MXPA99011245A - Method of detecting corrosion in pipelines and the like by comparative pulse propagation analysis - Google Patents

Abstract

A method of dectecting corrosion on an elongate member (10), such as a pipe (10). Far side and near side electric pulses are transmitted into a magnetically permeable pipe (10) at spaced locations to travel toward one another. These are synchronized to intersect at various locations on the pipe (10). The resulting waveforms are analyzed by combining adjacent waveforms resulting from pulses intersection at spaced locations. Two combined waveforms are analyzed by subtracting one from the other to produce a difference waveform and the difference waveforms are compared to detect corrosion.

Description

METHOD FOR DETECTING CORROSION IN PIPES AND SIMILAR, BY COMPARATIVE ANALYSIS OF THE PROPAGATION OF IMPULSES Field of the Invention The present invention relates to a system, apparatus and method for testing elongate objects, such as tubes, and is directed toward the problem of detecting corrosion, defects or other abnormalities in pipes under conditions in which access and / or the visual inspection of the tube is not possible or is not practical.

BACKGROUND OF THE INVENTION In petrochemical and oil processing plants, as well as in other industrial environments, it is common to have numerous pipes extending between various locations in the plant, with these pipes transporting fluids or gas (for example, petroleum products), often under intense heat and high pressures. These tubes are invariably made of steel and can have an inside diameter that varies anywhere from two to sixty inches, or even outside this range. The exterior of these pipes is often insulated, with the insulation layers being as large as about 1/8 to 5 inches or more thick, or even outside this range. For a variety of reasons (safety, potential environmental hazards, prevention of costly stoppages, etc.) the integrity of these pipes or pipes must be maintained. Tube defects can occur for a variety of reasons. One of which is that the moisture condenses and deposits between the insulating layers and the tube, thus causing corrosion (ie, rust). Visual inspection of the steel tube that is encapsulated in an insulator is not possible, unless the insulation layers are removed and subsequently replaced. However, this is costly and time consuming, and for practical reasons it must be economically feasible to carry out inspections with reasonable frequency. It is the object of the present invention to provide pipe inspection means under the circumstances given above so that corrosion, other defects and / or abnormalities can be detected with a relatively high degree of reliability, and in a form in which the Various inspection difficulties, such as those mentioned above, can be eliminated and / or mitigated.

OBJECTIVES OF THE INVENTION The method of the present invention makes it possible for corrosion in an elongate and electromagnetically permeable member, such as a tube, to be detected quite effectively. More specifically, this method makes it possible for much of the irrelevant information (reflections, electromagnetic noise) to be removed from the waveform, and then the waveforms processed in a particular manner enable the clearer identification of the variations in the shape of the waveform. wave that would indicate corrosion. In the method of the present invention, electromagnetic or electric pulses (waves) of near side and far side are transmitted from, respectively, near and far side transmission locations, spaced, on an elongated member. The pulses (waves) travel towards each other to intersect at intersecting locations on the elongated member. Far-side pulses are received as waveforms at a reception location after intersection with related pulses from nearside. The transmission of the near-side and far-side pulses is synchronized so that the intersections of the pulses (waves) of the near side and far side occur at separate intersection locations on the elongated member.

The waveforms of at least two of the pulses (waves) on the far side that are spaced apart combine to form a composite waveform. A variation or variations of the composite waveform is determined as a means for detecting corrosion. In the preferred way, one of the waveforms of the two waveforms that are combined is inverted and then added to the other of the waveforms that are being combined to create a difference waveform, and the variations of the waveform are determined. waveform as a way to detect corrosion. Also, in the preferred embodiment, the near-side pulses passing through intersection points that are adjacent to each other, are considered sequential pulses of near side, with the sequence order being the same order in which the points of intersection they are spaced along the elongated member. The combination of the near-side waveforms is carried out in a pattern such that adjacent, first and second waveforms are combined to form a first composite waveform, the second waveform and a third waveform adjacent are combined to form a second composite waveform, the third waveform is combined with a fourth adjacent waveform to form a third composite waveform, with the pattern repeating itself with subsequent pairs of waveforms from of adjacent pulses from far side. Compound and adjacent waveforms are compared with others as a means to detect corrosion. A reference waveform is established by creating composite waveforms resulting from pulses that intersect in areas without elongated member corrosion, and identifying composite waveforms that differ from the reference composite waveform by a phase change and / or dispersion and / or amplitude and / or wave distortion.

