CA2434111A1

CA2434111A1 - System and method for the powering and fault protection of remote telecommunications equipment

Info

Publication number: CA2434111A1
Application number: CA002434111A
Authority: CA
Inventors: Clive Mullins; Martin Bijman
Original assignee: Catena Networks Canada Inc
Current assignee: Catena Networks Canada Inc
Priority date: 2003-06-30
Filing date: 2003-06-30
Publication date: 2004-12-30
Also published as: US7099461B2; WO2005006514A3; WO2005006514A2; US20050053228A1

Description

SYSTEM AND METHOD FOR THE POWERING AND FAULT PROTECTION OF
REMOTE TELECOMMUNICATIONS EQUIPMENT
[0001] The present invention relates generally to telecommunication networks and specifically to a system and method for the powering and fault protection of electronic equipment where said electronic equipment is connected to one end of a subscriber loop and the power source used to power said electronic equipment is connected to the other end of the subscriber loop.
BACKGROUND OF THE INVENTION

[0002] In telecommunications systems, some equipment may be located in a remote location where accessing a power source to power the equipment is not economical, or not desirable for the cost or marketing of the equipment. In these situations, the remote equipment may be powered from a power source through the same type of subscriber loop twisted pair wires (subscriber loops) that are normally used to deliver telecommunication services from service provider equipment to subscriber premises. In this situation, the powering scheme for the remote equipment is called loop powered.
Such remote equipment that may be loop powered may hlclude remote terminals (RT), pairgain devices, loop extenders, network interface devices (NID), optical network termination (ONT), integrated access devices (IAD), and subscriber communication equipment such as a POTS telephone, IP telephone, FAX, set top box, and data modem.

[0003] Referring to Figure 1, the overall environment for loop powered remote equipment is shown generally by numeral 100. The power source that sources charge 102 to the subscriber loop 104 is typically an earth referenced poweo~ supply with an output impedance of typically less than 5 ohms. The subscriber loop 104 is located physically among the external elements and may be subject to faults I06 from lightning and mains power lines. Primary protectors 108 are provided on each wire at each end of the subscriber loop. The remote equipment 110 consists of a protection circuit 120 which includes protection electronics and also a chaxge storage circuit with an input impedance of typically less than 5 ohms. Subsequent electronics in the remote equipment may include a power supply 130 that typically utilizes a transformer to isolate the application electronics 140 from the high voltages that may occur on the subscriber loop 104 due to faults 106 and a regulator to develop stable power supplies to power the application electronics 140.

[0004] In subscriber loops, due to their physical location among the external elements, it is commonly understood by those skilled in the art that faults may occur between the subscriber loops and foreign potentials, including, lightning induced current and voltage, and power cross and induced AC current from mains power wires. For subscriber loops that are used for the delivery of mainstream telecommunications services to subscribers who utilize, for example, P~TS telephones, modems, fax machines, and data modems, there is considerable prior art for the protection of subscriber loop electronics that source charge on one end of the subscriber loop, typically in the line card of the service provider, and electronics that sink charge from the other side of the subscriber loop, typically the subscriber location. However, this prior art is not applicable to subscriber loops that are used for the loop powering of remote equipment, in that the series impedance of electronics on each end of the subscriber loop utilized for loop powering, typically a few ohms, is considerably lower than the series impedance of the electronics on a subscriber loop used fox mainstream applications, typically 100 ohms or more. For example, the currents developed from a lightning strike of 1000V would be 20 times higher in the loop powered circuits, since the input impedance is 20 times smaller.

[0005] Almost always on all subscriber loops, primary protectors are provided that will shunt charge to earth when the potentials across the primary protector exceed several hundreds of volts. All circuits connected to the primary protectors must operate with consideration for the independent behaviour of these primary protectors.

