Self-sealing electrical cable having a finned inner layer

[0002] This application is a continuation-in-part of application Ser. No. 09/730,661, filed Dec. 6, 2000 which is a continuation-in-part of patent application Ser. No. 09/228,482, filed Jan. 11, 1999, now U.S. Pat. No. 6,184,473, which is relied on and incorporated by reference.

[0003] Insulated solid and stranded electrical cables are well known in the art. Generally stranded cables include a central stranded conductor with a protecting insulation jacket disposed around the conductor.

[0004] The most frequent cause of failure of directly buried aluminum secondary cables is a cut or puncture in the insulation inflicted during or after installation. This leads to alternating current corrosion of the aluminum and finally to an open circuit. When a conductor is exposed to wet soil, upon damage, leakage current may flow, and cause localized electrochemical conversion of aluminum to hydrated aluminum oxide and eventually to an open circuit of the conductor.

[0005] In the U.S., thousands of such instances occur annually and the repair (location, excavation, repair, and replacement) can be very costly. As a result of the failures and in response to this problem, a tougher insulation system was introduced and became an industry standard. The tougher cable is described as “ruggedized,” and generally consists of two layers: an inner layer of low density weight polyethylene and an outer layer of high density polyethylene. This design is more resistant to mechanical damage than one pass low density polyethylene, but still can result in exposure of the aluminum conductor if sufficient impact is involved.

[0006] Investigations show that AC electrolysis current can approach half-wave rectification when the current density is high. This accounts for the rapid loss of aluminum metal frequently experienced in the field. A caustic solution (pH 10-12) develops at the aluminum surface and dissolves the protective oxide film.

[0007] The mechanism of aluminum cable failure is the formation of hydrous aluminum oxide. As the aluminum oxide solids build up, the insulation in the vicinity of the puncture is forced to swell and splits open, making larger areas of the aluminum conductor surface available for electrolysis, thus increasing the leakage current and accelerating the corrosion process. Rapid loss of aluminum by AC electrolysis continues until ultimately the cable is open-circuited. A caustic environment is created at the aluminum, electrolyte interface, which dissolves the protective oxide film.

[0008] The ruggedized or abuse resistant type insulation was supposed to protect the cable from physical abuse. While it helped this problem, it did not eliminate 600 V cable failures. Utilities have recently repotted varying numbers of 600 V aluminum underground distribution cable failure rates scattered between 70 and 7000 per year. Failures are evidenced by an open circuit condition accompanied by severe corrosion of the aluminum conductor.

[0009] All the reasons for 600 V failures are unknown, but several have been postulated by cable users. These cables seem to experience a high degree of infant mortality, followed by failures occurring over decades. The infant mortalities are usually directly related to damage caused by adjacent utilities, damage inflicted by landscaping and planting, or damage to the cable prior to or during installation. The failures occurring years later are harder to explain. There have been postulations of lightning damage, manufacturing defects, or insulation degradation over the life of the installation.

[0010] In order to better understand the insulation characteristics, studies of the AC breakdown, and DC impulse breakdown were conducted. AC breakdown studies on several different cables showed a high safety margin of performance. Each of these cables had a 0.080 inch wall thickness. Tests were conducted in water filled conduits. The AC breakdown strength of all of these cables was consistently above 20 kV, far in excess of the operating stress.

[0011] Impulse breakdown studies have also been performed on several 600 V cable constructions having different insulation formulations. The impulse breakdown level of these cables was approximately 150 kV. This exceeds the BIL requirements of a 15 kV cable system and should well exceed the impulses on 600 V secondary cables during operation.

[0012] The above margins of electrical performance were measured on new cables. They are far above what is needed to operate on a 600 V system since most of these cables operate at 120 V to ground. One of the tests during compound and product development is a long term insulation resistance test performed in water at the rated operating temperature of the insulation. For crosslinked polyethylene cables the water temperature is 90° C. The insulation resistance must demonstrate stability and be above minimum values for a minimum of twelve weeks. If there is instability indicated, the test is continued indefinitely. Relative permittivity is measured at 80 v/mil and must meet specific values. Increase in capacitance and dissipation factor are also measured in 90° C. water over a 14 day period. Insulation compounds used in present day cables easily meet these requirements.

