller to soft-configure devices to respond to a shortened "local" identifier. A typical track is unlikely to have more than 256 devices, and usually will have fewer than 64, more typically fewer than 16 devices; this means that 8, 6 or 4 bits respectively should be sufficient to identify each device uniquely. This can be achieved by the controller reading the long device identities, assigning different short local identifiers to each device found, for example sequentially, and configuring the - 25 - devices to respond to the local device identifier. In this way, after the short local id is sent to a track unit, by using the long identifier for addressing, subsequent addressing may be based on short local ids.

Special interrogating messages and associated replies allow the control unit to establish the identities of all the track units within a reasonably short time. This process operates by sending interrogation messages to which units respond with a specially formatted reply. These interrogation messages have parameters which identify groups of identities. All units with identities within the given group reply simultaneously. The replies consist of two bytes sent in the normal signalling format (that is, with start, new message and stop bits and so on), but are interpreted differently to normal messages. The 16 bits which form the contents of the replies will normally be all "l"s, but any one unit shall pull any one bit to the "0" condition in order to indicate a part of its identity. The 16 bit positions shall allow units to indicate their identities 4 bits at a time and the wired OR nature of the bus means that two or more units can safely pull the same bit low.

The "plug-and-play" image may be enhanced by arranging each unit to transmit a relatively foil textual name on an appropriate command from the controller. Once the unit's identity has been established by the process described above, the textual name is passed to the controller via an appropriate sequence of messages in the normal format, and is then made available to the user. This confirms to the user that the unit has been registered correctly by the system, and clarifies the type of unit. An example of a textual name might be:-

20 WATT LAMP CONTROLLER

TYPE : 13 VER 1.1 PART CODE : 17-50-80

It will be understood that the present invention has been described above purely by way of example, and modifications of detail can be made within the scope of the invention. For instance, although the present invention has been described with reference to a track lighting system, the invention is applicable to any power distribution system in which a low voltage alternating current power supply feeds a power track to which electrical loads are connected. The electrical loads need not be lighting units but may be any device - 26 - which requires electrical power and to which data may be sent or from which data may be received. Although the invention may used most advantageously in a low voltage power distribution system, the invention may also be extended for use with higher voltages.

Each feature disclosed in the description, and (where appropriate) the claims and drawings may be provided independently or in any appropriate combination unless otherwise stated.

Reference numerals appearing in the claims are by way of illustration only and shall have no limiting effect on the scope of the claims.

Claims

- 27 -CLAIMS

1. A control unit for use with a low voltage power distribution system which comprises an alternating current power supply and a power track for supplying power from the power supply to at least one electrical load, the control unit comprising: control means for determining a signalling time in the voltage waveform for transmitting data; switch means for disconnecting the power supply from the power track at said signalling time; and transmitting means for transmitting a plurality of data bits over the power track during a signalling period while the power supply is disconnected.

2. A control unit according to claim 1 which is adapted to transmit at least four bits, and preferably at least eight bits, in said signalling period.

3. A control unit according to any of the preceding claims which is adapted to transmit messages which span at least two signalling periods.

4. A control unit according to any of claims 1 to 3 wherein the control means is adapted to determine a signalling time which is close to, but offset from, a zero crossing instant of the voltage waveform.

5. A control unit according to claim 4 wherein the signalling time is offset from a zero crossing instant by less than 4 milliseconds and preferably between 10 microseconds and 3 milliseconds and more preferably by at least about 0J millisecond and at most about 2 milliseconds and most preferably about 1 millisecond.

6. A control unit according to any of the preceding claims wherein the signalling period begins on one side and ends on the other side of a zero crossing instant of the voltage waveform.

7. A control unit according to any of the preceding claims, the switch means comprising a pair of transistors with opposite polarities. - 28 -

8. A control unit according to claim 7 wherein the transistors are field effect transistors, preferably MOSFETs, connected in series.

9. A control unit according to claim 8 further comprising gate drive means for supplying to the gates of the transistors a voltage sufficient to turn off at least one of the transistors during said signalling period, and a voltage which follows the supply voltage with a substantially fixed offset, such that both transistors remain on, at other times in the voltage waveform.

10. A control unit according to any of the preceding claims for use with track units which are arranged so that, during the signalling period, the track has a nominal impedance within a predetermined range.

11. A control unit according to claim 10 further comprising an impedance element which substantially determines said nominal impedance.

12. A control unit according to any of the preceding claims wherein the transmitting means is adapted to transmit data by connecting and disconnecting a current source to the track.