(0006] The power supply that sources charge to one end of a subscriber loop is typically earth referenced, thus protection circuits from faults beyond the primary protectors may be crafted with relatively simple circuits or very often no additional circuits at all, that would shunt the fault energy locally to earth. Such protection circuits may be devised that maintain the connection between the power source and the subscriber loop throughout fault events, however this is not always the case.

[0007] Equipment located remotely that is loop powered must sink current from a subscriber loop however cannot be earth referenced rather would prevent the conduction of current to earth for voltages within the primary protector activation voltage range.
Further the energy that enters the remote equipment may be common mode when the fault influences both of the subscriber loop twisted pair wires, or differential which would occur when one of the primary protectors activates prior the other primary protector on the subscriber end of the subscriber loop. AlI things considered, the protection circuits for the electronics that sink energy from a subscriber loop that form the power supply of remotely powered electronics equipment are a new challenge.

[0008] For the protection of remote equipment, beyond the primary protectors, the prior art is to use simple electronic circuits that have several drawbacks. Fuses may be used however would require a service technician to visit after every fault.
Thyristors may be used to activate prior to the primary protectors or in coordination with the primary protectors, however as this design method requires the thyristor to conduct most of the energy in the fault event, they must therefore be large, and also the voltage developed across the thyristor may have large peaks during the fault event that is presented across the subsequent electronics, thus the subsequent electronics must be over-designed and expensive. Standards bodies are now testing more stringent, thus solutions based on thyristors will have greater difficulty achieving compliance. A relay or solid state switch to isolate the remote electronics may be utilized that is activated when sensors detect a fault event, however, a relay is mechanical thus prone to wear and tear, and either relay or solid state switch is utilized to disconnect the subsequent electronics when a fault occurs and later re-connected to the subscriber loop when 'the fault is cleared thus would result in a service interruption. These protection circuits are thus expensive to deliver and maintain, and result in interruption of service due to fault events.

(0009] It is also notable that a new generation of remote telecommunications equipment that is loop powered with copper subscriber loops and relies on optical fiber for all transmission would be subject to disruption of service if the remote;
equipment was susceptable to outages due to fault conditions that affect the copper subscriber loop but not the optical fiber. Such remote equipment may be required to add battery power if minimum service disruption were an objective, which is more expensive in capital and maintainance.

[0010] It is an object of the present invention to obviate or mitigate at least some of the above-mentioned disadvantages.
SUMMARY OF THE INVENTION

[0011] In accordance with several aspects of the present invention there is provided a circuit and process that forms the charge storage and protection circuit of remote equipment, that consists of a circuit that stores charge sunk from a subscriber loop interface, a circuit that connects the charge storage circuit to the subscriber loop interface, a circuit that connects the charge storage device to the subsequent electronics of the remote equipment, a process that decides when to activate these connections, and sensors that monitor the conditions of the circuit to facilitate sensing events that trigger different aspects of the process, such that the remote equipment remains powered thus continues to operate without going out of service during most fault events.
BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Embodiments of the present invention will now be described by way of example only, With reference to the following drawings in which:
Figure 1 is a block diagram representing the location of the protection circuit in the overall environment (prior art) Figure 2 is a block diagram of the protection circuit Figure 3 is a block diagram of the process 280 DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] For convenience, like numerals in the description refer to like structures in the drawings. Referring to Figure l, the environment in which the loop powered remote equipment resides is illustrated generally by numeral 100. In this environment and in the presence of the illustrated external protection electronics, a number of fault conditions may be presented at the input nodes 122 and 124 of the remote equipment.

[0014] When lightning may strike in close proximity to the subscriber loop i 04, current is induced in the subscriber loop 104 which develops a potential at nodes 122 and 124 that exceeds the activation voltage of the primary protectors 108. Due to the independent nature of the primary protectors 108 one protector 108 may change from open circuit to closed circuit before another one does, which results in a differential voltage appearing across terminals 122 and 124. These faults may have either polarity. This event shall be called a lightning event, and typically lasts tens of microseconds.

