Electricity: measuring and testing – Fault detecting in electric circuits and of electric components – For fault location
Reexamination Certificate
2000-04-07
2003-01-21
Le, N. (Department: 2858)
Electricity: measuring and testing
Fault detecting in electric circuits and of electric components
For fault location
C324S535000
Reexamination Certificate
active
06509740
ABSTRACT:
BACKGROUND OF THE INVENTION
The invention relates to the diagnosis and location of faults or other conditions of interest in electrical conductors such as electrical wiring or cables. More particularly, the invention provides methods and apparatus for time domain reflectometry according to which a fault or another region of interest may be located in an electrical conductor.
A time domain reflectometer (TDR) is a device (see
FIG. 1
) that generates an initial electrical signal, couples that signal to a pair of conductors through a drive impedance ZD, and observes the electrical waveform at that coupling point. That observed waveform is comprised of an amount of the initial signal and an amount of each signal, if any, that is reflected back from any non-uniformity in the impedance of the conductor pair. In this discussion, that observed electrical waveform is called the “TDR waveform”. The relative amplitudes of the initial signal and the reflected signals are dependent on the relative impedances at each of the signal origination points and the drive impedance. V
2
in
FIG. 2
represents the voltage amplitude of the initial electrical signal. V
1
in
FIG. 2
represents the voltage amplitude in the TDR waveform between the time of the start of the initial electrical signal and the time of the reflection from the far end of the cable under test. This voltage is calculated by the formula V
1
=V
2
*(ZC/(ZC+ZD)). The elapsed time (T) between the observation of the initial signal and the observation of the reflections is dependent on the physical distance (D) between those signal points and the velocity of propagation (VOP) for the conductor pair. This VOP is usually between ½ and 1 times the speed of light in a vacuum. Since the signal travels from the coupling point to the reflection point and then back, the elapsed time is equal to twice the physical distance divided by the velocity of propagation: T=2*D/VOP. In this discussion, T (=2*D/VOP) is called the “electrical length”. One of the common uses of the TDR technique is to determine the distance between the point of connection and a fault in a cable. The fault (shorted wires, broken wire, electrical leakage due to moisture, crushed dielectric, etc.) usually manifests itself as a non-uniformity in the impedance at the point of the fault and is thus observed as a reflected signal. The distance is calculated by the formula D=VOP*T/2. The time is measured from the start of the initial electrical signal (T
1
) to the start of the change in voltage from V
1
to V
2
, as shown in FIG.
2
. This start in the change of the voltage is referred to as the “Knee”.
These TDR measurements require at least two conductive parts in the measurement path. One is the conductor being measured, and the other represents the conductive environment around the first conductor. This environment can be the second wire in a twisted pair, the shield around a center conductor of a coaxial cable, the multitude of wires in a wire or cable bundle, the metal conduit in which a single wire or multiple wires are located, or the dirt and the earth in which a wire is buried. The reference in this discussion to “wire pair” or “cable” means any wire and its conductive environment. The initial electrical signal used in TDR measurements can be of any shape, but the common shapes used are a narrow pulse or a steep step.
When the cable is relatively noise free, and consists of just one cable with no splices or impedance variations such as those caused by crushed dielectric, the TDR waveform is easily measured (
FIG. 2
) and interpreted. The major element of this waveform is the reflection from the end of the cable and the electrical length to that end can easily be measured with common TDR timing techniques. However, if the cable under test is actually constructed of two or more cables with one or more splices, the TDR waveform will contain elements of those splices and impedance variations due to cable differences, and will be difficult to interpret. Some elements of the waveform will be indicative of normal conditions in the cable and its installation, and some will be indicative of a cable or connector failure. For example, a normal household AC wiring situation could contain 12 gauge wire from the circuit breaker to a switch, and then only 16 gauge wire from the switch to an overhead light fixture (FIG.
3
). The perturbations in the TDR waveform (
FIG. 4
) caused by the change in wire size, the spread of the individual conductors at the switch, and the switch itself, are significant in size when compared to actual failures in such a cable. These perturbations could, in the absence of adequate waveform analysis, be mistaken for flaws or failures in the cable. An additional example can be found in cable TV systems, where 75 ohm coaxial cables are joined together with small coaxial connectors (FIG.
5
). The impedance variations in the cables and connectors cause perturbations in the TDR waveform (
FIG. 6
) that reach 25% of the normal signal amplitude. Such variations are, in the absence of adequate waveform analysis, indistinguishable from failures due to crushed dielectric or improperly attached connectors. In the normal use of a TDR-based cable fault measurement tool, the desired information is the distance to the fault, and not the locations of normally occurring splices and connectors. The portion of the TDR waveform that contains the information needed to determine the distance to the fault is the transition from a horizontal signal (small increase in voltage as time increases) to a vertical signal (large increase in voltage as time increases) and is called the ‘knee’ region (FIG.
2
). There may be several knee regions in a TDR waveform, and detailed analysis is required to identify the real fault data in the midst of many fault-like data groups in the measurement.
The analysis of such multi-element TDR waveforms involves the measurement and storage of many data points along the waveform. A common method used for measuring the TDR waveform is to sample the voltage at regularly spaced time intervals and save the data in a memory (see, e.g., U.S. Pat. Nos. 5,521,512 and 4,755,742). These samples can later be analyzed by a software program or a trained human technician. The absolute time values involved in the TDR waveform are in the nanosecond and sub-nanosecond range, and the voltages involved range from tens of millivolts to several volts, requiring expensive components for the sampling portion of the hardware. Many thousands of samples are needed for a comprehensive analysis, and this requires expensive memory. The analysis software itself can be large and can require expensive processing in order to reach a valid conclusion in a reasonable period of time. For example, a 3000 foot cable (electrical length) sampled at 6 inch intervals would require 6000 samples. And, since the range of voltage to be sampled is greater than 200 to 1, at least an 8 bit sampling device is required. Using low cost hardware, these samples could take 100 microseconds (&mgr;s) each to acquire, and then take 250 microseconds (&mgr;s) each to analyze. This would result in a measurement time of over 2 seconds, with a storage requirement of 6000 bytes.
Another common method of detecting cable faults with TDR measurements is to select a voltage and observe the time at which the TDR waveform crosses that voltage (see, e. g, U.S. Pat. No. 4,739,276). Such a method requires that the initial signal generated by the TDR device is sufficiently sharp to avoid dispersion effects. Additionally, the user is expected to select or vary that voltage and interpret the time readings. The problem with such techniques is that in the absence of detailed wave shape analysis, a large but inconsequential impedance variation is easily mistaken for the fault.
As more homes and small businesses install more computer and communications systems, the need for very low cost and very easy to use wire and cable test equipment has grown. Manual analysis of the complex TDR wavefor
Needle Dave
Shepard Stan F.
Allen Dyer Doppelt Milbrath & Gilchrist, P.A.
Jovial Test Equipment, Inc.
Kerveros James
Le N.
LandOfFree
Methods and apparatus for time domain reflectometry does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods and apparatus for time domain reflectometry, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods and apparatus for time domain reflectometry will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3014329