Corrosion that is present between two adjacent points of intersection of the elongated member is detected by comparing a composite waveform resulting from the combination of the difference waveform that overlaps the point of intersection with difference waveforms on opposite sides. of the composite waveform of overlap. Also, the corrosion that is present at an intersection point of two waveforms can be detected by deriving two difference waveforms by combining the waveform at the corrosion point with the adjacent waveforms to form two waveforms. difference, which are then compared. Also, two additional difference waveforms, which are on opposite sides of, and adjacent to, the two difference waveforms that are analyzed to detect corrosion, are compared to the two difference waveforms that are combine at the point of intersection as a way to detect corrosion. Other features of the present invention will become clearer from the following detailed description. % Brief Description of the Drawings 20 Figure 1 is a somewhat schematic view of the system of the present invention being in its working position where it is being used in the testing of a section of insulated pipe; Figure 2 is a graph illustrating a way in which the data can be taken and represented according to the present invention. This graphic graph shows the propagation time against the distance from A to B and again from B to A, giving an inverse profiling; Figures 2A and 2B are two schematic drawings illustrating the intersection of two pairs of pulses at separate adjacent locations; Figure 3 is a graph that displays a curve at the bottom of the graph, which represents a composite waveform that results from both the near side and far side pulses traveling along the lower tube section test, and the curves in the upper part of Figure 1 showing the resulting waveforms in different locations, using the method of the present invention; Figure 4 is a graph that is similar to the upper part of the graph of Figure 3, displaying separately a first waveform resulting and identified at the zero location shown in Figure 3; Figure 5 is a graph similar to that of Figure 4, but shown separately the resulting waveform and identified at location 25 shown in Figure 3; Figure 6 is a graph similar to that of Figures 4 and 5, but showing separately the resulting and identified waveform at location 50 of Figure 3; Figures 7A to 71 are a series of simplified waveform illustrations to demonstrate certain principles of different waveforms; Figures 8A to 8E, as well as Figures 9A to 9E are two series of Figures similar to those of Figures 7A to 7E, and to further illustrate certain principles of different waveforms; Figure 10 is an illustration of the trajectories of the electromagnetic wave components traveling along a pipe section; Figures 11 - 16 are waveform representations illustrating the difference waveforms produced according to the method of the present invention; Figure 17 is a graph illustrating the waveforms in the first step of the third embodiment of the present inventions; Figure 18 is a graph similar to that of Figure 17, showing the waveforms of a subsequent stage in this third embodiment; Figure 19 is a schematic drawing of the placement of the antennas in this third embodiment; Figure 20 is a graph similar to that of Figures 17 and 18, illustrating a third stage in the third embodiment; Figure 21 is a graph illustrating three of the waveforms of Figure 20, plotted on a scale that emphasizes the vertical dimension of the waves; Figure 22 is a graph similar to that of Figure 21, showing the waveforms of Figure 21 and also the difference waveforms derived therefrom; Figure 23 is a graph similar to that of Figure 21, showing three of the waves displaced jointly; Figure 24 is a graph showing a plurality of difference waveforms, wherein two corrosion areas are being detected; Figure 25 is a graph derived from preliminary waveforms illustrating the difference in amplitude of difference waves where corrosion exists; Figure 26 is a graph based on Figure 24, further emphasizing the difference in amplitude; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The basic test apparatus and method of the present invention will now be described with reference to Figure 1. A tube 10 having a section 11 which is under test is shown. This tube 10 is or could be a pipe or pipe that would typically be used in the petrochemical or petroleum industry, where the pipe is made of steel surrounded by a coating or insulation layer. The apparatus 12 of the present invention is shown in some schematic way in its operating position, testing the section 11 of the tube 10. This apparatus 12 comprises a pulse generator 14, a signal analyzer 16 and an interactive computer 18, and two transmit / receive antennas 20 and 22. There are two cables 24 and 26 (or other means of transmitting signals or pulses) interconnecting the antennas 20 and 22, respectively, to the pulse generator 14. There is a second pair of cables or cables. other transmission means 28 and 30 connected between the respective cables 24 and 26, and to the signal analyzer 16. When the transmitted pulse is received by one or the other of the antennas 20 or 22, this pulse is in turn transmitted to the analyzer of signals. Some analysis can take place immediately in the signal analyzer 16. Alternatively, the information related to the pulse can be stored and analyzed at a later time. The computer performs certain control functions in the proper transmission and reception of pulses and other functions. There are several ways in which an apparatus such as the apparatus 12 can be used in detecting corrosion in pipes, and two of these are described below. There is a first method wherein a single pulse is transmitted from the pulse generator 16 to one or the other of the antennas 20 or 22 to cause the wave to be formed to travel from the location of this antenna 20 or 22 along the tube 10 to the location of the other antenna 20 or 22 where the signal of the waveform is received. The distance between the delivery location 20 and the reception location 22 is determined with precision, and the time of transmission of the pulse from the antenna 20 or 22 to the other antenna 20 or 22 is measured very accurately (desirably to a fraction of a nanosecond or even a fraction smaller than a nanosecond). If the section between the two test locations 20 and 22 is not corroded, and if the tube is uniform along its extension, then the pulse will arrive at the receiving location 22 in a waveform that is with the same pattern general (except possibly for distortions, such as a neighboring magnetic field, electromagnetic noise, etc.). Also, the pulse travel speed will remain substantially constant, as long as the tube remains without corrosion and is uniform. However, when corrosion is found between points 20 and 22, corrosion will affect the waveform by retarding its velocity by decreasing its amplitude, and also possibly by changing the actual waveform itself. One method for using this technique is to send the pulse from the transmission location to the receiving location on a non-corroded section of the pipe of a known length and diameters, and known characteristics relating to its transmission of electromagnetic waves. This would set the travel time of the wave from the transmission location to the reception location and the expected waveform configuration at the reception location. Later, the various sections of the pipe are tested, as illustrated in Figure 1. When there is a delay in the predicted arrival time of the waveform and / or deviations from the reference waveform for tube without corrosion, then it will be presumed that this is due to corrosion of the tube. However, it should also be understood that some other distortions (for example, near electromagnetic noise, presence of some other object that would produce distortion of the electromagnetic field) could also affect the waveform, and this should be taken into account. The second method that will be further described in this text is that termed the "double pulse" method, described in U.S. Patent No. 4,970,467. "In this method the same apparatus as shown in Figure 1 can be used. However, instead of using a single pulse or series of simple pulses, as in the method described above, both antennas 20 and 22 are used both as transmission antenna and receiving antenna in the same time frame. while a pulse is transmitted from the antenna 20, a pulse is also transmitted from the antenna 22. These pulses travel towards each other and "collide" at some intermediate location along the tube.This encounter of pulses will cause variations in both forms of wave as they move through the collision area to the other antenna that is in its reception location, appropriately coordinating the precise time in which the pulses are transmitted In the two locations 20 and 22, the point of collision along the length of the tube can be made to occur at any desired location along the length of the tube. Then by changing the relative times of transmission of the pulses in small increments, this point of collision can be staggered along the length of the tube. As described in the aforementioned patent, when the point of collision occurs in a location where corrosion exists, the waveform resulting from corrosion will be different from a reference waveform that would occur where the point of collision is in a section of the tube without corrosion. In this way, not only are there means to detect corrosion, but also means to detect the location of said corrosion. Also the antennas 20 and 22 could only be used as transmission antennas and two additional antennas could be used as reception antennas. In addition, other transmission and reception devices could be used, such as making a direct electrical connection to the tube. The present invention is particularly adapted to extract information from the waveforms resulting from the double pulse method described above.