[0015] Similarly lightning may strike in close proximity to the subscriber loop 104, where both primary protectors on the wire connected to terminals 122 and 124 activate at approximately the same moment, thus the fault potential is experienced on both nodes simultaneously with the same polarity. This fault is a common mode potential fault event.

[0016] Due to the close proximity of power lines to the subscriber loop, currents typically as high as 100mA will be induced in the subscriber loop that will appear common mode on both wires at each end of the subscriber loop that, once they find a current path, will not create sufficient potential to activate the protectors 108. This fault is a common mode induced current fault, and may actually occur indefinitely during normal operation.

[0017] Similarly due to the close proximity of power lines to the subscriber loop, physical events may occur that result in the power line connecting to the subscriber loop, typically through some resistance due to physical aspects of the fault connection. This fault is called a power cross event, and may be sustained for long periods of time until the physical fault connection is removed and repaired.

[0018] Referring to Figure 2, a circuit block diagram of the charge storage and protection circuit (CSAPC) in accordance with an embodiment of this invention is illustrated generally by numeral 200. The CSAPC illustrated in Figure 2 would replace block 120 in Figure 1. The basic function of the components are as follows, however, these components may have other functions based on vaxious aspects of the invention.
Circuit 250 provides EMI suppression. Input diode 210 serves to block all reverse current from the subscriber loop that attempts to enter at node 204 and leave at node 202.
Switch 220 and resistor 222 which is typically 400 ohms serve to limit inrush current, and switch 224 is activated after the inrush event is completed. Charge storage circuit 260 consists of capacitor 262, and capacitor 266 through resistor 254 which is typically 0.5 ohms.
Lightning protection is provided generally by circuit 230. Linear circuit 232 presents the voltage sensed across switch 224 to one input of comparitor 234. Threshold selection circuit 240 outputs one of two voltages to the other input of comparitor 234, either Vlow when switch 242 is on, which occurs whenever switch 224 is on, or V:high when switch 246 in on, which occurs whenever switch 224 is off, as inverted by inverter 244. A very short delay 241 implements the change in threshold output from block 240 a short time after a change in state of switch 224, typically l0us. When the positive input to comparitor 234 is greater in potential than the negative input, its output would be on, and would turn on switch 224 depending on the state of AND gate 236. Switch 270 connects the charge storage circuit to the subsequent electronics connected to nodes 206 and 208.
Linear circuit 282 presents the voltage sensed across resistor 264 to the process 280, said voltage is proportional to the current through resistor 264. Linear circuit 284 presents the voltage sensed across capacitor 262 to process 280. Process 280, which in its physical implementation may be powered directly from capacitor 262, executes in a timed sequence a process based on linear input valtages Current and Voltage and outputs voltages at Inrush, Arm, and Connect that activate switch 220, switch 224 depending on the state of AND gate 236, and switch 270, respectively.

[0019] Refering to Figure 3, a block diagram of the process 280 in accordance with an embodiment of this invention is illustrated generally by numeral 300.
Irditialize operation 310 is the state of process 300 where the CSAPC is attempting to settle the charge storage circuit 260 to a stable voltage. Normal operation 312 is the state of pre~eess 300 when it is not in Initialize operation. Vmax is the maximum Voltage that can be tolerated by the subsequent electronics at nodes 206 and 208 under normal operation. Vpwm is a voltage margined several volts below Vmax that would be the Voltage delivered to the subsequent electronics when in pulse width modulation (PWM) mode. Imax is the Ioad current that is beyond the expected current range of the subsequent electronics under normal operation. Vmin is the Voltage that is too low to sustain the basic operation or functionality of the subsequent electronics. PWM mode is the state of process 312 of actively preventing charge from entering the charge storage circuit 260 such that Voltage does not rise higher than Vpwm. Linear mode is the state of process 312 when it is not in PWM mode. When in PWM mode, process 330 selects process 331 to execute the PWM, Process 332 detects when PWM is no longer needed, such that process 330 may select process 333 and enter linear mode. Process 320 detects if Voltage exceeds Vmax, and selects to enable PWM mode. Process 340 detects conditions which result when too little charge is entering the charge storage circuit 260 or too much charge is being drawn from charge storage circuit 260 such that Current exceeds Imax or Voltage :is less than Vmin respectively, and selects to reduce operating conditions 341 or power down 342 and enter Initialize operation.

