Numerous systems exist where a signal of some type is superimposed on a power signal. In these systems, the communication information is separated from the power signal in the frequency spectrum such that through proper frequency separation techniques and with proper encoding and decoding techniques the information can be extracted from the power signal. These systems can perform very well and meet many design needs. Systems comprising a high frequency carrier, upon which is impressed the encoded information, requires a tightly specified and controlled transmission medium such that the high frequency carrier is not lost and signal integrity is maintained. These systems generally tend to require more complex encoding and decoding hardware. If the proper techniques are not used, transmission distances and data transmission integrity can be limited. If good techniques are used, very high data rates can be achieved.
In the design center, there are often requirements to use existing wiring such as in sprinkler systems where it is both cost effective and much less disruptive to the environment to use existing wiring. A typical system may have an individual wire to each valve and a common ground. This type of lay out is not well suited to high frequency encoding systems where transmission line characteristics are important. With conventional systems, adding additional valves means installing a new wire for each new valve.
Fowler (U.S. Pat. No. 4,093,946) teaches the fundamentals of communication over a two-wire system and method for interrogating a plurality of data gathering devices for actuation of selected ones from which data is received. Transducers, as taught by Fowler, are typically remotely located from the interrogating and receiving apparatus and are connected via a single two-conductor path over which power is conveyed to the transducers and data signals conveyed between the transducers and the receiving apparatus. Fowler teaches that turning on current source and causing a constant current to be propagated down cable can produce a binary signal. He continues that voltage modulation can be can be induced by alternately placing a high impedance and a short circuit across the conductors of the cable, while the current source is continuously enabled. The voltage on the cable can be caused to alternate between two binary states.”
Shimada (U.S. Pat. No. 4,139,737) applies the two-wire communication concept through the addition of a time division multiplex transmission system in which electrical power is transmitted to remote terminals from a central unit simultaneously with address and control signals.
Shimada also references that adding modulators and demodulators to power lines was well known in 1979. “Alternatively a pair of transmission lines have been arranged between the central unit and the respective terminal units and, further, modulators and demodulators have been inserted between the central unit and the said transmission lines and between the respective terminal units and the transmission lines so that the address, control, response and the like signals have been superposed on electric power waves to be transmitted between the central unit and the respective terminal units. However, with this arrangement, the modulators and demodulators have been required, the system formation has been complicated and the cost reduction has not been able to be attained.” This technique is widely used in both AC and DC two-wire systems. Systems with high frequency carriers require special care in impedance matching and layout and are limited in distance. They also require a certain level of cost and complexity.
Horn (U.S. Pat. No. 4,208,650) in 1980 teaches the concept of using a basic message protocol where “The transmitter is connected to a data containing unit and operates through a data cycle which addresses the data unit to provide a plurality of serially arranged message frames during each data cycle. Each message frame includes a marker bit, a sync word, data words having either digital or analog information, address and error words, a checkword, and various parity and start bits. The receiver, which is connected to another data unit, is adapted to recognize each message frame by means of the marker bit and the sync word, and evaluate the message validity by means of the parity and start bits, and the checkword.”
These four patents teach the following principles:
There have been significant refinements and variations over time including the following. Each variation was developed to meet specific requirements demanded by the projected usage.
Demeyer, et al. (U.S. Pat. No. 5,089,974) in 1992 teaches a system similar to the present invention. The Demeyer abstract reads:
“A building power management controller comprises a plurality of modules connected by a two-wire network. Each module comprises a data transceiver device, controlled by a microprocessor, to both transmit data to the other modules and to a central unit via the two-wire network, and to receive information via this two-wire network. The modules are supplied with power by the two-wire network. When two modules transmit simultaneously, one takes priority so as not to disturb the messages transmitted.”
Demeyer utilizes the concept taught by Fowler above, where a current source is applied and an alternate sink and high impedance load is used to generate the timing for the coding scheme. Demeyer also teaches that super imposed communication signals can be handled by sensing an out of phase bit through a comparator means where a difference in state between the output of the transmit bit and the input from the two-wire level sense comparator is sensed. The comparator is will indicate by an output level if there is a difference between the lines while the module is in the low state. The microcontroller in the module can sense this level and withdraw from the communication cycle letting the other device continue.
