WO1990007235A1 - Cable systems or the like - Google Patents

Abstract

A system and method is described for expanding the load capacity of a cable system (15) or the like supplied with AC power from a source (10), without increasing the voltage-drops caused by load currents, by coupling additional loads to such a cable system (15) only during intervals ('Windows') of each half-cycle of the AC power during which instantaneous current values for existing loads (21, 25) are zero or near zero. The existing loads may be cut off from the power source (10) during such intervals. Also described is an arrangement for locating and replacing defective amplifiers (123) in a cable system, and for controlling gain, using a field strength meter (151), and for temperature compensating, using a sensor (165, 166), in a cable system by interposing or removing amplifier stages (123).

Description

CABLE SYSTEMS OR THE LIKE The present invention relates to cable systems for transmitting signals such as television, audio, computer, data, facsimile, or other signals, and particularly transmitting signals over co-axial cables in a distribution network. The present invention concerns the provision of enhanced power- carrying capacity and gain control for existing cable com¬ munication systems, such as cable television or CATV. Background of the Invention In cable systems, a multitude of signals, such as television, facsimile, audio, radio frequency, or computer data signals may be transmitted separately or simultaneously through a coaxial cable in a distribution network. Conventional cable communication systems usually include a plurality of broad-band radio-frequency amplifiers spaced along a coaxial cable system in a tree-and-branch arrangement. Their function is primarily to restore the desired radio-frequency signal or signals to ap¬ propriate levels to overcome the effect of attenuation by signal losses associated with cable dielectric and conductor resistance. Such signals may be fed by the cable system to various outlets or taps, such as cable subscriber homes, for communication, entertainment and educational purposes.

To provide operating power to the amplifiers of such a system, it has been conventional practice to supply energy to the same cable at various power injection or feed points, such as from local alternating current power sources, which then supply power to nearby power-consuming devices such as amplifiers along the cable system. ^■ A common power source may have an alternating voltage of 60 volts at a frequency of 60 hertz, with a current capacity of 14 amperes. This power is transmitted along the cable to each amplifier location, and is used at the amplifier by power supply circuits in the amplifier which generate the necessary voltages required by the active circuitry of the amplifier for transmitting the desired radio- frequency signals along the cable system. It is usual to combine this power transmission on the same cable with the transmission of desired smaller-magnitude radio-frequency signals being relayed to customers or subscribers.

Such power injection points are provided at spaced intervals along a cable system. The separation between power injection points is determined by the voltage-drops caused by current flow along the cable, which in turn depends mainly upon the power requirements of the amplifiers needed to boost the intelligence (e.g., radio-frequency) signal to maintain a signal level useful for the subscriber taps. In general, from a 60-volt source, voltage drops providing amplifier power supply input voltages down to 45 volts may be tolerated, but not less. A power injection or feed point may provide power transmitted in several directions, both "downstream" and "upstream" from the feed point, to appropriate power-using devices. The power requirements for a system are determined when the system is first designed, and the power injection points are chosen to meet the designed power requirements, with appropriate allowance for expansion. The locations of these power feed points are determined in cooperation with the power company which supplies the energy used by the CATV power supply units, and other utility companies (e.g. telephone companies) which may provide f cilities for supporting and maintaining the cable system. It is costly and impractical to proliferate such power feed points beyond what is the anticipated requirement, possibly with allowance for anticipated expansion. Typically five amplifier locations may separate each adjoining pair of power injection points along a cable system. In many instances it is desired to add an additional service on the same cable system, or to extend the system beyond what was anticipated during system design. If the equipment required for this expansion draws more power than the system was designed to provide, it may be necessary to add additional power injection points. This very expensive step is necessary because neither the current nor voltage may be increased indiscriminately. Fur example, increasing the voltage above 60 volts is prohibited by the National Electric Code. An uncontrolled increase in the peak current could lead to intolerable voltage drops along the cable, or induce noise, in the coils and ferrites of the system's passive components.

It is an object of the present invention to provide a system in which the power available in the system may be used more efficiently, allowing additional devices to be supplied with power without increasing the peak current drawn. This eliminates the necessity to repower the system, which would be very costly.

In this way, new services requiring additional power- consuming equipment may be added to an existing cable system by including control circuitry in the equipment which will control the specific point in time and the time duration during which the new equipment can draw current. In this manner, the new equipment will draw electric current only when current drawn from the existing equipment is at or near zero.

According to the present invention, additional power is transmitted along the cable to new equipment at intervals of time when power is not being transmitted to existing equipment. In this manner., current consumption does not add on an instan- taneous basis and peak current is minimized. Since peak current is minimized, the voltage drop caused by the resistance of the cable is also minimized, thus increasing efficiency.

According to one feature of the invention, this time¬ sharing method of transmitting power can be applied to an entirely new system. In this application, the timing control circuitry will be included in all power-consuming equipments in the cable distribution system. The control system will optimize the current consumption such that the simultaneous current drawn by different devices will be minimized. This will minimize peak currents which will minimize voltage drops on the cable and thus provide greater efficiency. According to another feature of the invention, minimizing current peaks in the cable system will minimize the induction of noise and hum modulation in the coils and ferrites found along the path of such systems.

This improvement to supplying power to cable systems is equally applicable to any wire or guided wave transmission system, whether using paired wires or coaxial cable, or to a master antenna system, to transfer signals with necessary power to operate remote active equipment.

By way of illustration, it may be desired to add a new service formed of frame store units (FSU) on an existing cable system. Such FSU requires a specified amount of power, which most existing cable systems do not have sufficient excess capacity to serve, either because any reserve power capacity has been earmarked for and is being conserved for later expansion, or the existing system may be using fully the capacity of its power supply units. While such additional FSU units may be supplied with separate direct power from power lines, this is not economical or practical because it requires special construction to meet electrical codes and special installations at each location by the local utility company, with substantial expense and long lead times for scheduling costly utility installations.

Particularly where such added power-consuming devices have special voltage requirements (such as in low voltage systems) not only would long lead times be required for utility installations, but there would be in addition a high cost for additional multiple individual power supplies, and possibly a high cost for paralleling the existing cable system with new cables. Similarly, redesign and rebuilding of the system power sectors by changing feed points to obtain sufficient power for both the existing system and the additional FSU's would entail considerable expense, require long lead times to obtain new utility power locations, and might cause disruption of the cable system operation.

According to the present invention, a system is provided which can add further power-consuming loads with or without added power sources for them, on existing systems without requiring additions to the physical cable itself.

According to another aspect of the invention, an improved arrangement is provided for carrying signals (such as television, facsimile, audio, radio frequency or computer data signals) thorugh a coaxial or fiber optic cable in a distribu¬ tion network.

Such a network is usually composed of (1) a super- trunk, comprising a cable and associated amplifiers intended to transport signals from a head end or source to one or more hubs which in turn serve different geographic areas, and (2) trunks which supply such signals from a hub to sub-hubs and distribu¬ tion or feeder cables serving individual recipients within the service area of a hub such as an area of 20 to 100 square miles, depending on the type of system. Such systems carry signals over long distances, desirably with a minimum of electrical noise and distortion. Most of such super-trunks and trunks are restricted in overall length to less than ten miles because of noise and distortion limitations. Trunk cables are split at strategic locations to branch the network repeatedly to cover a larger area, but are generally not tapped to feed subscribers directly, since the losses incurred by such tapping is generally high and would reduce the distance the trunk could properly serve. Instead, the trunk at spaced intervals usually feeds bridger amplifiers which serve feeder cables which in turn serve subscribers. Due to normal cable attenuation, trunk cables employ trunk amplifiers to reestablish signals to proper amplitude. Such amplifiers typically have a gain between 18 and 26 decibels.

Emanating from each bridger amplifier are feeder cables which are routed past subscribers' residences in the bridge amplifier service area. At each cable supporting pole, or at a property line where underground cables are installed, a tap may be provided installed for the purpose of obtaining a portion of a signal from the feeder cable, to serve the subscriber drop. Insertion loss at each subscriber tap is relatively high, dependent upon the relative strengths of the signal on the feeder cable at that point and the signal level needed for the drop. To overcome such insertion loss, line- extender amplifiers are used to boost the signal level at spaced locations alone the feeder cable. However, the number of line extenders which may be cascaded is limited to a small number, due the high output levels typically used in the industry, which increase distortion.

According to this aspect of the present invention, a minimal loss cable architecture is supplied, based on the concept of Neperian amplifier spacing. This concept uses an optimum spacing for radio frequency amplifiers which limits a cable signal level drop or attenuation to approximately 8.7 db and consequently uses a like gain for each amplifier along the cable, providing minimum noise and distortion of signals. This requires a much closer spacing than the amplifier spacing typically used in the cable television industry today, where cable system designers generally choose amplifiers with the highest gain possible, to limit the number of amplifier stations in the system, in an effort to keep equipment costs to a minimum. Thus, a few amplifiers, operating at relatively high gains which afford greater amplifier station spacings, are generally used to favor economy and cost, but at the expense of performance and increased complexity to maintain good perfor¬ mance.