[0020] Refernng to Figure 1, current supplied from the power source 102 on the far end of the subscriber loop 104 is normally DC in nature, which is required to enter node 122 and leave node 124. The voltage from node 122 to node 124 may vary. The voltage delivered by the power source 102 on the other end of the subscriber loop is called Vsupply, and may range from lOV to 200V. The subsequent electronics 130 and current load is called Iload. A subscriber loop 104 resistance may vary from one loop to the next from 0 ohms to 2000 ohms, in applications, the maximum P;loop that can be tolerated for a given Vsupply and Road may require the parallel use of several subscriber loops.

[0021] Refernng to Figure 2, protection diode 210 results in only a marginal decrease of the supplied voltage, less than one volt, at the subscriber loop interface at nodes 202 and 204 but serves several purposes. First, it prevents the CSAPC from being subject to reverse potentials at nodes 202 and 204, if the subscriber loop is connected backwards.
Further, diode 210 serves to reduce approximately half of all fault energy on average as the diode blocks all faults that cause the current in the subscriber loop to reverse. Thus, any faults that generate a potential at node 204 that is more positive than the potential at node 202 are blocked by diode 210 and are thus benign. Third, during a fault event where the current in the subscriber loop is AC in nature, diode 210 half wave rectifies the current and in essence supplies additional charge into the CSAPC that may be used to power the remote equipment during such a fault. As such it is an aspect of the present invention that the remote equipment could be normally powered directly from a mains AC power source located near the remote equipment rather than the powered subscriber loop, considering mains AC power sources are available in abundance and convenient.

[0022] Upon connection of the remote equipment to the powered subscriber loop, or when process 280 begins at initialize operation, Inrush is on which tunzs on switch 220, Arm are off which turns off switch 224 through AND gate 236, and Connect is off which turns off switch 270. Vhigh is set to be SV greater than the voltage seti;led across switch 224 at the end of a lightning event, which depends on the application selection of Vsupply, maximum Iload, which determines the maximum Rloop allowable, and resistance of resistor 222. For example, for Vsupply= 190V, Rloop= 600ohms, and resistor 222 selected to be 400ohms, Vhigh would be 75V. Vlow is set a few volts higher than the worst case maximum of two voltages. First is the voltage developed across switch 224 which corresponds to the desired current flowing through switch 224 in a lightning event at which point switch 224 is to be turned off. For example, if the on-impedance of switch 224 is selected to be 1.25 ohms when the maximum desired current flows through switch 224 at 5 amps, the first voltage is 6.25V. The second is the maximum decay across the charge storage device for one cycle of the AC power fail fault frequency under maximum Road. For example, in the charge storage total capacitance is SOOuF and maximum load is 700mA, the decay in one cycle of 60Hz is 6V. Thus for these two examples, Vlow would be set to lOV.

[0023] The power up process is now described. When the circuit is first powered, current from the subscriber loop initially flows to storage capacitor 262, through switch 220 and through resistor 222 which serves to limit the inrush current. As capacitor 262 charges, capacitor 266 charges virtually simultaneously through resistor 264. As capacitor 262 charges and the potential across it increases, the potentia across switch 224 decreases proportionally until it is lower than Vlow, such that comparitor 234 would remain on whether it is comparing the voltage across switch 224 to either threshold Vlow or Vhigh.
During the inrush event, process 280 monitors the rate of change of Voltage to determine when the inrush event has sufficient settled to ensure the voltage across switch 224 is less than Vlow, at which point Arm and Connect are set to on. With comparitor 234 on and Arm on, switch 224 turns on through AND gate 236 which effectively provides a parallel current path that shorts out resistor 222, thus connecting the subsoil>er loop interface nodes 202 and 204 to charge storage circuit 260, which more rapidly fully charges the charge storage circuit 260. With Connect on, switch 270 is on Which connects the fully charged storage circuit 260 to the subsequent electronics at nodes 206 and 208. Process 300 changes to normal operation.