13. A control unit according to any of the preceding claims further comprising a reception circuit for detecting data voltage levels on the track.

14. A control unit according to any of the preceding claims for use with a low voltage track lighting system.

15. A power supply unit for a low voltage power distribution system comprising a low voltage transformer and a control unit according to any of the preceding claims.

16. A track unit for use with a low voltage power distribution system comprising an alternating current power supply and a power track for supplying power from the power supply to at least one electrical load, the track unit comprising: control means for determining a signalling period in the voltage waveform, the signalling period corresponding to a period during which the power supply is temporarily - 29 - disconnected from the track for data transmission; and transmitting means and/or receiving means for transmitting and/or receiving data over the power track during said signalling period.

17. A track unit according to claim 16 for connection between the power track and an electrical load, forther comprising switch means for disconnecting the electrical load from the power track during said signalling period.

18. A track unit according to claim 17 wherein the switch means is adapted to transmit data by connecting and disconnecting the electrical load.

19. A track unit according to claim 17 or 18, said switch means comprising a pair of transistors connected in series with opposite polarities.

20. A track unit according to claim 19 wherein the transistors are field effect transistors, preferably MOSFETs, with commonly connected gates.

21. A track unit according to claim 20 forther comprising gate drive means for supplying to the gates of the transistors a voltage sufficient to turn off at least one of the transistors during said transmission period, and a voltage which follows the power track voltage with a substantially fixed offset, such that both transistors remain on, at other times in the voltage waveform.

22. A track unit according to claim 16 for connection to a sensor for sensing an ambient condition forther comprising transmitting means for transmitting data from the sensor over the track.

23. A track unit according to claim 22 forther comprising a sensor selected from one of a heat sensor, a movement detector, an infrared detector, a sound detector, a light detector, a temperature detector or a voice recognition device.

24. A track unit according to any of Claims 16 to 23 forther comprising memory means for storing a program.

25. A track unit according to Claim 24, arranged to execute said program in - 30 - the absence of control commands from a track controller.

26. A track unit for use with a low voltage power distribution system comprising an alternating current power supply and a power track for supplying power from the power supply to at least one electrical load, the track unit comprising means for storing a program and means for controlling the electrical load based on the stored program.

27. A programmable track unit according to any of Claims 24 to 26 wherein the track unit includes non-volatile memory means, for storing a program, or program parameters, so that the unit can be programmed after manufacture.

28. A track unit according to Claim 27 arranged to receive the program or program parameters over the track.

29. A kit comprising a plurality of track units according to any of Claims 24 to 28, each unit having a respective unit program stored therein, the unit programs not all being identical, so that, when the units are installed on a track, the units together execute a desired ensemble program.

30. A kit according to Claim 29 wherein the units include means for signalling between units.

31. A track unit or kit according to any of claims 16 to 30 for use with a low voltage track lighting system.

32. Device for connecting and disconnecting a low voltage alternating current power supply from a load comprising: a first field effect transistor connected in series with a second field effect transistor in a back to back configuration, said first and second field effect transistors being connectable between the power supply and the load, said first and second field effect transistors having commonly connected gates; and a drive circuit which is adapted to supply to the gates of the transistors a drive voltage which is sufficient to turn off at least one of the transistors when the power supply is to be disconnected from the load, and a voltage sufficient to turn on both of the - 31 - transistors when the power supply is to be connected to the load, whereby said drive voltage follows the supply voltage with a substantially constant voltage offset when the power supply is to be connected to the load.

33. Apparatus for transmitting data in a low voltage power distribution system which comprises an alternating current power supply, a plurality of electrical loads, and a power track for supplying power from the power supply to the plurality of electrical loads, the apparatus comprising: means for disconnecting the power supply from the power track during a portion of the voltage waveform; means for disconnecting the electrical loads from the power track during a corresponding portion of the voltage waveform; and means for transmitting data over the power track while the power and loads are disconnected.

34. A track lighting system comprising an alternating current power supply, a power track connected to the power supply, and a plurality of lamps connected to the power track, forther comprising the apparatus of claim 33.

35. A track lighting system according to claim 34 forther comprising means for receiving data and means for controlling the lamps.

36. A method of transmitting data in a low voltage power distribution system which comprises an alternating current power supply, a plurality of electrical loads, and a power track for supplying power from the power supply to the plurality of electrical loads, the method comprising: disconnecting the power supply from the power track during a portion of the alternating current cycle; disconnecting at least one of the electrical loads from the power track during a corresponding portion of the alternating current cycle; and transmitting data over the power track while the power and loads are disconnected.