Demeyer does not teach the ability to use this capability to automatically configure the system.
This invention teaches communication and power distribution design wherein both the power for operation of remote sensing elements and remotely actuated elements can be transmitted down a single pair of wires together with the communication or networking protocols necessary for operational control of and data gathering from the remote devices. While the techniques discussed are applicable to many control and data gathering activities, this invention is especially suitable for the control of heating and air conditioning systems, sprinkler control systems, security and lighting controls and industrial automation controls. The present invention teaches a novel and robust protocol particularly well suited for two-wire systems, such as irrigation systems, that were not specifically designed for data communications.
This two-wire system facilitates expansion by extending the existing wiring with additional two-wire lengths to reach the new locations. The two-wire system can be connected in a loop configuration that provides a redundancy level not available in other topologies. A loop provides the ability to continue to operate with a single break in the two-wire loop. In the heating and air conditioning environment, it is often difficult to add wire, and being able to use existing thermostat wiring can be an installation and cost advantage. The same holds for lighting and security control.
This invention also teaches a novel method to automate the discovery of new devices added to a two-wire system to facilitate automatic configuration. The method teaches using a specialized protocol that allows primitive wiring systems to be used to permit multiple devices to engage in sophisticated and reliable data communication.
During typical operation, the base unit 200 receives data from and transmits data to a control computer 100 using a standard interface 103 such as an industry standard Universal Serial Bus (USB) or an RS 232 interface, or alternative, a proprietary bus. Remote unit commands are encoded and delivered to the one or more remote units 250 for action on the two-wire bus 140. Responses are received from the remote devices 250 by the base unit 200 over the two-wire bus 140. The base unit 200 manages the basic two-wire bus 140 by detecting shorts and opens on the two-wire bus 140. The base unit 200 also has the ability to detect the basic quality of the transmission received, and monitor degradation of the two-wire bus 140 by accurately measuring the time between each transition and computing a quality factor based on the variation of the transition timing with the expected timing. The base unit 200 is composed of a microcontroller 205 that provides the computational processing necessary. The power source 220 provides power at two levels. The high power level, activated by raising the high power line 210 to a logical one, provides operational power for the devices. The device is current limited such that a shorted wire will not harm the system. The microcontroller 205 monitors the power provided to the two-wire bus 140 such that if it exceeds a preset level for a fixed length of time, the microcontroller 205 will shut down the power to the two-wire system and report to the controlling computer 100 that an “over current” situation has occurred. The microcontroller 205, using the current sense 235 and voltage sense lines 230 provided, monitors system power usage and is able to measure the voltage drop on the two-wire connection 140 during communication with a remote unit 250. This voltage drop is transformed into a resistance measurement of the wired connection, which provides an objective measurement of the quality of the two-wire bus 140, and can indicate degradation over time. The two-wire bus 140 is switched to low power mode in preparation for communications by lowering the high power line 220 and by raising the low power line 215, thus providing the appropriate power for the communication process. In another cost effective embodiment, the function of the controlling computer 100 is embedded in the base unit 200.
Following the postamble 325 of a command message 320, the protocol allows the appropriate remote unit 250 to respond. The response is analogous to the command sequence just described except in the response phase, the remote unit 250, rather than the base unit 200, encodes, then asserts, the message on the two-wire bus 140. The preamble phase of the command phase ends in the high voltage state 330, analogous to the initial high voltage state 305. A response preamble 310′ is asserted by pulling the two-wire system 140 to the low state for a specified period of time, ending with a high voltage assertion 315′. Following the response preamble 315′ a multi-byte response message 320′. Following the response message, the remote unit 250 asserts a postamble phase 325′ for a proscribed period time, and then returns to the high voltage state.