[0013] Manufacturing defects in cable insulation are found during production by either of two methods. During the extrusion process, the cable is sent through a spark tester, where 28 kV DC, or 17 kV AC, is applied to the insulation surface. Any manufacturing defect resulting in a hole in the insulation will initiate a discharge, which is detected by the spark tester. Most manufacturers use this method. Another test that is also often employed is a full reel water immersion test. In this test 21 kV DC, or 7 kV AC is applied to the cable after immersion for 1 hour or 6 hours, depending on whether the cable is a plexed assembly or single conductor, respectively. The actual voltages used for these tests are dependent on the wall thickness. The above values are for an 0.080 inch wall.

[0014] The above testing has demonstrated electrical performance that is stable and far surpasses the requirements of the installation for 600 V cable. This does not explain a sudden cable failure after many years of operation. Such sudden failure can be explained by a better understanding of the failure mechanism. Aluminum corrosion in the presence of an alternating leakage current is a combination of two different mechanisms. Aluminum is normally afforded a great deal of corrosion protection by a relatively thin barrier layer of aluminum oxide, and a more permeable bulk layer of oxide. However, flaws or cracks exist in these layers which provides a spot for the corrosion reaction to begin. The metal in contact with water undergoes an anodic (positive ions moving into solution) and a cathodic cycle, sixty times per second.

[0015] During the anodic half cycle of leakage current, aluminum ions leave the metallic surface through these flaws and combine with hydroxyl ions in the water surrounding the cable. This reaction results in pitting of the metal and the formation of aluminum hydroxide, the whitish powder evident in corroded cables. Another important reaction also occurs. The hydroxyl ions are attracted to the metal surface during this half cycle, which increases the pH, causing a caustic deterioration of the oxide layer, further exposing more aluminum.

[0016] During the cathodic half cycle another reaction occurs. Hydrogen ions are driven to the aluminum surface. Instead of neutralizing the caustic hydroxyl concentration, the hydrogen ions combine and form hydrogen gas, which leaves the cable. The hydrogen depletion has the effect of further concentrating the caustic hydroxyl ions, thus furthering the deterioration of the surface oxide. No pitting occurs during this half cycle since the aluminum ion is attracted to the metal. A caustic solution develops, hydrogen evolves, aluminum pitting takes place, and aluminum hydroxide forms during this reaction.

[0017] A critical current density is necessary to sustain the corrosion reaction. Below this current density corrosion will be very slight, or almost imperceptible. Once the current density is high enough, the reaction can be swift. The necessary current density is below 1 mA/in². The current density of a damaged 600 V cable is influenced by the voltage, leakage resistance, and the area of exposed metal. Variables affecting this can include dampness of the soil, chemistry of the soil, degree of damage, etc.

[0018] The toughest cables on the market today will not always stand up to the rigors of handling, installation, and operation. And exposed aluminum will eventually deteriorate. The solution, then, is to find a way to economically prevent the corrosion process.

[0019] Attempts have been made to prevent the ingress of moisture by introducing a sealant between the strands of the conductor and between the conductor and the insulation. See U.S. Pat. Nos. 3,943,271 and 4,130,450. However, it has been found that the mere introduction of a sealant into such spaces is not entirely satisfactory. Attempts to prevent moisture from reaching the conductor, such as using water swellable material, have not met with technical and/or economic success. For example, voids may be formed in the sealant during the application thereof or may be formed if the cable is accidentally punctured. Any such spaces or voids form locations for the ingress of moisture which can lead to corrosion of the conductor and conventional sealants used in the cables cannot eliminate such voids.

[0020] A prior art attempt to minimize the flow of moisture or water within the interstitial spaces of a stranded conductor came in the form of compacted or compressed stranded conductors. The stranded conductor itself was radially crushed in order to reduce the diameter of the conductor and to fill the interstitial spacing with metal from the individual wires themselves. The drawback to this method is that even though some deformation of the individual wires does take place, and some of the interstitial spacing is filled, there is still the possibility of cable insulation damage through which moisture can enter the cable and contact the conductor.

[0021] Another attempt at correcting moisture flowing within interstitial space consisted of filling the interstitial space with a foreign substance which physically prevented the flow of the moisture or water within the conductor structure. These substances typically comprised some type of jelly base and a polyethylene filler material. At slightly elevated temperatures, this compound becomes fluid and viscous and can be applied as the conductor is being formed. The individual wires used to form the conductor are fed into an extrusion die where the moisture blocking compound is extruded onto and around each individual wire and, as the wires are stranded into the conductor, the interstitial space is filled with the jelly-like material. Upon cooling, the filler becomes very stable and immobile and does not flow out of the interstitial spaces of the stranded conductor. Once the filling compound is applied within the interstitial spaces of the stranded conductor, it tends to remain in place. The problems encountered in applying such a filling substance revolve around precise metering of the material into the interstitial spaces as the stranded conductor is being formed. If too much material is extruded into the conductor, the outer insulation will not fit properly. If too little material is applied, the interstitial spaces will not be filled and therefore will allow moisture to flow within the conductor.