One object of the present invention is to greatly reduce the cost of amplifiers, to make the higher performance Neperian architecture economically desirable. This is done by simplifying amplifier design, which reduces component count and hence power requirements and cost, and which also improves reliability of operation of the amplifiers. Reliability is also enhanced by another feature of the present invention, providing essentially fail-safe operation by including bypass switches to route signals around a failed or defective amplifier. Such bypass switches may also be used to provide an automatic level control for the amplifier cascade, to keep operating levels within appropriate limits for preserving effective use and performance, and to provide temperature compensation.

According to the present invention, cable trunk and/or distribution system is provided having multiple sections of cable and amplifiers connecting a headend with one or more remote terminating locations or connecting a hub to bridger amplifiers, and/or connecting bridger amplifiers to subscriber taps. Located at the headend is a control computer with outdoor temperature sensing equipment, a cable modem for com¬ municating control data to the individual trunk and/or dis¬ tribution amplifiers, and a receiver modem connected through any of a number of media (such as phone lines, radio transmis¬ sion links or RF cable) to receive data on the status and performance of signals received at the terminal end. At the hub site or at the end of any line of amplifiers an automated RF level sensor is employed to measure the amplitude of signals (which may include a calibrating signal) received from the cable trunk or distribution feeder. Also located at the terminal end is an outdoor temperature sensor to obtain temperature data*. A feedback loop for control purposes is provided from the terminal end back to the head end. All amplifier and terminal end equipment may be small enough to be pole-mounted in an environmental housing. Terminal end equipment may also be referred to as "hub" equipment for the purpose of this description.

The control computer at the head end collects the data from the hub as to signal status, as well as data on temperature at the hub,and processes these data to determine the optimum operating configuration of the trunk system, as discussed below.

For optimum performance, the gain of the amplifiers in the trunk or distribution feeder is kept relatively low, at or near the 8.7 db. optimum gain of Neperian theory. According to one feature of the invention, each amplifier station along the cable consists of one or more gain blocks or stages, which may be cascaded, each bypassed by a set of relay switches. Each gain block provides the full gain required for the amplifier station, such as between 8.0 and 10 db. For normal operation only one gain block (the primary one) is utilized in the RF signal path. The other (secondary) one is bypassed, for the time being, and does not contribute to the gain provided by the amplifier station.

The bypassed gain blocks provide fail-safe redundancy for the amplifier cascade. If a primary gain block fails, it can be bypassed .and a secondary gain block switched into the system under control of the computer at the head end. In this manner the amplifier station with a failed gain block maintains the same level of gain, and failure creates no change in the operation of the amplifier cascade for the trunk.

In a second fail-safe mode, a normally unused (redun¬ dant) gain block may provide fail-safe operation in the event of a catastrophic failure of a different amplifier station in the cascade. Upon such failure, where both gain blocks of an amplifier station, or even the power supply which serves both gain blocks, may fail, the entire amplifier station may be bypassed by the RF relay switches. Since the gain of an amplifier station is low, of the order of 8.7 db., the input to the following amplifier will fall only by a similar amount. The following amplifier may then be commanded to switch its secondary gain block into the RF path in series with the primary gain block. The overall gain of this amplifier station will then be twice normal and will essentially compensate for the failed amplifier station. Thus, the performance of the entire amplifier cascade will be nearly unaffected, being degraded only by a lowered input to one amplifier, which will have minimum effect and further may be compensated by the adjacent amplifier station. Preferably the bypass is arranged to be fail-safe, in that when no control signal is applied to the RF relay, the relay switch will always be in the bypass position. Thus, for example, if the amplifier station were to lose its power supply, the amplifier station would be automati- cally bypassed rather than interrupting signal flow down the amplifier cascade.

The ability to bypass amplifier gain blocks also permits a form of gain control according to the invention. Since gain blocks of relatively small gain can be readily switched in and_, out of the cascade, the overall performance of the cascade can be held within narrow limits,as during possible extreme temperature excursions and associated deviations of cable attenuation. Unlike automated gain control of conventional trunk amplifiers lere the gain control is provided at individual amplifiers throughout the cascade, the control for the amplifier cascade of the present invention is centralized at the control computer at the head end. This control computer maintains proper levels by two means.

First, there is a temperature-sensing and attenuation reference. The control computer at the head end is fed with temperature data from outdoor temperature sensors at either or both the head end and the remote hub. Alternatively, tempera- ture sensor sites other than these (such as at the cable company's local office) may also be used. The control computer then uses the temperature data to reference a look-up table to determine which amplifier blocks should remain effective by being interposed in the cascade, and which should be bypassed, to overcome temperature effects on gain.

Thus, the amplifier cascade may be designed to have amplifier spacings which, at a normal temperature of 68°F have slightly less cable attenuation than the gain of the amplifier gain block. At 68°F therefore the output level of each amplifier station will be slightly over the output level of the preceding amplifier station, which level change becomes cumulative for the amplifier cascade. The look-up table would then be preprogrammed, upon being provided with information that the temperature is 68°F, to bypass every N-th amplifier station so that the (N+l)th .amplifier station would then have a lowered output to prevent the signal level at the hub from increasing beyond desired limits. In effect, a desired reduction in overall gain is effected by distributing it over the whole amplifier cascade, bypassing enough amplifier stations to accomplish the total gain reduction desired.

If the temperature should increase, the attenuation of the cable will also normally increase. The input level of the successive amplifier stations will then progressively decrease. The control computer will reference the look-up table for the new higher temperature, which will instruct the computer to redistribute bypassed amplifiers to longer inter- vals. At the maximum temperature designed, the attenuation of the cable will be at its maximum and a minimum number or even no amplifiers will be bypassed. Stated differently, at the maximum temperature, where the attenuation of the cable is maximum, a maximum number of amplifiers are effective, and none need by bypassed. At lower temperatures, where the attenuation of the cable would be less, tending to result in higher than desired final signal levels, certain of the amplifiers may be programmed to be bypassed, so as to prevent the actual signal level from exceeding appropriate maximum limits. For example, as the outdoor temperature decreases from 50° to 30°F, the control computer may change the frequency of bypassing from perhaps every fourteenth amplifier to every twelfth amplifier. Thus, over the total length of the amplifier cascade, as the cable attenuation changes as a function of temperature, the control computer will increase or decrease the overall gain provided by the amplifier cascade on a distributed basis by bypassing the appropriate amplifiers. This will balance the total gain of the amplifier cascade against the total loss by attenuation. As a second form of automatic gain control, the signal level may be sensed at the end of the cascade, for predetermined signals (either communicated intelligence or a special control signal) . This data is relayed to the control computer via a data modem through an appropriate medium, such as a telephone line or other means. If the sampled data indicates that gain levels are not within proper limits, the control computer will be programmed to shift its reference in the look-up table to adjust the overall gain of the cascade by selection of bypassing. If the gain discrepancy between the sampled signal and the desired signal at the end of the cascade is greater than a pre-set limit, an alarm may be generated. Thus, automatic gain control is centralized in the control computer at the head end, augmented by temperature sensors and signal-level sensors at the remote hub. It will be understood that the computer controlling this system can be used to control other circuits and/or devices in the amplifier. According to one feature of the invention the amplifiers are greatly simplified to consist essentially only of one or two low-gain blocks, RF bypass switches, a modem or decoder for receiving control data, and the amplifier station power supply. As a result the amplifiers may be inexpensive in relation to conventional CATV amplifiers. This cost reduction more than compensates for increasing the number of amplifiers in the cascade in a Neperian design, where the number of amplifiers may be two to three times that of a conventional design. A further benefit of the present invention is its fail-safe feature even if a complete amplifier station fails. In a conventional design, failure of an amplifier block would completely disable the trunk, since signals downstream of the failed amplifier would be suppressed. Even if RF-controlled bypasses were used, the high gain and corresponding spacing of conventional amplifiers would cause the input to an amplifier following a failed unit to be unacceptably low and result in an undesirably low signal to noise ratio at the hub. However, in the present system, the input to an amplifier following a failed unit is changed but little, and the failure of a unit therefore has little impact on system performance. This is a consequence of the present system design where the lower gain of the stages and smaller spacing along the cable causes less of a reduction of signal level even when an entire station is bypassed. In addition, with the Neperian design, normal amplifier outputs are lower and inputs are higher than in conventional design. Thus, even with reduced input, the input level is not much lower than a conventional amplifier input without occurrence of a failure. Noise figures of the amplifiers are therefore not significantly deteriorated.

If a failure should occur in a complete amplifier station in the present system, the output of the entire amplifier cascade either falls or is shut off completely. The signal level sensor at the hub detects this and forwards its data to the control computer. The control computer then enters a diagnostic routine to locate and isolate the failed amplifier station. This is accomplished by providing each amplifier station in the cascade with a unique address. The computer may sequentially address each amplifier station and instruct the addressed amplifier to go to bypass mode. After each such addressing command, the computer receives data from the signal detector level at the end of the cascade. If the signal is reduced by 8 to 10 db. or if normal service is not restored, then the amplifier being addressed is not at fault, and may be returned to normal operation. Thereafter, the succeeding amplifiers are addressed until one is found whose bypassing does not reduce the level at the hub or which does restore the signal at the end of the cascade. This is then the failed amplifier, and its identity becomes known from its address. Appropriate correction measures are then readily supplied, either by substituting the secondary gain block for the failed unit, or if that does not succeed, then by bypassing the entire amplifier station (as to both gain blocks) and increasing the gain of the succeeding station, by utilizing both the primary and the secondary gain blocks.