In the following text a first embodiment of the method of the present invention is described, with reference to Figures 2A-2B to Figures 9A-9E. A second embodiment is also described here, later, using in part the same principles as those of the first embodiment, and this will be described later with reference to Figures 10-16. Reference is first made to Figures 2A and 2B, which are schematic illustrations of the operation of the double pulse method. In Figure 2A, an extension of one hundred feet of tube is schematically shown. The pulse will be considered to travel along the length without corrosion at the rate of one foot per nanosecond.

In Figure 2A, the near-side pulse is transmitted within the tube at the NS location (near-side location) at a point at the time indicated at zero. The second pulse is transmitted inside the tube at the far side location (designated FS), in this particular example, it is considered that the second far side pulse is transmitted within the tube forty seconds before the time when the side pulse close is transmitted within the NS location. Therefore, it can be seen that when the near-side pulse has traveled along the extension of the tube for forty seconds to reach a location indicated at the forty-foot location, the near-side pulse is transmitted at time zero from the near side location. The pulses of near and far side travel toward each other, each traveling thirty-five feet until they intersect at the thirty-five foot location on the one-hundred-foot tube. At the intersection, the two pulses interact with each other, and the near-side pulse continues its travel path to the near-side location (NS). Also, the near side pulse after passing through the intersection point continues its course of travel to the far side location (FS). At this time, it is important to note that each of the pulses is a complex waveform in some way. First, as a waveform travels along the length of the tube, it is subject to attenuation, distortion, interference and scattering. In addition, each wave can be considered as having what could be termed wave components conformed to first and subsequent arrivals. There is a first arrival that will travel the shortest course from the transmission location to the reception location. Thus, if both the transmitter and the receiver are on top of the tube, the first arrival will travel along the upper surface of the tube in a straight line. Then there are secondary arrivals that are pulse components that follow a helical path once around the tube to arrive a short time later. Then, there is a third, fourth, fifth, ... etc. Arrivals that reach later times. In addition, there are fairly common sources of external interference, such as sources of electromagnetic radiation, nearby objects that can interact with the waveform traveling along the tube, and so they are activated and in turn transmit their own electromagnetic radiation back to the tube under test. In addition, there are reflections and refractions. Returning to Figure 2A, first consider that a 100-foot section of tube under examination is free of corrosion. After the near-side and far-side pulses intersect at the thirty-foot location, there is a resultant waveform that reaches the near-side reception location, which is composed of both the original near-side pulse, and of the pulse of far side, with these having been modified or affected to some degree by reason of the intersection. It has been found that if the intersection of the pulses of near and far side takes place in a location on a tube that is not corroded, then the resulting pulse traveling through the intersection location will have certain characteristics typical of a situation where the intersection takes place in an area without tube corrosion. However, if the intersection of the near-side and far-side pulses takes place at a location where corrosion exists, the two pulses interact in a different way, and the resulting waveform of each of the intersected pulses has characteristics different However, the analysis of the waveform as a means to detect corrosion is difficult to quantify. There are features such as elevation time, slope, amplitude, and phase change, all of which are important characteristics of the waveform.

To illustrate this, reference is made to the lower curve shown in Figure 3. This curve, designated 50, is a composite curve resulting from the combination of pulses from both near and far sides. In this example, a transmission takes place on the near side, and the receiving antenna is also located on the near side. The part of the curve indicated at 52 represents the near-side pulse that is being transmitted inside the tube at the transmission location. The part of the curve indicated in the general area of 54 represents a portion of the composite wave that reaches the near-side receiving location, this being a combination of the components of the far-side wave and the near-side wave. . As indicated above, there are reflections, refractions, late arrivals, etc., which complicate the waveform. Reference is now made to Figure 2B, which shows a second double-pulse operation in which the transmission time of the far-side pulse has been delayed by four nanoseconds, so that it is transmitted thirty-six nanoseconds before transmission of the pulse. close side pulse. It can be seen that after the pulse of the far side has traveled thirty-six feet, the transmission of the pulse from the near side takes place. Thus, when the near-side pulse is transmitted, the far-side pulse is at the sixty-four foot location, and the two pulses intersect at the thirty-two foot location. With the foregoing being presented, the method of the present invention will now be described. Consider that the double-pulse test method is being carried out and that the near-side and far-side pulses are timed (as indicated in Figure 2A) so that there is an intersection at the thirty-foot location on the tube. one hundred feet Suppose further that the waveform that is received at the near-side location looks the same, or is similar to that shown at the bottom of Figure 3.