The multi-byte command message 320 and the multi-byte response message 320′ are shown in more detail in middle diagram of
Diagram 370 illustrates the encoding of a typical byte. Each byte is encoded as a serial stream of bits by alternating from a low to high state. In the preferred embodiment, the bytes are encodes as 8 bits bytes numbered 1-8. Bits 1, 3, 5, and 7 (381, 383, 385, and 387) are encoded on low voltage cycles and bits 2, 4, 6 and 8 (382, 384, 386, and 388) are encoded on high voltage phases. The first bit 381 is pulled low by the encoder sinking current to bring the voltage on the two-wire system to a logical zero. If the communicated bit is a “one” then the bit is held low for three units of time. If the communicated bit is a “zero” the bit is held low for only one unit of time. In the example illustrated in timing diagram 370, bits 1, 4, 5, and 8 are “ones” and bits 2, 3, 6, and 7 are “zeros” resulting in an encoded byte value of 10011001. At the conclusion of the required time specified by the bit being a “one” or a “zero” the encoder releases the line and the current source in the power source drives the two-wire potential to full voltage. The voltage remains at the high level for one unit of time if the next bit is a “zero” bit or three units of time if the bit is a “one” bit. This process continues until each bit of the data field, in the communication cycle, has been impressed on the two-wire bus. A final “low” on the two-wire for a set amount of time signifies the post-amble or end of message. In the preferred embodiment, “zero” bits are one unit of time long, and the “one” bits are three units of time long, but any encoding where the time units are of substantially different durations will suffice.
The basic concept of utilizing one time period for a logical one and different time period for a logical zero is well known in the art and was first used between 1970 and 1980 to encode computer data on computer magnetic tapes. The prior art technique is disfavored in many modern applications because it requires at least one transition per bit and contains a DC component that is difficult to decode accurately in a magnetic read channel. However, in the present invention, the disfavored technique has been adapted in a novel way in the present invention to be very useful, resulting in a cost effective solution that works well in an electrically hostile environment. The present invention enhances the technique by adding the necessary definition, preambles and postambles necessary to form a unique and cost effective communication solution for hostile environments, such as sprinkler control systems with existing two-wire systems already in place. In the preferred embodiment, the two-wire signal moves between zero volts and the power supplies voltage and therefore is well adapted to a signal containing a direct current (DC) component.
In the preferred embodiment, both the command message 320 and the response message 320′ are composed of a number of eight bit bytes. A protocol is imposed on the format of the command and response messages so that the desired functionality can be achieved. The system operates on a command and response operation model. The controlling computer 100 sends a command 320 to a base unit 200. The system provides an addressing method for communicating with multiple base units 200. This command 320 is either processed by the base unit 200 or the encapsulated message is forwarded to the addressed remote unit 250 over the two-wire system 140. When the addressed base unit 200 has processed the command, if the command requires a response, the base unit 200 composes a response message 320′ and sends it back to the computer. If the command is a pass-through command, the encapsulated message is sent over the two-wire network 140 to the addressed remote unit 250. The base unit 200 will then wait for a response 310′ from the remote unit 250. This response 310′ will be encapsulated by the base unit 250 and then sent forward to the controlling computer 100. If there is no response from a remote unit 250 within the allowable time, then an error message is sent to the controlling computer 100 indicating that no message was received. In the preferred embodiment, all messages are protected by checksum codes to verify message integrity. The base 140 unit does not interpret messages that are passed-through to remote units 250.
At the conclusion of the reception of a command message 320, the remote units 250 can use this time to perform any commands sent during the command sequence. At the end of the minimum specified processing time, the remote unit 250 addressed by the command message activates and sends a preamble. The remote encoder 240′ sinks the current supplied by the base unit power source 220 bringing the voltage on the two-wire bus 140 to a logical zero. The low level indicates that communication from the remote unit 250 will follow. After a prescribed period, the remote encoder 240′ then allows the voltage to rise to a logical high level allowing the base unit 200 to prepare for the communication cycle to follow. The remote devices 250 use the same protocol for communication and hold the bus low for one unit of time if the bit is a “zero” and three units of time if the bit is a logical “one”. The entire message is composed of nine bytes. Each byte is composed of eight bits. The first seven bytes composed the body of the response message. The last two bytes are a checksum composed of the sum of the previous six bytes. The base unit computes the checksum on the first seven bytes and compares it to the received checksum. If they are identical, the message is valid. If they are not, the message has been corrupted during transmission and is discarded.