In rare cases, many amplifiers may be rendered inoperative. This may be caused by the failure of a sector main (AC) power supply and/or its stand-by batteries. Upon loss of power, every amplifier affected will automatically bypass gain stages. While loss of gain in many successive amplifiers might have an adverse effect on signal quality, the amplifier bypassing will maintain path continuity. This continuity can be useful to the cable operator in determining the nature and location of a failure before repair personnel are dispatched. A cut cable would cause loss continuity. Loss of continuity indicates that the problem is not amplifier-or power-supply-related.

Another method according to the invention for bypassing a failed amplifier gain stage is an automatic method which is self-contained in the amplifier station. Detectors at the input and output of the gain stage sample the power level of the signals. If the differential between the input and output signal levels decreases by more than a predetermined amount, the gain stage is assumed to be malfunctioning and is bypassed. If both input and output signal strengths decrease by a similar amount, it is assumed that a previous amplifier, or other source, is at fault and bypassing does not occur. This method is particularly useful in open loop systems, where there is no return path from the hub to the headend to complete the control loop. In a closed loop architecture, this method would not be necessary.

It will be understood that both closed loop and open loop systems may work in conjunction with each other. A closed loop trunk system to a hub may exist as previously discussed with feedback control signals from the sensors at the hub being sent back to the control computer at the headend. At some midpoint in this trunk cascade, additional trunks to different geographic areas may split off from the main trunk through signal splitters. These additional trunks may go to other hubs where closed loop control can be employed, or they may be open loop systems.

In the case of open loop trunks or feeders, gain control for temperature variation will be processed by the central computer at the headend, according to pre-defined look- ' up tables, and restoration (bypassing) of failed amplifiers will be controlled by the above automatic, self-contained method.

The principles and advantages of the present inven- tion will be more completely understood from the following description of preferred embodiments taken in conjunction with the appended drawings, in which: Fig. 1 is a diagram of an exemplary cable system useful in explaining the performance of the present invention.

Figs. 2A, 2B, 2C and 2D illustrate typical voltage and current waveforms at the input to cable system amplifiers. Fig. 3 is a circuit diagram of a typical power-supply circuit for a cable system amplifier.

Fig. 4 is a block diagram illustrating the principles of the present invention.

Fig. 5 is a circuit diagram of one form of the present invention.

Fig. 6 is a timing diagram for the operations of the circuit of Fig. 5.

Fig. 7 is a circuit diagram of another form of the present invention. Fig. 8 is a schematic block diagram of a general cable system according to another aspect of the present invention.

Fig. 9 is a schematic diagram of an amplifier station according to the invention. Fig. 10A through 10E are flow diagram useful in explaining the operation of the present invention.

Fig. 1 indicates schematically a cable system having a conventional 60-volt, 50 or 60 hertz, power source 10

(usually supplying square-wave power) , which feeds a sequence of trunk amplifiers 11a, lib etc. over a cable or transmission line illustrated as having a section 15a between the power source 10 and the first trunk amplifier 11a, a second cable section 15b between trunk amplifier 11a and trunk amplifier lib, and a further cable section 15c beyond trunk amplifier lib for feeding additional amplifiers. Each trunk amplifier 11a, lib, etc. passes power to power-consuming devices (line extenders) on adjoining feeder lines. For example, trunk amplifier 11a may pass power to line-extender amplifiers 13a,

13b, etc. over .cable sections 17a, 17b. At the same time, amplifier 11a may provide power to another feeder line having line-extender amplifiers 18a, 18b etc. over cable sections 19a,

19b, etc. In similar fashion, trunk amplifier lib is shown illustratively as passing power to two branches, such as line- extender amplifiers 21a, 21b, etc. over cable sections 23a, 23b, etc. as well as line-extender amplifiers 25a, 25b, etc. over cable sections 27a, 27b etc. It will be understood that this branching arrangement is illustrative only. Any one trunk amplifier may pass power to none, 1, 2 or more branches, while at the same time passing power along the main trunk cable sections 15b, 15c etc. In turn, each feeder line may have any appropriate number of line-extender amplifiers, depending upon the distribution of subscribers requiring program-information signals (which are radio-frequency signals as contrasted with power-frequency signals). By way of example, trunk amplifiers 11 may be Jerrold SJ-450 amplifiers (requiring .65 amperes at 60 volts) and the line-extender amplifiers 13, 15, 21, 25, etc. may be Jerrold X-450 amplifiers (requiring .44 amperes at 60 volts or .52 amperes at 51 volts).

It will be understood that the various cable sec¬ tions, including the trunk cable sections 15 and the various branch cable sections 17, 19, 23, 25, not only transmit power to the various amplifiers but, in addition, transmit program information in the form of radio-frequency signals, to the various subscribers who are connected to taps along the cable sections (not shown) . However, ordinarily these taps require no power from the power source 10 and for present purposes are disregarded.

An important problem encountered in a system of this type is that the power source 10 has a limited current capacity (such as 15 amperes). It is, of course, desired to supply as many amplifiers (both trunk and line-extender amplifiers) from this power supply as is feasible. Another limiting factor for any one power supply is that the current drawn along the cable creates voltage drops (IR drops) which- reduce the input voltages at amplifiers which are downstream from the power supply source. These voltage drops are proportional to the current appearing on the cable at any instant, since the current draw of any amplifier is dependent upon the voltage available at its input, cable systems are very sensitive to the addition of devices which change the balance between the input voltage and current draw of attached power consuming devices". Thus, such amplifiers customarily require more current at lower input voltage than for higher input voltage. The voltage available at the input to each of the various amplifiers thus depends upon the current passing through the line sections ^' feeding the respective amplifier, and the resistance (dependant on length) of those line sections. A limiting factor is therefore when the total voltage-drop from power supply to an amplifier is such as to create an input voltage for the amplifier below its minimum requirement. The following il¬ lustration for an exemplary system will demonstrate this point. As indicated in Fig. 1 , the first trunk amplifier 11a may illustratively be at 1,000 feet from the power source 10, with a normal cable resistance of 0.77 ohms. The current in section 15a is the sum of all currents required by the entire system to the right of and includinrr trunk amplifier 11a. Illustratively, this may be 5.14 amperes, which creates a volt- drop of 3.96 volts in cable section 15a. As a result, the input voltage to trunk amplifier 11a would only be about 56 volts. The branch amplifiers 13a, 13b may together with trunk amplifier 11a draw 2.41 amperes. Between trunk amplifier 11a and trunk amplifier lib, the cable section 15b is illustra¬ tively shown as 2,000 feet in length, producing a resistance of 1.54 ohms. The current requirement for trunk amplifier lib and beyond is illustratively 2.73 amperes, which creates a voltage- drop of 4.2 volts in cable section 15b, leaving a voltage of only 51.8 volts at trunk amplifier lib.

By way of example, feeder cable 27a may have a resis- tance of 1.75 ohms, with a required current of 1.04 amperes which produces an IR drop of 1.82 volts. Thus, at line extender amplifier 25a, there would be available only about 50.0 volts. If line extender 25b is also 1000 feet further along, its current requirements of .52 amperes would produce an IR drop of .91 volts, providing only about 49.0 volts at its input in this example. Thus, the successive trunk and feeder line sections and amplifiers draw current in such a way as to reduce the initial 60 volt supply from power source 10 to substantially lower values. If such value is reduced lower than say 45 volts, the amplifier becomes impractical to operate. Accord¬ ingly, the power source 10 can supply only a limited number of trunk amplifiers and line extender amplifiers, depending upon the current drawn by each and the resistance of the cable line sections connecting these amplifiers. It will be apparent that if an additional power- consuming device, such as a frame storage unit (FSU) , were to be coupled to the power line at the location of a trunk amplifier, the added current drawn would increase the usual IR drops, and may reduce input voltages to an extent not tolerable for the operation of amplifiers there or down-stream.

For example, if an FSU were associated with each trunk amplifier, requiring an additional 90 watts, the addi¬ tional current drawn from the power supply would reduce the voltages available at the various branch points so that a branch amplifier previously having 49.1 volts may now be reduced to 43.9 volts which would not be tolerable with the existing system. Hence, the additional FSU units could not merely be added to the existing system. To get the additional power needed would require increasing the voltage of the power supply. However, most cable systems operate with a source voltage of 60 volts, the maximum allowed by the National Electrical Code. It is therefore usually impossible to increase the line voltage at a particular amplifier without adding new power injection points. Because a cable system has many more amplifiers than just described, it has become common to separate the system into power sectors, each of which provides power for the amplifiers in its range. Such power sources are placed near the centers of the power sectors so as . to provide power both upstream and downstream (with respect to the RF signal flow), the locations of the power source being determined by the nature of the cable system, and by the requirements of local utilities, since such power sources are customarily mounted on power or telephone poles of local utilities and must meet all of the basic requirements of those utility companies.