Now suppose that a second test operation is going to be started and that the far-side pulse is delayed by four nanoseconds. However, the transmission at the near-side location remains constant, in terms of time, and is still transmitted at time zero. As illustrated in Figure 2B, the intersection takes place at the location of thirty two feet. What this would effectively mean is that the portion of the composite curve that is likely to be attributed to the far-side pulse would have been delayed, relative to the transmission time of the near-side pulse, and that, with reference to the lower curve of the Figure 3, the portion of the composite waveform that is contributed by the far-side pulse would have been changed some way to the right, which is shown in the bottom curve of the graph in Figure 3. For the purpose of To extract significant information about the condition of the pipe, the following is done. First, the composite waveform that results from the transmission and intersection of the pulses as shown in Figure 2A is stored in the memory. Next, the second composite waveform resulting from the transmission of the near-side and far-side pulses according to Figure 2B is also received. Subsequently, the second composite waveform resulting from the test operation of Figure B is subtracted from the composite wave resulting from the test operation in Figure 2A. At this point, it is very important to keep in mind that the near-side pulse has been in both instances (in the operation of Figure 2A and in the operation of Figure 2B) transmitted at time zero. Thus, in the operation of both Figure 2A and that of Figure 2B, the near-side pulse has no position change. Of these, it becomes clear that the contribution of the near side pulse to the composite waveform is essentially subtracted from the composite waveform resulting from the operation of Figure 2A. Now, let's turn our attention to the far-side pulses of the test operation in Figure 2A and 2B. With the far-side pulse having been delayed by four nanoseconds, the wave component of the far-side pulse has now been changed from the first location in the first operation of the four nanoseconds of Figure 2A to the composite curve of the second location of the second test operation of Figure 2B. With the first complete composite curve being subtracted from the second complete composite curve, what can be termed a "difference waveform" remains. It has now been found that if in double-pulse operations where the intersection locations are staggered within a reasonably close distance, one from the other, and if tube is found without corrosion at both intersection locations, the waveform of The resulting difference is a reasonably well-defined and identifiable peak. Reference is now made to Figure 3. It can be seen that the curve at the top of Figure 3 shows three separate peaks and designated "zero", "twenty-five" and "fifty", respectively. Each of these peaks is the result of using the method of the present invention, wherein the point of intersection of the two adjacent double pulse operations has been staggered in a range of approximately 5 feet. To provide clearer representations of the waveforms, Figure 4 illustrates the simple curve indicated at "0" in Figure 3; Figure 5 illustrates only the curve identified in "25" in Figure 3; and Figure 6 illustrates only the curve indicated at "50" in Figure 3. It should be noted that the waveforms indicated at "0", "25" and "50" at the top of Figure 3 are waveforms. reals extracted from adjacent waveforms and similar to those shown at the bottom of Figure 3. It is important to note that if the composite waveform is not formatted correctly, the difference between adjacent waveforms does not provide the waveforms of "effect" shown in Figure 3.

To review how to correctly format the information, consider the two active waveforms on the tube, one from the near side (NS) and the other from the far side (FS). When the information analyzer is synchronized with the near-side pulse, the component of the near side of the composite pulse will not move (that is, change of position). However, the FS (far side) pulse, which is synchronized to the master clock, will scroll through the screen from left to right and will modify the composite waveform for each intersection along the tube. When the composite waveforms are subtracted from each other, two important things happen: 1. The effect of the NS pulse, which has a very large amplitude with respect to the FS pulse is canceled, since this NS component in the waveform compound is fixed in time. 2. The resulting difference waveform represents the difference between the two adjacent FS pulses that have intersected with the NS pulses at two different points on the tube. When this difference happens, then the "effect" (elevation of time, slope, amplitude, dispersion and absolute time, among many parameters that are affected by corrosion) influence the configuration of the different waveform. The difference waveform will shift in time with respect to other adjacent pairs. This displacement in time is a good indication of the condition of the tube, as long as it can be interpreted in a meaningful way. The difference waveforms shown here are examples of waveforms that are well defined but are very difficult to extract real-time information. (See Figures 4, 5 and 6). Notice in particular Figure 6 on the "knee" of the waveform is not well defined, and could be selected anywhere from a point close to three thousand two hundred to four thousand two hundred, a range of one hundred nanoseconds. The automation of a selection of the location of the absolute time of the knee is very difficult and sometimes impossible. However, the peak is well defined. The peak is not only a voltage difference between two different response waveforms, but is determined by the configuration factors involved with the leading edge of the two adjacent waveforms. For example, if the two adjacent waveforms are displaced more in time than any other two adjacent waveforms, it will result in an increase in amplitude. Hence, "? E" (peak) is a function of time. It is also a function of the difference in actual amplitude between two different waveforms. Also, if the leading edge of a waveform is distorted as a result of corrosion, this distortion will result in a change in amplitude in the difference waveform and in a change in the position of the peak with respect to time. When the tube is in good condition, the peak is very sharp and the waveforms configuration of extremely uniform difference. When the tube has anomalies (eg, corrosion), the configuration of the difference waveform is significantly altered and the effect of corrosion (EC) displaced by significant differences in the leading edge. These differences result in an effective peak change that may be directly related to the quality of the tube. This peak change is much easier to instrument and measure than other parameters. This change in peak is also an indicator of the cumulative effect of all individual parameters and the effect of the electromagnetic response, even if these could be very difficult to measure individually. As the tube degrades, the peak is more easily distorted due to the complex contribution of all the driving forces. In a perfect system, each pulse of difference must be identical. Hence, the measurement of time associated with the first peak occurring after an indefinite knee of a differential pair provides an effective way to extract critical information and measure the effect of corrosion.

Obviously, a stable source is required and is being used by this system. From the included waveforms, it will be obvious that the measurement of the peak is easier than measuring the time related to the knee. When the peak is not well defined, it will be indicative of different abnormalities in the tube. The process of measuring the effect of corrosion is designed to impose the quality of information and reduce the time required to collect and analyze information in the field. To illustrate in a more simplified form certain aspects of the present invention, with respect to the subtraction of one waveform from another, reference will now be made to Figures 7A to 7G, and also to Figures 7H to 71. Figure 7A shows a rather simple waveform 60 which is plotted, for convenience, in straight lines. Figure 7B shows the same wave at 62, but one unit out of waveform 60. Figure 7C shows the summation of waveforms 60 and 62 as waveform 64. It is possible to derive meaningful information from the form of wave in Figure 7C where the waveforms are added, but it is preferred to first invert one of the waveforms and then add them together. This is done in Figure 7D, which shows the waveform 60, with the waveform 62 offset and inverted, and the summation of the waveform 60 and the inverted wave 62 performs a subtraction of the waveforms of Figure 7 A and 7B. This results in the difference in the difference waveform 66 shown in Figure 7E. As indicated above, these curves are somewhat artificial and, in reality, these simple waveforms would not be formed with these straight lines. Rather the Figures T? and 71 would be more real, where we observe the waveform 68 and a very similar waveform 70 offset in the waveform 68. In 71, a waveform of difference 72 is shown which would result in subtracting wave 70 from the waveform 68. It can be seen that the difference waveform 72 is configured in a better defined peak.