(0024] Note fundamentally that it is one advantage of the present invention that the charge storage circuit 260 is located on the subscriber loop side of the protection switch 270 that serves to disconnect the subsequent electronics of the remote equipment from the subscriber loop only when necessary in a fault event, thus providing that the charge storage circuit 260 may continue to be connected to the subscriber loop during fault events, and as such will collect charge from all or a portion of the fault event for the purpose of powering the remote equipment. This advantage allows in most cases the remote equipment to remain in service throughout a fault event, as opposed to the prior art in which the intent is to disconnect the fault event energy from the remote equipment.

[0025) Note further that it is yet another advantage of the present invention that the presence of the process 280 allows flexibility to implement the optimum.
behaviour of the CSAPC for various fault events, as sensed by Current and Voltage, processed by process 280 with control of outputs Connect, Inrush, and Ann, such that the subsequent electronics are subj ect to minimum stress and rapid response to the fault event, resulting in the CSAPC and subsequent electronic components in the remote equipment need not be oversized or more rugged than required to deliver their .functic>n under normal operation, which reduces cost of the remote equipment. Further this optimization maximizes the probability of avoiding service interruption due to fault events. Further that by implementing process 280 as software rather than a state machine allows ease and flexibility to adapt to present and future powering and fault requirements and features.

(0026] In the event of a lightning fault where node 202 is more positive than 204, the transient event is recognized within the industry to have a rise time to peak voltage and further fall time to half the peak voltage of either l0us to 1000V and l'~
000us further to SOOV, or 2us to 2500V & l0us to 1250V, respectively. This transient speed typically exceeds the cumulative response time of sensors, process algorithms, and subsequent switching events, mitigating the need for a local independent circuit to deal with the fault.
In the present invention, such circuits are provided by circuit 230. As lightning current flows through switch 224, a voltage develops across it proportional to it on-impedance.
When this potential exceeds Vlow, comparitor 234 turns off, and turns off switch 224 through AND gate 236. The lightning current continues to flow safely through switch 220 and resistor 222 into the charge storage circuit, thus adding more charge to the charge storage circuit to power the subsequent electronics during the fault.
With switch 224 off, threshold circuit 240 outputs Vhigh to comparitor 234. After switch 224 turns off the fault increases to its maximum voltage as seen at nodes 202 and 204, then decays until the voltage across switch 224 decreases to a voltage dependant on Vsupply, Rloop, Iload, and the resistance of resistor 222 which under all application conditions settles below Vhigh, at which time eomparitor 234 turns on which turns on sv~ritch 224 through AND gate 236, after which the potential across switch 224 falls below Vlow more rapidly than the delay 241 prevents the switching in the threshold selection circuit to Vlow, such that the comparitor 234 remains on thus switch 224 remains on through AND gate throughout the termination of the lightning fault event, and the circuit:
continues under normal operation.

[0027] In the event of a common mode potential fault, both leads 202 and 204 will simultaneously reach a potential at the activation voltage of the primary protectors, which is typically several hundreds of volts. In the loop powered remote equipment, it is required that all interfaces to earth potential block voltage to a level greater than the worst case activation voltage of the primary protectors, typically 1000V. Thus as all voltages developed in the remote equipment are below the voltage blocked at all interfaces to earth potential, thus this fault it benign.