The elements of a cable system are usually designed for "power passing", for distribution of the 60-hertz energy throughout the system. Amplifiers, splitters and taps are provided with by-pass circuits to allow the 60-hertz energy needed downstream or upstream to pass without affecting the RF signal transmission. Isolating power circuits from RF is usually provided by inductors, and the current-carrying capacity of these inductors is one of the factors in determin- ' ing the extent of a power sector. Thus, power-using devices in the system have the capability to allow power to pass through them to the next power-using component of the system. The usual arrangement is to feed power at a point from which power flows in two directions, approximately half the current each way, until the power sector meets an adjoining sector. For example, a 15 ampere power source may power a sector in two directions, approximately half each way, providing seven and one-half- .ampere capacity for each direction. The most common power- passing capacity of equipment allows up to 10 amperes of current to pass on the system.

In many cases the extent of a system sector is deter¬ mined by the IR drop, which reduces the voltage by virtue of the flow of current through the resistive elements of the cable and the power-passing components. The lack of sufficient voltage to drive the powered device frequently calls for inserting a new power sector.

The nature of such a cable system makes it generally considered impractical to add additional power-consuming devices, except to the extent that available reserve capacity was designed into the original system. However, when such reserve capacity is used up or dedicated for appropriate purposes, a limit is reached for the number of power-consuming devices on the cable system. The present invention provides a way of being able to add additional loads or power-consuming devices to a customary power sector, without requiring addi- tional wiring along the system, and without reducing amplifier input voltages, as explained below.

It has been determined that the usual current and voltage characteristics in such a cable system afford an oppor- tunity to accomplish this result. This is illustrated with respect to Fig. 2A which shows in the upper part of the figure the voltage waveform at the input of a representative cable system amplifier, and in the lower part of the figure the waveform of the corresponding current. As seen, the voltage is a close simulation of a square wave, although actually a trapezoidal waveform with a flat top. This voltage waveform is provided by a power supply such as 10, in order to obtain maximum power capacity without exceeding the voltage rating of the power-consuming devices connected to the cable line. Each amplifier requires unidirectional voltages for its operation. These are provided by a rectifier at each amplifier, which rectifies the trapezoidal alternating voltage waveform to provide unidirectional and substantially constant operating voltages. In order to accomplish this, a relatively large capacitor is provided at the input of each amplifier to filter out the variations from the input alternating voltage. Fig. 3 schematically illustrates a circuit diagram for the power circuit of a representative amplifier, comprising a transformer 31 (usually with a ferrite core) coupled to the power line terminals 33 by radio frequency isolation chokes 35. The output of transformer 31 supplies a full-wave rectifier 29 having a relatively large-capacitance storage capacitor 30. The voltage across storage capacitor 30 is supplied through a voltage regulator 35 and radio-frequency filtering capacitor 37 to the output terminal 39 which may, for example, supply unidirectional operating voltage for the amplifier. The power needed for other down-stream power-consuming devices is bypassed around the amplifier circuit in conventional manner.

In customary manner, the storage capacitor 30 tends to maintain uniform the voltage supplied to the regulator 35, and hence the voltage across the load at output 39. Should the voltage output at 39 tend to decrease, the regulator draws charge from storage capacitor 30 to try to keep the output voltage uniform. In turn, the storage capacitor 30 draws current from the rectifier to replenish its charge. However, during an appreciable portion of each half-cycle the electric charge in the capacitor is sufficient to maintain a voltage higher than the input voltage, so that current is drawn from the rectifier (and hence from the power mains) only during a portion of each power half cycle. Current tends to be drawn at the peak voltage of the input which is at the latter part of the half-cycle for most CATV voltage sources.

Accordingly, during the interval shown at Y at Fig. 2A of each half cycle, there is little current, and no IR drop is created. The unidirectional voltage output at 39 is in effect maintained uniform by the voltage across storage capacitor 30. The operation of the amplifier would therefore not be affected even if it were cut off from the source of power during that interval Y. In effect, during the period Y, the IR drop caused by amplifier current is at or near zero so that the voltage at the input to this amplifier (which is passed on to the next amplifier) is not then reduced by the current drawn at this amplifier.

It has been found that little or no current is drawn by an amplifier during about 1/3 to 1/2 of each cycle. Fig. 2A shows that the current waveform drawn by a single amplifier may have only a narrow pulse of current 32. This current pulse on the cable becomes wider as more amplifiers load a given power supply. Figs. 2B, 2C and 2D show representative waveforms for different cable systems. In each instance, there is an interval (shown at Wl or W2) of zero or low current during each half-cycle, when the IR drops attributable to this current would be of low value.

If an amplifier were disconnected from the power mains during its low-current interval Y, there would be essentially no effect on its operation, since little if any current flows to it during this period. Energy needed by the amplifier circuitry is supplied from the storage capacitor 30, which will be recharged during the period of high-current supply. This affords an opportunity for the voltage source to supply current to another load during the interval Y without affecting the input voltage to the first load. The second load may in turn be cut off from the main power during the interval that the first load draws current.

In general, whether or not power is cut off for existing load devices, additional load devices may be added constrained to draws power only during intervals of each power cycle when the existing load devices do not normally draw power or are cut off from the power source.

While the RMS current or average current would be increased by the addition of such a second load, the peak current will remain the same, so there will be no increase in the maximum IR drop observed on the cable. Thus, the added load would not diminish the input voltages for the other downstream power-consuming devices in the systems.

According to one feature of the invention, called "shared time current", additional loads are placed on the cable system so as to draw current only during these intervals Wl and W2. Such additional loads therefore do not essentially affect the current drawn by pre-existing loads during periods other than Wl and W2, so that they do not alter the power or voltages applied to the pre-existing loads. Concomitantly, the power applied to the supplemental loads is not affected by the IR volt- drops created by the current drain from the pre-existing loads. This arrangement . therefore permits adding further or additional loads to a pre-existing system, without affecting the voltages available and power flow to the pre-existing loads, such as these amplifiers. Fig. 4 schematically illustrates in block diagram form one way of accomplishing this result. The power mains 41 supplies power to a transformer 43 having connected to it the customary bridge rectifier 45 and storage capacitor 47 which then applies the required unidirectional voltages for amplifiers or like circuitry, schematically designated by the block 49. An electronic power switch 51 is interposed between transformer 43 and rectifier 45, under control of a controller 53, which includes a zero-cross-over detector 55, counter timer 57, and switch driver 59.

Fig. 5 shows a circuit embodiment of Fig. 4. Three integrated circuits (IC's) IC-1, IC-2 and IC-3 are coupled to the secondary of transformer 43 in cascade fashion. Each IC is a onostable multivibrator. Transformer 43 is for isolation and to supply power to the control logic and is not to be confused with the transformer of the additional load.

The output of IC-3 controls an FET circuit 61 which serves as electronic switch 51 of Fig. 4 to cut off the input AC line 33 from the output 63.

The operation is as illustrated by the timing diagrams of Fig. 6. The waveform of the 60-hertz line voltage is shown in idealized form at 67A, being essentially a trapezo- idal waveform like 32 of Fig. 2A. IC-1 detects when that waveform crosses the zero axis, at which time a short pulse wave is produced as shown at 67B. The pulse 67B starts a pulse 67C at IC-2, whose duration T_Q is selected by adjustable resistor 64. This duration T₀ determines the delay between the zero crossing of the AC wave and the beginning of current draw of the con¬ trolled power supply, to correspond to the low-current-period Wl for the pre-existing load on the cable system.

IC-3 is supplied with the pulse 67C, whose failing edge initiates a pulse 67D. The width of pulses 67D is adjusted by adjustable resistor 66 to have a duration T^ equal to the desired on-time for the pre-existing load. The result is to modify the mains voltage waveform 67A to be as shown at 67E at the amplifier input. This supplies essentially the same power to the pre-existing load as before. However, during the intervals Wl, W2, power may be drawn from the system, without affecting the voltages at the inputs to the pre-existing load. This is done by the power switch 61 coupled between the mains line 33 and output 69. Additional loads may be coupled to the mains line 33 through a corresponding power switch which permits current flow only during intervals Wl, W2. In this way, an additional load may be switched on to the power line during periods of each cycle when the regular load is essentially not drawing instantaneous power. The loads thus may draw current alternately, permitting extra loads without causing increased IR drops which would reduce the input voltage to the regular loads. Alternatively only the addition¬ al load is switched since its presence does not materially affect the original load when the latter is receiving no or little power. Fig. 7 shows a more detailed form of the present invention. The power mains 33 (which preferably supply a square-wave voltage, which may be 60 volts, 60 hertz) are coupled through RF-suppressing inductors 35 to a center-tapped transformer 31, which supplies a full-wave bridge rectifier 29. The rectified bridge output supplies a storage capacitor 30

(illustratively of 10 microfarads capacitance) and a voltage regulator 35 (which may be an integrated circuit chip, type

7812), whose output 39 supplies the unidirectional operating voltages required for the switch circuitry of present system. This circuit therefore forms a conventional DC power supply.