For purposes of further analysis, in Figure 7F, waveform 74 is shown, in exactly the same shape and position as waveform 60 of Figure 7A. Subsequently the same waveform is shown in Figure 7F at 76, inverted and changed two units from the waveform 74. Then when the waveform 76 is subtracted from the waveform 74, there is the difference waveform 78 shown in Figure 7G. It will be noted that with the waveform 76, spaced two units away from the waveform 74 (instead of a unit, as in Figures 7A and 7B) it has an amplitude that is twice the amplitude of the waveform of 66 difference. This illustrates that if the time displacement of waveforms increases, we would expect the amplitude of the difference waveform to increase. This is simple by way of illustration, and refers only to a particular facet of corrosion detection from the difference waveform. For purposes of further analysis, reference is made to Figure 8A to 8E and to Figures 9A to 9E. In Figure 8A, a waveform 80 is shown, and in Figure 8B a second waveform 82 is advanced and has been attenuated and delayed, presumably due to corrosion encounter in the tube. Figure 8C shows the sum of these, this being the waveform 84. Figure 8D shows the same waveform 80 and the adjacent waveform 82 inverted. Figure 8E shows a difference waveform 86 that results by subtracting waveform 82 from Figure 8B of waveform 80 of Figure 8A. In Figure 9A, these same steps are followed. Figure 9A shows a waveform 88 that is the same as the waveform 80 of Figure 8A. Figure 9B shows a second waveform 90 delayed in one unit, and having a different slope along the leading edge. Figure 9C shows the waveform of the summation at 92. Figure 9D shows the waveform 88 and the second waveform 90 inverted. Figure 9E shows the difference waveform at 94.

When reviewing Figure 7A to 71, Figures 8A to 8E, and also Figures 9A to 9E, four of these figures show difference waveforms, these being the difference waveform 66 in Figure 7E, the form of difference wave 78 in Figure 7G, difference waveform 86 in Figure 8E, and difference waveform 94 in Figure 9E. It can be appreciated that the different characteristics of these difference waveforms emphasize the difference between the adjacent waveforms. It should be kept in mind that these waveforms of Figures 7A-7I, 8 A-8E and 9A-9E are not the most complex composite waveforms as shown in 50. These are simplified waveforms that are simply provided for show some of the principles involved. What the method of the present invention accomplishes is the elimination of a large amount of irrelevant information. There is a tendency for the near-side pulse to sink the far-side pulse, mainly because the near-side pulse has a substantially larger amplitude, since it is closer to the transmission location. The components of the pulses of near side are substantially eliminated. Beyond this, subtracting the changed wave components and attributable to the pulses from far side, one another, is how a difference comparison is provided. If the tube is substantially uniform along its extension (without corrosion) and if the point of intersection of the pulses is staggered in uniform increments along the tube, then the same or very similar difference waves would be expected to occur. get. As indicated above, the difference curves shown in the upper part of Figure 3 are quite similar, indicating no corrosion or possibly minimal corrosion. Thus, the difference curves as shown in Figures 4 to 6 provide meaningful information, without being confused by extraneous wave components.

A second embodiment of the present invention will now be described with reference to Figures 10 to 16. As an introduction, much of the focus on the analysis of waveforms for detecting corrosion has been directed to the leading edge of the form of wave or at least the first portion of arrival of the waveform. To some extent, it has been recognized (or at least conjectured) that valuable information would be contained in the final arrival portions of the waveform. However, the problem is how such information could be identified and / or extracted. As indicated earlier in this text, it can be considered that the propagating waveform is a composite of a number of wave components formed at least in part from first and subsequent arrivals. To illustrate this graphically, reference is made to Figure 10, which shows a relatively short section of the tube, where there is a transmission location 102 and a reception location 104. In this example, these locations 102 and 104 are both the top of the tube and aligned. The straight line in the direction of the length between the points 102 and 104 is indicated at 106. Since this axis 106 is the shortest path between the points 102 and 104, the path of the first arrival would be along the path indicated at 108, which is coincident with axis 106.