(0028] In the event of common mode induced current fault, the potential at the powered end of the subscriber loop is tied to earth through typically a few ohms, thus does not develop any significant potential, and is expected to continue in zuormal operation indefinitely in the presence of this fault. In the loop powered remote equipment, the common mode induced current fault currents which typically are not exceeding 100mA
per subscriber loop wire are blocked with respect to earth potential at all interfaces, thus do not flow into the CSAPC, as a result they must flow through thc; subscriber loop resistance of typical maximum 1000 ohms per wire to the powered end of the subscriber loop which is tied to earth, such that the common mode potential developed at node 202 and 204 reaches maximum positive or negative 100V, which will not activate the primary protectors, nor does it exceed the voltage barrier to earth as previously discussed is typically 1000V, thus this fault is benign.

[0029] In the event of power cross fault, or when powered from local .AC mains power source, AC current enters the remote equipment that is half wave rectified by input diode 210. In the instance that the AC current arrives when the remote equipment is running in normal operation, charge from the AC current will be is added to char;;e storage circuit 260, and Voltage will rise. Depending on the charge contributed by the AC
current, Voltage may not rise above Vmax to require any action, and process 280 will continue in linear mode. If Voltage rises above Vmax as detected by process 320, PWM mode would be enabled. PWM mode is implemented by process 331 by pulse width modulating Arm to control the state of switch 224 through AND gate; 236 to regulate Voltage at Vpwm while the CSAPC is under load by the subsequent electronics, where inductors 252 and 254 of EMI circuit 250 serve to minimize the cuwent peaks of the regulation. Inrush should be turned on/off. The PWM process 331 would phase lock to the half wave rectified waveform using Voltage and Current, and would turn Arm on during the half period when the waveform is blocked by diode 210 to reduce noise, and modulate the turn off timing of Arm to occur sometime during the half period when the waveform is not blocked by diode 210 when charge is being added to the charge storage circuit 260. The maximum decay of on the charge storage circuit 260 must be less than a margined amount below Vlow in order to ensure that comparitor 234 remains on constantly during the PWM process, in order for switch .224 to activate through AND
gate 236 when Arm is turned on, and to ensure circuit 230 will respond in the event of lightning while in PWM mode. In the example given earlier, if the charge storage total capacitance is SOOuF and maximum load is 700mA, the decay in one cycle of AC
current frequency 60Hz is 6V, in which case the selection of Vlow at lOV would ensure proper operation. As such the operation of the remote equipment may continue indefinitely in the presence of added AC current. When the fault is removed or for instance the AC
current is reduced, the additional charge added to charge storage circuit 260 will be reduced, Voltage will decrease, the pulse width modulation will no longer be required as detected by process 332 and operation will return to linear mode by process 333.

[0030] It may occur due to a fault event that the equipment providing the power at the powered end of the subscriber loop will activate its own protection circuitry, and disconnect from the loop. In this instance, the operation of the remote equipment would have to be sustained solely from the charge delivered by the fault, which may not be sufficient, in which situation Voltage would decrease. Also, it may occur that under normal operation of the remote equipment that excessive functionality or services are enabled or an equipment failure occurs, or a fault occurs within the subsequent electronics, in which situation Current would increase. It is yet another aspect of the present invention that process 340 may determine if Current exceeds Imax, or if Voltage 1 S is below a voltage Vmin, to execute process 341 which would reduce in graduated fashion the operating conditions the subsequent electronics. Such a change in operating conditions may include the dynamic reduction in services provided by the remote equipment. If the changes in operating conditions cannot be reduced further, process 342 is selected to power down the remote equipment, and proceed to enter Initialize operation.

[0031] Note that it is a further advantage of this invention that the subscriber loop used to loop power the remote equipment may also be utilized to carry the transmission information to the remote equipment using a transmission protocol, including for example the POTS frequency band, ISDN, xDSL, or other protocols as can be applied in this application by one skilled in the art.

[0032] Although the invention has been described with reference to certain specific embodiments, various modifications thereof will be apparent to those ;9killed in the art without departing from the spirit and scope of the invention.