One terminal of the logic-grounded center-tapped secondary of transformer 31 is supplied through a current- limiting resistor 71 and noise-bypassing capacitor 73 to an overload-protecting diode 75 (which may be a type IN4148 zener diode) to an input to monostable multivibrator 77 (formed of one half of an IC type 4538). Operating voltage for multi¬ vibrator 77 is supplied from output terminal 39. The multi¬ vibrator 77 is connected to turn "on" to produce an output when its input goes positive at or near the passage through zero of the input alternating voltage from transformer 31, after a time delay determined by RC circuit 79, 81, which determines the duration of the multivibrator output pulse. The multivibrator responds only to positive inputs and does not respond to negative inputs. The output of multivibrator 77 is passed through a zener diode 83 and input resistor 85 to a second monostable multivibrator 87, which may also be one-half of an IC type 4538. Multivibrator 87 is turned on by the trailing edge of the output pulse from multivibrator 77. Multivibrator 87 is caused to turn off after a time interval which is determined by the RC circuit 89, 91. The output of multivibrator 87 is then supplied through another zener diode 93 to the input of a metal oxide silicon field effect transistor or MOSFET 95 (which may be an International Rectifier type IRF 622), acting as a switch, which is on for the period of the pulse received from multi- vibrator 87. Switch 95 is supplied with input power (which may be 60 volts) at terminals 33 and supplies its switched output to output terminals 99.

The circuit thus far described will then produce an output shown by the pulse 101 of Fig. 6, during positive half- cycles of the input power. A similar circuit (having the same reference numerals with the suffix "a" ) is connected to the opposite terminal of the secondary of transformer 31, and provides a corresponding pulsed output shown at 103 in Fig. 6 during the negative half-cycles. As a result, during each half-cycle, there is a controlled time interval during which current may be drawn by the new load. The time intervals would be selected so as to draw power only during the times which would not cause excessive voltage drops at the inputs to the original power-consuming devices. It will thus be seen that the present invention interleaves power supplied from the voltage source to respec¬ tive power-consuming devices, on a time-division basis, operating on each cycle of the alternating current power.

In certain instances, instead of switching power at every half-cycle, ', the power supplied to the original and sup¬ plementary loads may be switched at intervals of several cycles. This would be applicable where the power- consuming device is supplied from a rectifier or other power supply circuit, having a sufficiently large . storage capacitor to maintain the operating voltages for the power-consuming device within tolerable levels over a period of several cycles. In such instances, a first load may be cut off for a period of several cycles, with the storage capacitor maintaining neces¬ sary operating voltages. During that period, power is supplied to a second similar load, and charges up its storage capacitor. At the end of the period, the reverse occurs, with the first end being powered and charging up its capacitor, while the second load is cut off. In such case, the multivibrators described above would be preceded by counting circuits, to count the predetermined number of cycles before each multi¬ vibrator is to be actuated to commence or terminate the signal determining the on-time for each power circuit. Here again, effective against the other load current drawn for one load does not cause a voltage-drop effective against the other load.

The present invention has therefore provided a way for increasing the power-carrying capacity of a cable system or the like, by limiting peak currents and the subsequent voltage drop on interconnecting cables. This improves the efficiency of power distribution through the cable system since less power is lost to lr losses on the cable.

Another aspect of the present invention concerns controlling amplifiers, gain and temperature compensation in a cable or cable system.

Referring to Fig. 8, at the head end 110 of a cable system, at the left of the diagram, there are the usual cable channel processors and modulators designated by the rectangle 111. They supply the usual intelligence signals (i.e., television and/or radio and/or computer and/or other data signals) to the cable designated at 113, through a conventional coupling device 115.

A computer 117 whose functioning is described below provides signals to a radio frequency modem 119, at a carrier frequency different from those used by the output of the rec¬ tangle 111, to the coupling device 115 for transmission over the cable 113.

Cable 113 extends to the hub 121 at an appropriate distance from the head end 110. The cable 113 is provided with a sequence of amplifier stations 123 indicated as N in number, at spacings and with gains as described below. The signals transmitted over cable 113 are provided at the hub 121 to a conventional splitter device 124, one output 125 of which proceeds to a distribution device 127 for the various intel¬ ligence signal feeds 129 in conventional manner. According to one aspect of the invention, the amplifier stations 123 are spaced apart a distance for which the normal attenuation or loss in the cable 113 is substantial¬ ly 8.7 db. (i.e., between about 8 and 10 db) . Each amplifier station provides this low gain to compensate for the ap- proximately equal attenuation of a section of cable immediately preceding it (i.e., between it and the preceding amplifier station) . This arrangement provides a cable system which overall has essentially no loss, and has minimum noise and distortion. To maintain this condition reliably, notwithstanding possible failure of individual amplifiers, a special amplifier arrangement shown schematically in Fig. 9 forms part of this invention. As seen in Fig. 9, the various signals coming over the cable 113 are supplied to a signal splitter 131. The cable 13 serves as the transmission medium not only for the intel¬ ligence signals, and control signals, but also for power current. For enhanced simplification of circuitry to reduce the number of components and thereby increase the reliability (i.e., mean time between failures), power is preferably transmitted over the cable system as direct current.

Thus, the splitter 131 serves to direct onto line 133 the power current which is fed to a power supply circuit 135 through an inductor or other high-frequency-suppressing element 137. Power supply 135 supplies required power to the two stages or gain blocks 123A and 123B of the amplifier station 123 of Fig. 9, by way of leads 139. Power supply 135 also forwards the power current onward to the next section of cable 113 through lead 141 and high-frequency-suppressor 143.

Splitter 131 also aids in deriving from the cable 113 the control signals for the particular amplifier station 123 illustrated. Thus, it would be understood that each of the amplifier stations 123 has an individual "address", which is incorporated into the control signals supplied from computer 117 through modem 119 to the cable 113. Splitter 131 serves by conventional means to separate these control signals and supply them to lead 145 and then to a modem-controller decoder 147. The function of controller 147 is to derive from the trans¬ mitted control signals those which individual to the particular amplifier station 123, and produce from them those control signals needed to determine the operation of the particular amplifier station 123. Amplifier station 123 is formed of a multiplicity of stages or gain blocks, illustrated by way of example as two in number, namely stages 123A and 123B. A switching arrangement formed by switches SW1, SW2, SW3 and SW4 determines which if any of the amplifier stages of the amplifier are effective. Splitter 131 derives from the cable input 113 the intelligence signals plus the totality of control signals, whether or not individual to the specific amplifier station. These signals are supplied over lead 149 to amplifier stage 123A when SW1 is in the up position. With SW2 also in the up position, these signals (after amplification by stage 123A) are supplied either to stage 123B or directly to the output of cable 113 through switch SW4 (when "down"). The arrangement of switches is such so that either stage 123A nor 123B or both stages 123A or 123B or neither stage 123A or 123B, may be interposed in the path of the intelligence plus control signals on lead 149, as shown by the following table.

Effective Stages Stage A only

(stage B bypassed)

Stage B only (stage A bypassed)

Stages A&B

(no bypass)

No stages (full by-pass) dn dn dn dn It is thus clear that switches SW1 and SW2 essential¬ ly put stage 123A in the cable circuit or out of it, while switches SW3 and SW4 essentially put stage 123B in or out of the cable circuit. Switches SWl and SW2 may therefore be ganged or operated simultaneously (as by being double-pole switches) , and the same may be done with switches SW3 and SW4.

Control of the switches SWl to SW4 is determined by controller 147. It will be understood that the switches and controller may be conventionally operated, and are preferably constituted by an integrated circuit or chip designed in a conventional manner to produce the functioning just described. Alternatively, conventional relay switches may be used. Preferably, the switch elements and controller are passive in nature, but if a power source is required, it may be derived from the power supply 135.

To provide a form of automatic gain control, the signals arriving at the hub 121 as seen in Fig. 8 are supplied through a splitter 124 to an automated field strength or signal level meter 151 to provide a signal on the output lead 153 indicative of the amplitude level of the intelligence signals on cable 113 arriving at the hub 121. The signal on lead 153 may be an average or mean of the intelligence signals, or a separate control signal dedicated for gain control. The computer control 155 cooperating with an RF or telephone modem 157 forwards these amplitude-level data back to the head end over a separate communication channel 159. Those signals are received at an RF or telephone modem 161 at the head end 110 and converted into control signals on lead 163 to cause computer 117 to forward appropriate control signals to the N amplifier stages 123-1 to 123-N along the cable to restore the amplitude level at hub 121 to the predetermined desired level.

Thus, according to' one aspect of the present inven¬ tion, the signal level at hub 121 is continually monitored. If the signal level sensed is lower than the predetermined desired level, this is an indication that one or more of the amplifier stages may have become disabled. This in turn triggers an interrogation cycle created by computer 117. The sampling or interrogation of the various amplifier stations is accomplished in known manner. One procedure is to assign a unique digital address to each amplifier station. When interrogating that station, a signal 113 sent by conventional means from the control computer over the cable system, which signal contains the digital address plus digital control data for instructing the addressed station to take the action desired (e.g., as to bypassing and/or substituting a stand by amplifier stage). The controller 147 (Fig. 9) at the amplifier station will respond only to its unique address data, and will then provide the appropriate switching in SWl, SW2, SW3 and/or SW4 called for by the control data. Presuming, for example, that the initial state of the system is that all stages A are switched into the system and all stages B are bypassed, under this interrogation cycle, each amplifier station 123 is successively addressed and its switches are activated to bypass its stage A.