In addition to the wave component 108 of the first arrival, there are two wave components of the second arrival, the travel paths of which are indicated at 110 and 112. It can be seen that each of these are helical trajectories, which travel longitudinally and through of a 360 ° helical curve. Then the third arrivals are indicated at 114 and 116, and these are also helical paths, but with a total traveling circumferential component of 720 °. Obviously, the second arrival has a longer travel path than the first arrival, the third arrival has a travel trajectory still longer than the second arrival, etc. If there is corrosion in the tube, at least some of these late arrival pulse components will pass through the corrosion area or areas and that path component will be delayed, attenuated and / or modified in some way. With the foregoing given as background information, reference will now be made to Figure 11, which illustrates the waveforms obtained by the second embodiment of the present invention. In Figure 11, the vertical axis represents the voltage (measured in volts) and the horizontal axis measures time, with each increment representing ten nanoseconds. Thus, the numeral one thousand actually represents one hundred nanoseconds, the numeral two thousand represents two hundred nanoseconds, etc. It can be seen that the waveforms presented in Figure 11 extend over the whole of five hundred nanoseconds. The particular tests from which these curves were obtained were made in a pipe section one hundred and sixty feet in length (that is, the transmission location was one hundred and sixty feet from the receiving location). In addition, the double pulse method was used, as indicated above. Since the entire tube is surrounded by electromagnetic energy, the effect of corrosion on any part of the tube will appear in the difference waveform obtained by the intersection of the pulses at the location of the corrosion. The first steps in the second embodiment in the method of the present invention are substantially the same as those in the first embodiment. More specifically, a first test operation was carried out by transmitting the near-side pulse and the far-side pulse in a time-regulated relationship so that they are at a certain point of intersection. There is a composite waveform that results from this first test operation and that is stored. Subsequently, as described in the presentation of the first modality, there is a second test operation in a time-regulated relation of the far-side pulse so that it is either advanced or delayed for the point of intersection to be changed, and the result was a composite wave that is recorded and that had components of the far-side pulse somehow changed from the previous composite wave. As described in the first embodiment of this method of the present invention, one of the composite waveforms is subtracted from the other to obtain a difference waveform. These steps are performed in the second embodiment of the present invention and it will be recognized, of course, that these are substantially the same steps as described in the first embodiment. From this point forward in the method of the second embodiment, a subsequent analysis is carried out as described below. For purposes of description, we will consider the sequence of difference waveforms and designate these as the waveform of difference 1, waveform of difference 2, waveform of difference 3, etc. It will be evident that the difference waveform 1 results from processing the composite waveforms 1 and 2 which result from the first and second double pulse test operations; the waveform of difference 2 is the result of processing the composite waveforms 2 and 3 that result from the second and third double pulse operations, etc. In Figure 11, the difference waveform 120 is first plotted, and this waveform is approximately five hundred nanoseconds in length. The next step consists of plotting the second difference waveform 122, but the second difference waveform 122 is shifted to the left, and also further underneath in some way so that the second difference waveform 122 is aligned with and a short distance below the first difference waveform 120. It will be seen in the representation of the waveform of Figure 11 that the two waveforms 120 and 122 tie each other in a rather close manner. These two difference waveforms 120 and 122 were derived from adjacent composite waveforms, and both composite waveforms resulted from a double-pulse operation where the pulse is intersected in an area without corrosion (or at least one area with very light corrosion) of the tube under test. As indicated on the right side of Figure 11, the upper composite wave results from a difference waveform where one reference point of intersection was that of the 125.3 foot mark, while the second waveform was of difference 122 was made up of composite and adjacent waveforms at reference location 129.6. Figure 12 shows two other reference waveforms 124 and 126, which result from two adjacent pairs of composite waveforms at reference locations at the locations of 34.6 and 38.9 feet above the tube section under test. These composite waveforms also resulted from the pulses of the far side and near side of each test operation intersecting in an area without corrosion (or very slight corrosion) of the section of the tube under test. Reference will now be made to Figure 13, where the waveform 126 is shown as the upper waveform, as shown in Figure 12, this difference waveform having a reference location of 38.9 feet in the tube. The lower waveform 128 has a reference of 43.2. This was in a section of the pipe somehow corroded and that had a corrosion index of 1.0262. (The corrosion index is a scale that is used by the inventor in the hierarchy of the corrosion areas, a categorization of 1,000 would be zero corrosion and the higher the number, the greater the severity of the corrosion). It will be appreciated in Figure 13 that the lower waveform 128 has in two areas somewhat of a phase change, indicated at 130 and 132. Reference will now be made to Figure 14, where there are two adjacent reference waveforms in locations along the pipe section at the locations of 151.2 feet and 155.5 feet. There was a corrosion index of 1055, which is higher than that of Figures 11, 12 and 13. These waveforms are indicated at 134 and 136. It can be seen that at location 138 there is rather an important phase change.

Next, attention is directed to Figure 15, which shows as the upper waveform the same waveform 136 which is the lower waveform in Figure 14; and a new difference waveform 140 taken at the reference location of the 160 feet in the tube. These waveforms resulted from composite waveforms that developed with intersecting locations being in the most highly corroded area. Several features should be noted. When observing the location of the peak at 142, and the locations of the two peaks at 144 and 146, it can be seen that there are substantial differences in amplitude relative to the second and third peaks between these curves 136 and 140. In addition there is a significant change in phase that is indicative of corrosion anomalies. It is remarkable to observe the difference curve 134 which is at the reference location 151.2 (Figure 14) and the curve 140 that is in the reference location 160 (Figure 15). It can be seen by coupling the curves 134 and 140 that these correspond slightly to one another closely, at least much closer to each other than what each one is coupled with the curve 136. This would indicate that the corrosion area is more probable in the area of reference location 155.5, presumably somewhere between the area of 153 to 158. Finally, reference is made to Figure 16, where there are two waveforms of difference 150 and 152. It can be seen that the Waveform of difference 152 has substantial similarities to curve 136 (see Figures 14 and 15). Furthermore, it can be appreciated that the coupling between the waveforms 150 and 152 is rather similar to the coupling of the waveforms 134 and 136 in Figure 14. More specifically, it can be seen that the second and third peaks 154 and 156, respectively, waveform 152 is of almost equal amplitude, and then there is a phase change in area 158. This is a pattern quite similar to that shown in area 138 in Figure 14. In this way, it can be seen that with the method of the second embodiment of the present invention, the presence of corrosion is detected by a method that could be called "full-wave analysis", which involves looking not only at the waveguide, but also at a much longer time duration of the waveform that also contains important information. It also becomes clear that valuable information is obtained from portions of the waveform as far along the waveform as two hundred to four hundred i nanoseconds or larger from the first arrival of the electromagnetic pulse.

In addition, the location of the corrosion can be located within reasonably close tolerances by appropriately synchronizing the pulses so that the point of intersection is known. Also, it should be noted that these readings were taken on the same section of the tube, but with the intersection being moved to different locations. Therefore, all of the waveforms developed for the data of Figures 11 ugh 16 passed over the same section of the tube. The key difference is that the point of intersection was moved. When the point of intersection of the waveforms that were combined to form the difference waveforms was in an area of corrosion, the variations of the difference waveforms became clear.