In addition to provide robust communications between the controlling computer 100, one or more base units 200 and one or more remote units 250, the present invention also teaches a novel, robust system and method for device detection. In the preferred embodiment, every remote unit 250 has a unique serial number. When the system is configured, each remote unit 250 is assigned a unique address on the two-wire system 140 such that it can be addressed easily and independently. During the configuration process, it is advantageous for the system to be able to perform an auto discovery function and detect all remote units 250 connected to the two-wire system 140. The method of the preferred embodiment of detecting potential collision and avoiding data corruption is novel and enables this robust auto detection and identification function. The two-wire power is supplied through a current source 220 that pulls the two-wire voltage 140 to the required voltage level. An encoder 240 on the base unit 200 or the encoder 240′ on one of the remote units 250 sinking the current and pulling the two-wire voltage low produces a logical two-wire low level. The encoders are open collector or open drain devices and as such can be activated concurrently without damaging the device. However, if two devices both answer and attempt to send a full response, the response from both devices would mix and produce a corrupted return. The method of collision avoidance requires that prior to each transition from a high two-wire voltage to a low two-wire voltage; the sending unit checks the level of the two-wire voltage before activating its encoder. If the voltage is low already, the remote 250 does not enable the encoder 240′ and terminates sending the response message 320′. This allows the competing response messages 320′ to be sent without corruption. The system discovers the remote units 250 on the system by transmitting a return serial number command to all remote units 250. Each remote unit 250 will attempt to transmit its serial number. Each remote unit 250 will wait a pseudo-random delay before attempting to send a response 320′. As each remote unit 250 begins to send a preamble 310′, it checks to see if the line is already low using the remote sense/control circuitry 260. If the two-wire 140 is already low, the remote unit 250 discontinues its response 320′ and will wait for the next command 320 from the base unit 200. Because the remote units 250 start their response messages 310′ with some time variance, most of the remote units 250 will drop off during the generation of the preamble 310. If by chance multiple remote units 250 happen to send a synchronized 310′ preamble, remote units 250 will dropout as they detect potential collisions with their data stream. As a remote unit 250 detects a potential collision, it will back off and wait for the next command 320, allowing the other remote unit 250 to complete without error. Because each remote unit 250 has a unique serial number, each remote unit 250 will be responding with a different data stream, which will guarantee that only a single remote unit will complete successfully. The remote unit 250 that communicates its serial number successfully is then given a command to “sleep” and the process is repeated until all remote units are found and put into a sleep state. Remote units 250 no longer respond when in the sleep state. All remote units 250 will have been found when there is no response 320′ to the return serial number command.
The description thus provided illustrates the preferred embodiment of the invention and is provided by way of illustration and not limitation. One skilled in the art can and likely would make variations that are nonetheless within the scope and spirit of the invention. For example, variations involving the exact ordering and definition of the command protocols, the voltage levels whether AC or DC used, the types of systems deployed are examples of the parameters and contemplated by the present invention. The invention should only be limited by the claims as set forth below:
This application claims priority of provision patent application No. 60/326,074 filed Sep. 28, 2001.
Number | Name | Date | Kind |
---|---|---|---|
6112246 | Horbal et al. | Aug 2000 | A |
6298377 | Hartikainen et al. | Oct 2001 | B1 |
6397220 | Deisinger et al. | May 2002 | B1 |
6473783 | Goshey et al. | Oct 2002 | B2 |
6615212 | Dutta et al. | Sep 2003 | B1 |
6801507 | Humpleman et al. | Oct 2004 | B1 |
6993585 | Starkovich et al. | Jan 2006 | B1 |
7269660 | Westfield | Sep 2007 | B1 |
Number | Date | Country | |
---|---|---|---|
20030097482 A1 | May 2003 | US |
Number | Date | Country | |
---|---|---|---|
60326074 | Sep 2001 | US |