As each successive amplifier stage A of the N amplifier stations 123 is bypassed, the amplitude level at hub 121 is sensed for any change. The bypassing of a stage A, if that stage is properly operative, would reduce the overall gain between the head end and the hub by approximately 8 db. If this gain reduction is sensed, the computer is instructed to terminate the bypass for that stage A, and to proceed to bypass the stage A of the next amplifier. This procedure is repeated until, upon bypassing a stage A, no change in output level at the hub is detected. This indicates that stage A is defective. Upon sensing that situation, the computer instructs the corresponding stage A to remain bypassed, and unbypasses its counterpart stage B, which would then add the required gain increment to restore the level at hub 121 to the desired value. When this is done, the interrogation cycle is terminated.

Similarly, if the signal at hub 121 completely fails, the computer again bypasses successive stages A until a signal is detected at hub 121. At that point, an open circuit stage A has become bypassed. To restore the gain level, a correspond¬ ing stage B is then unbypassed, cutting out the defective stage A from the system entirely. However, if both stage A and stage B of the same amplifier become inoperative, bypassing stage A followed by unbypassing stage B of the same amplifier will not produce a desired increase in signal level at hub 121. Under this cir- cumstance, stage A remains bypassed and stage B is restored to its bypass condition. Then stage B of the succeeding amplifier in the amplifier cascade is unbypassed to add its gain to its stage A, and thereby replace the gain not provided by the preceding defective stage B. It will be understood that the bypassing of stages A may thus be controlled independently of the bypassing of stages B, and both are controlled to maintain the desired signal level at hub 121.

It will also be understood that the gain of stage B may have a value less than the gain of stage A. For example, the gain of stage B might be half of the gain of stage A, or 4 to 5 dB. This will improve the resolution of gain control over the system to maintain a tighter tolerance of levels during temperature variations. In case of a failure of a stage A gain block, two stage-B gain blocks, one from a preceding amplifier station and one from a following amplifier station, may be substituted. Temperature Compensation

The present system also provides an automatic temperature compensation for the cable system. As is known, the cable attenuation or loss varies with temperature. As ^■temperature rises, so does the cable loss, and the input signal levels at successive amplifiers would then decrease, leading to undesirable noise. This is avoided by the present invention. An outdoor temperature sensor 165 is provided at the hub 121 to provide a signal to computer control 155 indicative of change (either increase or decrease) of temperature from a predetermined "normal" value. From established data it can be determined what the effect on attenuation will be for the change in temperature experienced by a cable system of its specific length. Therefore, by forwarding the temperature signal through a computer control 155, modem 157, line 159 to the head end 110, a determination can be made as to the needed change in gain to overcome the temperature-created change in attenuation. In one form of the invention, for each 8 to 10 db of gain increase needed, a previously bypassed stage A or stage B of one amplifier 123 may unbypassed to add its gain to the amplifier cascade. Similarly, for each 8 to 10 db of signal level increase occasioned by a temperature change, computer 117 may send out a signal addressed to an amplifier stage A, to bypass it, and thereby reduce the gain of the amplifier cascade.

Computer 117 may be provided with a look-up table, which for each temperature indicates which and how many gain blocks are to be effective or are to be bypassed. For example, at the normal temperature of 68°F, the gain of each block or stage may be designed to be slightly more than the cable attenuation between successive amplifier stations. The look-up table would then be pre-programmed so that at 68°F it would order bypassing selected amplifiers to prevent the level at the amplifier station succeeding the bypassed one from increasing beyond desired limits.

If the temperature increases, the computer will then reference the look-up table for the new higher temperature, and the look-up table will instruct the computer to redistribute bypassed amplifiers to have a longer interval between bypassed gain blocks. For example, at the maximum design temperature (e.g. 125°F) the attenuation of the cable will be at its maximum and no amplifiers need be bypassed.

When the temperature decreases, the cable attenuation decreases and the computer will then order more amplifier stations to be bypassed. For example, as the temperature decreases from 50°F to 30°F, the control computer may change the frequency of amplifier bypassing from perhaps every 14th amplifier to every 12th amplifier over the total length of the amplifier cascade. As the cable attenuation of the trunk cable changes as a function of the temperature, the control computer will thus increase or decrease the overall gain on a dis¬ tributed basis by bypassing the appropriate amplifier stations. This will balance the total gain of the amplifier cascade against losses in the cable.

Figs. 10A to 10E show flow or function diagrams illustrating the functioning of a preferred embodiment of the present system. Referring to Fig. 10A, at the starting point 101, the computer 117 determines at decision point 103 whether the signal level detector 151 at the hub 121 detects a signal level within the required limits, illustratively within 1-4 db of the desired value. If "yes", the system returns to the start point 101 and makes another determination. It repeats this loop at a desired periodicity such as 4 times per minute, and thus continually monitors the signal level of the system.

If a fault occurs in some amplifier station, the detector 151 will determine that the signal level is not within the appropriate limits, leading to a "no" result from decision block 103. The computer then determines whether there is any signal as indicated by decision block 105. If "yes", then as a first stage a determination is made as to whether the change in signal level which caused it to be outside the desired limits may have been caused by temperature change. Thus, the tempera¬ ture is detected (either at the hub or head-end or elsewhere), as indicated by block 107. Since cable attenuation is greatest for maximum temperature, the cable system may be designed for amplifier station spacing along the cable to provide gain compensating for the attenuation at such maximum temperature. For lower temperatures, the attenuation is less, requiring less amplifier gain to retain proper amplifier input levels and hub signal level. For some cables, say 5 miles long, the attenua¬ tion may change 30 db. for a temperature change of 120°. According to one form of the present invention, to determine which stage or stages need be changed to compensate for temperature changes, the computer 117 stores a table listing, for the particular cable system, which amplifier stations along the cable are to be effective for each tempera- ture experienced (e.g., for temperatures between -40°F. to +120°F. ) . This can be predetermined from known attenuation temperature characteristics of cables. Then as indicated at block 111' appropriate amplifiers are activated or deactivated to create the proper number and location of operative amplifier stages (i.e. stages A) in the cascade. This is done by successively addressing the amplifier stations and providing for each the control signals to cause its stage A to be bypassed or not, as the stored data dictates.

After the temperature correction (or confirmation) has been made in this manner, by assuring that the appropriate amplifier stations are bypassed, a further determination is made as indicated by block 113, as to whether the signal level detected is now within the required limits. If this is the case, then the "yes" result leads to terminal A, back to the starting point 101 (Fig. 10A) , resuming the standby monitoring of signal level indicated at block 103. Thus, the system automatically compensates for temperature variations by activating the proper number and/or locations of active amplifier stages. Accordingly, the system will normally have all stages A unbypassed (except those required to be bypassed for temperature compensation) and all stages B bypassed, as standby gain blocks.

It will be understood that this temperature compensa¬ tion arrangement be used separately from the other aspects of the present invention. Where the temperature compensation arrangement, is used by itself, no return channel from the hub (e.g., 159) is.necessary since the temperature may be sensed at the head end. Gain Control - Reduced Siσnal Output

However, if, after temperature compensation is ac¬ complished, the detected signal is still not within the required limits, a sequence is begun to determine which amplifier may be the cause of improper signal level and to restore the proper level. There may be a complete loss of signal or a partial reduction in signal level. If the latter, the "no" result of block 113' (Fig. 10A) leads to terminal C of Figure 10A, and thence to terminal C of Figure 10C, leading to block 115' indicating that the system will then address in turn each amplifier station (a particular amplifier being designated as No. N, where N may be a value from 1 to P, the total number of stations). A determination is made as shown by block 117' as to whether this amplifier N (i.e., its stage A) is normally bypassed; if so, it is disregarded and N is incremented to the next value as shown by block 119'. It is then determined at block 121' whether the new N equals P, the maximum number of amplifiers. If not, the sequence is repeated for successive amplifiers until it is determined at block 117' that the amplifier under examination is not normally bypassed. In such event, as indicated by the "no" result at block 117', a command is sent by the computer 117 to bypass that amplifier N, per block 123'. Then it is determined, as shown in block 125', whether the output signal level at the hub 121 dropped by the amount of customary gain of an amplifier stage, illustratively 8 to 10 db. If this does occur ("yes" at 125'), it indicates that the amplifier under view is not defective, and accordingly the bypass is lifted (at 126') and N is incremented (at 119') to search for and examine the next unbypassed amplifier. This sequence is repeated until an amplifier station is reached for which it is determined ("no" at block 125') that the detected signal level did not drop by the specified decrement. This indicates that there was no effect pon bypassing a normally unbypassed stage A, from which it is concluded that the stage A is defective. In that case, that amplifier stage A is left bypassed (block 127') and stage B is activated (unbypassed) as indicated at 171, Fig. 10E. If that results in proper increase in signal level (per "yes"_r block 173) then a trouble report or alarm is generated to show a defective stage A at that amplifier station, and the system (now with proper signal level) returns to its monitoring and examining mode by in¬ crementing N (per terminal F, Fig. 10C and block 119'), to inspect successively the following amplifiers in the amplifier cascade. Thus, stage B is substituted for defective stage A to restore the required signal level. However, if unbypassing stage B did not increase signal level ("no" at 173, Fig. 10E) then stage B is bypassed (per 175) so that the entire amplifier station is bypassed) The stage B of the next amplifier station is interrogated (at 177). This may be at the next amplifier station in the amplifier cascade, or the next station with an unbypased stage A, or a station assigned by the computer from its look-up table. If this restores the signal level ("yes" at 179) then the system proceeds with interrogating successive amplifier stations (by terminal F to Fig. IOC, to 119'). This action cuts out amplifier station N, and makes effective both stages A and B of a subsequent unbypassed station. Accordingly, stage B of a subsequent amplifier station is substituted for a defec¬ tive stage A where the latter's corresponding stage B is also defective.