A third embodiment of the method of the present invention will now be described. In this third modality, as in the previous modalities, the pulses will be transmitted from the far side and near side locations, and in this particular modality, the waveform that will be analyzed to detect corrosion is the far side pulse that arrives at a reception location adjacent to the transmission location on the near side. Also, in this third mode, the time intervals between the transmissions of the far-side pulse will remain constant. Thus, to synchronize the pulses so that the point of intersection is stepped along the length of the tube, for each transmission, the pulses of near side will advance in a time relation by a short increment so that the point of intersection of the pulses will be staggered in a direction from left to right through Figure 18. In the first step of this additional mode, the far-side transmitter is turned off, and a series of pulses are transmitted from the near side and these are collected by the near-side receiving antenna that is closely adjacent to the near-side transmitter. The time relation of the transmission of the near-side pulses is synchronized in time with respect to a dispatcher, except that each subsequent transmission is advanced by an additional time increment, and in this particular example it will be assumed that it is being advanced by four. nanoseconds for each pulse transmission. The first step in the method of this preferred embodiment will now be described with reference to Figures 17 and 19. In Figure 19, there is shown a section of tube 170 having a near side transmission antenna 172 and a power transmission antenna. far side 174. There is a reception antenna 176 spaced from the transmission antenna 172 a short distance to the far side transmission antenna 174. Initially, the far side transmitter remains inactive so that the Texan side transmission antenna 174 do not transmit any signal. The near-side transmitter is activated to transmit a series of pulses in a time relationship and which are synchronized at regular time intervals. This is done in a way in which each subsequent pulse is advanced four nanoseconds in relation to the preceding pulse. For example, if the pulses of near side are going to be transmitted every two seconds, less the time of the advance of the timing, the first pulse would be sent to zero seconds. The second pulse would be transmitted four nanoseconds before the interval of two seconds. The third pulse would be sent eight nanoseconds sooner than the four second interval. The first pulse would be sent twelve seconds before the interval of 8 seconds, etc.

In this way, as can be seen in Figure 17, the first pulse is transmitted at 0 nanoseconds, the next pulse indicating that it has an advance of four nanoseconds, the third pulse at 8 nanoseconds in advance, with these advances continuing down until the 18th pulse, which has been advanced in 72 nanoseconds with respect to the initial pulse to 0. i Each pulse coming from the antenna 172 on the near side passes through the receiving antenna 176, and the pulse is recorded, with the forms of wave indicated in 178. It will be recognized that in most cases there is a certain amount of electrical noise, electromagnetic noise, echoes, external refractions, etc. That tend to obscure or "hinder" the signal. Each of these pulses 178 is recorded in the memory of the control unit, including all the various influences foreign to the signal, plus the part of the signal attributable to the pulse itself. As discussed hereinafter, these pulses 178 which are recorded are used as reference pulses which are subtracted in a subsequent step in the method of the third embodiment. The next step will now be described with reference to Figure 18. The far-side transmitter is activated so that time-controlled pulses are regularly transmitted from the antenna 174 in the tube section 170, that is, without any advance or delay in the timing. Each transmission on the near side is synchronized with the transmission on the far side. However, each time the next side transmits a pulse, the next pulse from the near side is advanced four nanoseconds from the designated time period from the previous pulse. Thus, at the location zero in Figure 18, the far-side pulse is transmitted over a period of time so that the pulses from the near-side and far-side antennas 174 intersect at the location of the antenna reception 176. The next pair of pulses are transmitted with the near side pulse being advanced in four seconds, so that the point of intersection is separated two nanoseconds closer to the far side. The third pulse is advanced in eight seconds so that the intersection is spaced an additional two nanoseconds to the far side. It can be seen that in the first pair of pulses that intersect in the antenna 176, their peaks come close until they match. It can be seen that subsequent pairs of i pulses are transmitted and with the near-side pulses being forward four nanoseconds in each transmission, relative to the far-side pulse, whose wave pattern that is received in the antenna 176 comprises a first peak 182 which is attributable to the near side pulse passing through the antenna 176, and a second peak 184 which is attributable to the near side pulse that reaches the antenna at a later time. In this third embodiment, the far side pulse that is received by the antenna 176 is the one that is analyzed to determine if there is corrosion. To carry out this the following procedure is developed. Each of the waveforms 186 resulting from the second stage of this method is also stored in the memory. Then the waveforms 178 (shown in Figure 17) are subtracted from each of the corresponding waveforms shown in Figure 18 with the resulting waveforms that are being shown in 188, Figure 20.

What has happened is that when the waveforms in Figure 17 are subtracted from the corresponding waveforms of Figure 18, the waveform from the near side, along with strange noise, echoes, etc. it is canceled so that what remains is the waveform 188 that essentially represents the far side wave that is "unobstructed". The overall result is that this facilitates the detection of variations in the waveform that originated from the far side. It should be noted that there is a relatively small amount of corrosion, its effect on the intersecting waveforms at the location of corrosion is more difficult to detect. By performing the first three stages as described with reference to Figures 17, 18 and 20, these more subtle variations in the resulting waveform of the far-side pulse intersecting the near-side pulse in the corrosion area can be detected more easily. Figure 21 shows four adjacent waveforms 190 which are the same waveforms 188 of Figure 20, except that the vertical dimension has been substantially increased so that the slope of these waveforms 190 is steeper. It can be seen in Figure 21 that each of these four waveforms 190 are very similar to each other. This would indicate that there is little or no corrosion in the area where the waveforms 190 have intersected. A fourth step is now developed in the method of the present invention to further improve the ability to analyze waveforms to detect corrosion, and this will be explained first with reference to Figure 22. Figure 22 represents four adjacent waveforms that they result from four transmissions of pulses that follow one after the other in sequence. These waves are designated 192-1, 192-2, 192-3 and 192-4. Each waveform is subtracted from the preceding wave to obtain a waveform of difference. This is accomplished by first inverting the 192-2 waveform and then adding this inverted waveform to the 192-1 waveform to obtain a difference waveform that is 194-1 (this 194-1 being the difference waveform of the two waveforms 192-1 and 192-2). In a similar manner, a second difference waveform 194-2 is obtained by inverting the 192-3 waveform and adding it to the 192-2 waveform to obtain the difference waveform 194-2. The third difference waveform 194-3 is obtained in the same way, inverting waveform 194-4 and adding it to waveform 192-3. It can be seen in Figure 22 that each of the three different waveforms 194-1, 194-2 and 194-3 are very similar to each other and have substantially the same amplitude.

Figure 23 illustrates another technique used in this third mode. The four waveforms 194-1 to 194-4 are displaced to be closer together, while the waveforms are left unchanged. By moving these waveforms so that they are very close to each other, it is much easier to detect variations in the waveforms. Also, the different waveforms that would result from the arrangement of the waveforms in Figure 23 would be of a much smaller amplitude. The effect of this is, however, that the differences in amplitude between the peaks do not diminish when the waveforms are displaced so that they approach each other. This also accentuates the differences in waveforms. To illustrate the waveforms where corrosion is being detected, reference is now made to Figure 24. 18 adjacent difference waveforms are shown as shown in 194-1, 194-2 and 194-3 in the Figure 22. It can be appreciated that the fourth waveform 196 and the fifth waveform 198 are configured rather differently than the adjacent waveforms shown immediately above and below these two waveforms 196 and 198. It will be appreciated that between the initial "hump" 199 and the second "hump" 200 of the waveform 196 there is a major depression at 201. Furthermore, it will be noted that the second peak or "hump" 202 of the waveform 198 has a much greater amplitude. smaller. Furthermore, it can be seen that the peak 203 of the first elevation or "hump" 204 of the waveform 198 is shifted to the right. An alignment line 204 is plotted to show the change from the alignment line to the left. It will also be appreciated that there is a change in the peak 205 of a waveform that is the twelfth waveform from the top. This is also an indication of corrosion, but a smaller degree of corrosion compared to the corrosion detected in the area of the fifth and sixth waveforms. Figure 25 is a graph in which points of the peak amplitude for adjacent difference waveforms have been prepared by plotting lines that 3l connect points of adjacent peaks. It can be seen that the indicator peak at 206 is much larger than the rest of the peak points, and this would indicate an area of corrosion. The peak indicated at 208, although it does not have the peak height at 206, still rises above the others. This would indicate that corrosion would probably be found at the location of the tube represented by point 208, which would be the peak of the difference waveform of two adjacent waveforms where the corrosion was at or near the location of the intersection . Figure 26 is a graph similar to that of Figure 25, where the amplitude of the points in Figure 25 has been amplified to accentuate the differences. It should be understood that the terms "near side" and "far side" can be reversed. Furthermore, it is understood that although the third modality has been described with the pulse of the far side being the pulse that is analyzed, and the pulse of the near side that has been advanced to make the point of intersection be stepped along the elongated member (tube), this arrangement could be reversed. In addition, in the actual test operation, both pulses from near side that reach the far side, and pulses from far side that arrive at the near side could each be received and analyzed. Also, it should be understood that various modifications may be made in the present invention without departing from the basic teachings of the same. Furthermore, the terminology used in this description should, in the following claims, be given an equal interpretation with the scope of the invention and should not be construed as being limited to the specific procedures and operating components described herein.

Claims (9)

1. f 32 Novelty of the Invention 1. A method for identifying corrosion in an elongate, electromagnetically permeable member, such as a tube, said method comprising: a. transmit electric or electromagnetic pulses (waves) of near and far side from, respectively, transmission locations of near and far side, spaced on said elongate member, with said pulses (waves) traveling towards each other to intersect at intersection locations in said elongated member. b. receiving said far-side pulses as waveforms at a reception location after intersection with related near-side pulses (waves). c. synchronize the transmission of pulses (waves) from near and far sides so that the intersections of the waves from pulses from near and far sides occur at intersecting locations spaced at said elongate member. d. combining the waveforms of at least two of said far-side pulse waves, which are separated from one another, to form a composite waveform; and. investigate the variation or variations in said composite waveform as a means to detect corrosion.
2. 2. The method as recited in claim 1, wherein one waveform of the two waveforms to be combined is inverted and subsequently added to the other of the waveforms that are being combined to create a difference waveform, and variations in said difference waveform are investigated as a means to detect corrosion.
3. 3. The method as recited in claim 2, wherein far-side pulses passing through intersection points that are adjacent to each other are considered to be sequential, far-side pulses, with the order of sequence being the same as the order in which the intersection points are spaced along the elongated member, and the combination of the far-side waveforms is carried out in a pattern such that the first and second adjacent waveforms are combined to make a first composite waveform, the second waveform and the third adjacent waveform i are combined to make a second composite waveform, the third waveform is combined with a fourth adjacent waveform to make a third composite waveform, with the pattern repeating itself with subsequent pairs of waveforms from adjacent far side pulses.
4. 4. The method as recited in claim 3, wherein adjacent, composite waveforms are compared with each other as a means of detecting corrosion.
5. The method as recited in claim 4, wherein a reference waveform is established by creating composite waveforms resulting from pulses that intersect away from areas with elongated member corrosion and identifying composite waveforms that differ of the reference composite waveform in contrast to phase and / or dispersion and / or amplitude and / or wave distortion.
6. The method as recited in claim 5, wherein the corrosion that is present between two adjacent points of intersection in the elongate member is detected by examining a composite waveform resulting from combining the overlapping difference waveform. the point of intersection, with the difference waveforms on opposite sides of the composite overlap waveform.
7. The method as recited in claim 5, wherein corrosion is present at a point of intersection of two waveforms, and two difference waveforms are derived by combining the waveform at the point of corrosion with forms of adjacent waveforms, and these are compared with another waveform to investigate the presence of corrosion.
8. 8. The method as recited in claim 7, wherein two additional difference waveforms, which are on opposite sides of, and adjacent to, the two waveforms that are compared to detect corrosion i, are also compared to the two difference waveforms that are They combine at the point of intersection, as a means to detect corrosion. The method as recited in claim 1, wherein pulses from a far side passing through the intersection points that are adjacent to each other, are considered to be sequential, far-side pulses, with sequence order being the same as the order in which the points of intersection are spaced along the elongate member, the combination of the far-side waveforms being carried out in a pattern such that the waveform resulting from the intersection of Adjacent pulses of near, first and second side, with related pulses from far side, combine to make a first composite waveform, waveforms of near, second and third side pulses, which are combined with related pulses from far side they combine to make composite waveforms, with this pattern repeating itself for subsequent pulses.