However, if no signal level increase is experienced by unbypassing of that stage B ("no" at 177) the last cycle is repeated until an amplifier stage B is reached which is effec¬ tive, in which case the monitoring cycle described above resumes.

In summary, normally bypassed amplifiers are in effect skipped in scanning all of the amplifiers of the cascade. Any other amplifier which when bypassed created a signal level drop, is also skipped. But if bypassing does not create a signal drop of the required magnitude, the bypassing of stage A is maintained, and a stage B is inserted in the amplifier cascade. This stage B may be at the same station as a defective stage A, or at a subsequent station. The entire sequence of amplifier stations in the cascade is thus checked out until a correction is made or all have been tested. At that point, as shown in block 121' (Fig. IOC), the system reverts back to point A of Figure 10A and resume monitoring all stations.

Gain Control - No Signal Output

However, if it is determined at block 105' (Fig. 10A) that the detector has no signal, the system proceeds to point B of Figure 10B to address an amplifier N as shown by block 131'. If the amplifier is normally bypassed ("yes" at block 133') it is ignored; N is incremented and the cycle repeated for all of the amplifiers in the cascade or until an amplifier is examined which is not normally bypassed (which may be determined from the look-up table). In the latter case ("no" at 132'), the amplifier stage A is bypassed (at block 135'), and it is deter¬ mined whether the signal has been restored (at block 137'). If the signal has not been restored ("no" at 137'), then N is incremented at 139' and if the maximum number of amplifiers has not been examined, then as shown in block 141' the cycle is repeated. It will be understood that in Fig. 10B before incrementing N at block 139' the bypass imposed on the amplifier stage A under examination is removed (per 138'), so that the amplifier N is restored to its previous condition.

If during the cycle of testing in Fig. 10B, the signal is restored ("yes" at 137'), then that amplifier is left bypassed, as indicated at block 143' . A trouble report is generated as indicated at 145' to alert the maintenance staff that the defective stage A being viewed needs attention.

However, at this point a normally unbypassed stage has been bypassed, so that its gain is not available. This would result in a lower output signal level than desired. To overcome this, the action proceeds to point D (Fig. 10D) to activate stage B of the same station, by inserting it into the amplifier cascade, by appropriate actuation of the switches SW3 and SW4. Whether the output signal level has increased upon inserting stage B is tested at 153'. If "yes", the trouble report is generated (that stage A is defective) and the system returns to standby monitoring at A, Fig. 10A.

If the insertion of stage B does not increase signal level ("no" at 153'), then it is concluded that both stage A and B of this amplifier station are defective. The system then bypasses the stage B just tested (at 159') and activates (at 157') stage B of a subsequent amplifier station N + i, (where i may be 1, signifying the next station in the cascade, or a different value predetermined by the computer). If this in¬ creases the signal level to the desired value ("yes" at 155'), the system may provide a trouble report (at 145') to show the need to correct the defective stages A and B and then returns to the monitoring cycle. If the signal level does not increase at 135 ' , the system returns to block 157', to activate the next (or another predetermined) stage B to provide the necessary gain to make up the gain deficiency cause by the defective stages A and B. When all amplifier stations have been interrogated and examined, without curing the signal level deficiency ("yes" at 141') an alarm may be generated as at 147'. Summary

When a failure occurs in an amplifier station, the output of the amplifier cascade either falls or is shut off completely. This is detected by the automatic level sensor at the hub, to initiate the diagnostic routine at the control computer 117 to locate and isolate the failed gain block. By having each amplifier in the cascade with a unique address, a computer may sequentially address each amplifier and instruct the amplifier to go to its bypass mode. After each such command, the computer references new data from the level sensor at the end of the cascade. It will be understood that an appropriate computer program can be readily provided by conventional procedures to accomplish these described func¬ tions, or an appropriate microchip may be set up to do so. Alternatively, digital circuitry can be designed in a conven¬ tional way to carry out the functions just indicated.

Thus, the present system provides gain control for the amplifier cascade. If the primary gain block fails, it may be bypassed and a secondary gain block switched into the circuit by the control computer at the head end. In this manner, the amplifier station maintains essentially the same level of gain and there is no change in the operation of the trunk amplifier cascade.

Unlike automatic gain control for conventional trunk amplifiers, where the gain control is provided at each amplifier and is thus distributed throughout the amplifier cascade, the gain control for the present cable amplifier cascade is centralized at the control computer at the head end, resulting in lower cost for the amplifiers. Also the amplifiers are more simplified than conven¬ tional cable amplifier, consisting only of two low-gain blocks, the bypass switches, a modem for receiving control data, and an amplifier station power supply. This reduces costs over the conventional high-gain, more widely-spaced trunk amplifiers., and more than offsets the cost of a greater number of amplifiers than in a conventional design.

Moreover, the present invention provides fail-safe operation in the event of failure of a complete amplifier station in the cascade. In the case where both gain blocks (or the power supply which serves both gain blocks) fail, the entire amplifier station is bypassed. Since the gain of an

^• amplifier station is low, only about 8 to 10 db, the input to the following amplifier will fall only by a similar amount. A subsequent amplifier in the amplifier cascade is then commanded to switch its secondary gain block into the signal path in series with the primary gain block. The overall gain of that amplifier station will then be twice normal, and will compen¬ sate for the preceding failed amplifier. Preferably, the bypass switches are also arranged in a fail-safe configuration so that, in the absence of a control voltage, the switches remain in the bypass position. Thus, if an amplifier were to lose power, the amplifier station will be automatically placed in full bypass. The fail-safe feature is an important benefit. In a conventional design, a failure of a gain block would completely disable the cable or trunk. Even if bypass switches were used, the conventional high gain and corresponding spacing of the amplifier stations would cause the input to an amplifier following a failed unit to become unacceptably low, resulting in distortion. However, in the present system, the input to an amplifier following a failed unit is changed but little, and failure has little impact on system performance, even before a failed amplifier stage is bypassed and replaced by a secondary stage. This is caused by the use of lower-gain stages and spacing, which produces less of a reduction when a station fails and is bypassed. While it is preferred to bypass failed -.amplifier stages and to place a secondary replacement stage essentially in series with a bypassed failed stage, in some circumstances it is possible to open-circuit a failed stage and replace it with a secondary stage essentially in parallel with the open- circuited failed stage.

The present invention also provides the advantage of improved economy and lower cost. According to a recent cost analysis, the cost of a 5-mile cable system according to the present invention (including both fixed and per-mile costs) would be less than one-half the cost of a conventional cable systems, and with improved reliability.

It will be understood that obvious variations from the illustrative examples described above may be readily made without departing from the spirit of the invention, whose scope is defined by the appended claims.

Claims

WHAT IS CLAIMED AS THE INVENTION IS: 1. In a cable system or the like powered from an AC power source for energizing loads producing voltage-drops in said system, the method of avoiding increasing the voltage-drop experienced by a first load due to the addition of a second load to said system, including the steps of a) interrupting supply of power to said second load during intervals when said first load requires current, and b) supplying to said second load power having substantially no effect on voltage-drops experienced by said first load. 2. In a cable system or the like powered from an AC power source for energizing loads producing voltage-drops in said system, the method of avoiding increasing the voltage-drop experienced by a first load due to the addition of a second load to said system, including the steps of a) interrupting supply of power to said first load during intervals when said first load draws zero or near- zero current, and b) supplying said second load with power only during said intervals, whereby current drawn by said second load does not contribute to voltage-drops experienced by said first load. 3. The method of claim 2 wherein said interruptions occur at each half-cycle of said power. 4. The method of claim 2 including the steps of a) detecting the zero crossovers for the power input to said first load; b) producing a first signal in response to each said crossover; c) adjusting the length of said first signal to correspond to an interval of near-zero current; d) producing a second signal responsive to the termination of said first signal; e) adjusting the length of said second signal to correspond to an interval of substantial current to said first load; and f) suppressing input to said first load except during the interval of said second signal. 5. The method of claim 2 including the steps of a) detecting the zero cross-over for the power input to said second load; b) producing a first signal in response to said cross-over; c) adjusting the duration of said first signal to correspond to an interval of near-zero current; d) producing a second signal responsive to said first signal; e) adjusting the duration of said second signal to correspond to an interval of substantial current to said first load; and f) supplying power to said second load respon- sive to said second signal, and only during intervals other than said interval of substantial current to said first load. 6. The method of claim 2 comprising the steps of a) initiating a first signal in response to a zero cross-over of said power to said first load; b) terminating said signal at a selectable time interval after said initiation, and c) supplying said second load with power except during the duration of said signal; whereby the voltage-drop at the input of said first load is not substantially affected by the power supplied to said second load. 7. The method as in claim 6 further comprising selectably adjusting the time interval between said zero cross- over and initiation of said signal to correspond to a period of low power to said first load. 8. The method as in claim 7 a) wherein said first signal imitating responds only to positive half-cycles of said AC power source, and further comprising b) initiating a second signal in response to a negative-going zero cross-over of said first load power, c) terminating said second signal at a selec- table time interval after its initiation, and d) cutting off power to said second load during said time intervals. 9. In a cable system or the like powered from an AC power source for energizing loads producing voltage-drops in said system, the method of increasing the power-supplying capacity of said system comprising a) determining time intervals when a first load requires low current, b) supplying a second load only during such intervals, whereby the voltage-drop experienced by said first load is substantially unaffected by said second load. 10. A method as in claims 9 wherein s id intervals are determined in each half-cycle of said AC power. 11. In a cable system comprising an alternating power source and a plurality of loads coupled to said system at individual distances from said source, whereby current drawn by said loads causes voltage-drops between said source and each said load, the combination comprising: a) a switch at one said load for permitting power flow to said one load during intervals substantially less than one half cycle, said interval being arbitrarily set for the first load. b) a second load supplied by the same voltage source, and c) a switch at said second load for permitting power flow to said second load only during intervals when power flow to the first load is cut off, whereby neither load contributes to voltage- drops experienced by the other load. 12. A system as in claim 11 wherein said power cutoff is caused to occur at or adjacent to the current zero crossover point at each half-cycle. 13. A system as in claim' 11, including a) a zero-crossover detector responsive to input power, and wherein b) said electronic switch at the input of said one load is responsive to said detector for cutting off said power flow to said one load. 14. A system as in claim 13 including a) a circuit responsive to said detector for producing an adjustable-length signal, b) said switch at the input of said one load being responsive to said circuit. 15. In a cable system comprising an alternating power source and a plurality of power-consuming devices coupled to said system at individual distances from said source, whereby current drawn by said devices causes voltage-drops between said source and each said device, the combination comprising a) a first power-consuming device; b) a second power-consuming device; c) a switch at said second device for permitt- ing power flow to said second device only during intervals when power flow to the first load is low whereby said second device does not contribute substantially to voltage-drops experienced by said first device. 16. A system as in claim 15 wherein said two power- consuming devices are at the same location. 17. A system as in claim 15 including means for causing power flow to said second load only during intervals at which current to said first load is at or adjacent to zero value at each half-cycle. 18. A system as in claim 17 including a) a zero cross-over detector responsive to power input to said first device and b) an electronic switch at the input of said second device and responsive to said detector for permitting power flow to said second device only at intervals of each half-cycle at or adjacent to said zero cross-over points. 19. A system as in claim 18 including means for adjusting the initiation of said power flow within each half- cycle. 20. A system as in claim 15 including means for shutting off power to said first device during a portion of each half-cycle of power flow. 21. In a cable system or the like supplying both program signals and power to a first power-consuming device and including a power supply for producing unidirectional operating voltages for said device, where said power supply has a storage capacitor capable of maintaining said operating voltages at an operable level for a predetermined period of time, the method comprising the steps of; interrupting power flow to said power supply circuit for a time interval shorter than said predetermined time, applying an additional power-consuming device to said cable, and supplying power to said additional power- consuming device only during the said interval, whereby said operating voltages are continuously provided while at the same time permitting power to be supplied to an additional power-consuming device without incurring additional voltage-drops effective with respect to said first power-consuming device. 22. In a cable system having a signal input point and a remote output signal point, said two points being joined by a cable, said cable having processing stations along its length, each processing station modifying the signal present on said cable at its location, said stations including, for example, amplifie- rs, filters, power sources an switches, the improvement comprising means for remotely controlling the status of each station from a central point. 23. A cable system as in claim 22, comprising a central computer, to a transponder at the remote output signal point, and means algorithmicly determining the status of each station based upon data from said transponder. 24. In a cable system having a signal input point and a remote output signal point, said two points being joined by a cable, said cable having amplifier stations spaced along its length, each amplifier station having a normal gain ap- proximately equal to the attenuation of the cable section between said amplifier station and the preceding amplifier station, the improvement comprising each amplifier station having a pair of ampli- fier stages, the amplifier stages at each station having selectively a configuration permitting either or both or neither of said stages to contribute to the amplification of cable signals at said station. 25. A cable system as in claim 24, including means determining the configuration of at least certain of said amplifier stations by selectively bypassing either, or both stages thereof, means at said output point for determining the level of signals arriving thereat over said cable, and means responsive to a change in said signal level for changing the configuration of the amplifier stages of at least one selected amplifier station. 26. A cable system as in claim 24 wherein each amplifier station has a normally operative stage and a standby stage, said system also including means successively bypassing the normally opera- tive stage of each of said stations one at a time and concur- rently determining whether the output signal level has changed. 27. A cable system as in claim 26 further including means responsive to a change in said output signal level for discontinuing said successive bypassing and leaving the last normally operative amplifier stage bypassed. 28. A cable system as in claim 27 further including means for substituting a standby stage for the last bypassed normally operative stage. 29. A cable system as in claim 26 further including means responsive to failure of an amplifier station for bypassing both stages of said failed station. 30. A cable system as in claim 29 further including means for increasing the gain of an amplifier station subse- quent to said failed station, whereby the output signal level is restored to substantially the signal level value before said failure. 31. In a cable system for passing signals from an input end to a remote output end by means of a cable connecting said two ends, said cable having amplifier stations spaced there along at intervals, the cable section from each amplifier station to its preceding amplifier station having an attenua- tion approximately equal to the gain of said each amplifier station, the method comprising sensing the ambient temperature, and providing a number of amplifier stations opera- tive along said cable in correspondence with the temperature sensed. 32. A method as in claim 31 wherein said sensing is accomplished at said remote end. 33. A method as in claim 32 wherein said changing is accomplished by controlling said stations form said input end. 34. A method as in claim 31 wherein said providing is accomplished by controlling said stations from said input end. 35. A method as in claim 34 wherein said providing is accomplished by bypassing amplifier stations not desired to be operative. 36. A method as in claim 35 comprising unbypassing previously bypassed stations to provide additional operative stations. 37. The method as in claim 31 including the steps of determining which amplifier stations should be inoperative at each temperature sensed, and controlling said amplifier stations from said input end in correspondence with said sensed temperature to bypass such indicated inoperative amplifier stations. 38. The method as in claim 37 including providing such spacings for the cable attenuation expected at the highest ambient temperature designed. 39. The method as in claim 37 including increasing the number of operative amplifiers in response to increase in ambient temperature. 40. The method as in claim 37 including decreasing the number of operative amplifiers in response to decrease in ambient temperature. 41. The method as in claim 37, wherein said cable between each amplifier and its preceding amplifier has an attenuation of approximately 8 to 10 decibels, and the gain of each amplifier is approximately 8 to 10 decibels. 42. In a cable system for passing signals from an input end to a remote output end by means of a cable connecting said two ends, said cable having amplifier stations spaced there along at intervals, the cable section from each amplifier to its preceding amplifier station having an attenuation approximately equal to the gain of said each amplifier station, the method comprising monitoring said system by successively changing a gain characteristic at each station one at a time, sensing the output signal level at said remote end in response to each said change, and determining the change in said output level in response to each said change to indicate the status of the station changed. 43. A method as in claim 42 wherein said successive changing is controlled from said input end. 44. A method as in claim 42 comprising successively bypassing each said station. 45. The method as in claim 42 comprising determining whether the output signal level has been changed by an amount commensurate with the gain of an amplifier stage, and bypassing the amplifier stage causing such change. 46. In a cable system as in claim 42, wherein each amplifier station has a normally operative stage and a redun- dant stage, the method comprising determining changes in signal level at said output end successively changing each of the amplifier stations with respect to a characteristic indicative of loss of gain, determining in response to such change whether said signal level is changed beyond a predetermined value, thereby indicating by an absence of such value change whether a ^• change stage is defective. 47. A method as in claim 46, comprising interrogating each amplifier station by bypass- ing the operative stage thereof, and at each such bypassing determining whether the output signal level has changed by an amount commensurate with the gain of an amplifier station. 48. A method as in claim 47, comprising causing a standby amplifier stage to be sub- stituted for the normally operative stage in response to absence of such a change in signal level. 49. In a cable system for passing signals from an input end to a remote output end by means of a cable connecting said two ends, said cable having amplifier stations spaced there along at intervals, the cable section from each amplifier to its preceding amplifier station having an attenuation approximately equal to the gain of said each amplifier station, the method comprising successively bypassing each normally operative amplifier stage one at a time, and sensing an output signal level change from a predetermined value at each bypassing, terminating said successive bypassing upon restoring said signal level to within a predetermined dif- ference from said value. 50. A method as in claim 49 including indicating failure to restore said signal level to within said predeter- mined difference from said value.