[0022] Another drawback to this method of applying a moisture blocking material is that an extrusion head and an extrusion pump for applying the material is required for every individual layer of wires used to form the conductor. The problems described above regarding the regulation of the volume of material applied through an extrusion head are multiplied every time an additional extrusion pump and extrusion head is required within the conductor manufacturing system. Prior art efforts to manufacture an acceptable moisture blocked conductor revolved around methods for uniform application of the moisture blocking material to the conductor, but did not solve the problems created by handling and installation damage.

[0023] Applications of moisture blocking material to the spacing of concentric lay conductors is known within the industry. This can be found in U.S. Pat. Nos. 3,607,487; 3,889,455; 4,105,485; 4,129,466; 4,435,613, 4,563,540; and 4,273,597.

[0024] U.S. Pat. No. 4,273,597 shows a method of strand filling the interstitial spacing of a conductor with a powder. This is accomplished by passing the strands through a fluidized powder bed, where the interstitial spacing is filled with the powder. The stranded conductor then exits the opposite end of the bed where an insulating layer is applied which prevents the powder from vacating the interstitial spacing of the conductor.

[0025] U.S. Pat. No. 4,563,540 describes a conductor which is constructed by flooding a waterproofing material among the individual conductors which make up the core of the stranded conductor. This flooded core is then wrapped with a plurality of different layers of shielding material which prevents the influx of moisture into the stranded conductor.

[0026] U.S. Pat. No. 4,435,613 describes a conductor constructed of a plurality of layers of insulating material with the core (or conducting portion) of the conductor being filled with an insulating layer of polyethylene. This polyethylene layer is contained by other rubber and plastic and epoxy compounds which produce a conductor having a waterproof construction.

[0027] U.S. Pat. No. 4,129,466 deals with a method for the application of the filling medium which is applied to a stranded conductor. This method comprises a chamber into which are passed individual wires that will be used to form the stranded conductor. These wires have a filling medium applied to them in the chamber. After the application of this filling medium, the conductor is passed through a chilling chamber where the filling medium is cooled and allowed to solidify within the interstitial spaces. This method requires that the chamber containing the filling medium and the stranded conductor be both heated and pressurized. The heat applied to the chamber reduces the viscosity of the filling material, while the pressure assures introduction of the material into the interstitial spaces of the stranded conductor.

[0028] U.S. Pat. No. 4,105,485 deals with the apparatus utilized in the '466 method patent previously discussed.

[0029] U.S. Pat. No. 3,889,455 discloses a method and apparatus for filling the interstitial spacing of the stranded conductor in a high temperature flooding tank. The individual wires are fed into a tank containing the filling material, the material having been heated to allow it to become less viscous. The individual wires are stranded and closed within the confines of the flooding tank and the finished conductor is withdrawn from the opposite end of the flooding tank where it is passed through a cooling means. The disadvantages experienced here involve the practice of stranding the conductor beneath the surface of an elevated temperature moisture block pool. No access, either visual or mechanical, to the conductor manufacturing process is practical.

[0030] U.S. Pat. No. 3,607,487 describes a method whereby individual strands of wire are fed into a flooding tank which is supplied with heated filling material by a pump and an injection means. The stranded conductor is withdrawn through the opposite end of the flooding tank, wiped in a wiping die, wrapped in a core wrapper and then passed through a binder where it is bound. The bound, wrapped core is then passed through a cooler which sets the filling material. The above described process is repeated through another flooding tank, another cooler, another binding machine, another flooding tank, another extruder, another cooling trough, and is eventually withdrawn from the end of the manufacturing line as a product having a plurality of layers of moisture blocking compound which protects the conductor core. The disadvantages here comprise a complex manufacturing line whereby moisture blocking material is applied at many different locations, each having to be meticulously monitored and controlled in order for a proper conductor construction to be obtained.

[0031] It can be readily seen from the above referenced methods and apparatuses that moisture blocked conductors are known and it can also be recognized that there are major problems concerning the elimination of moisture contacting the conductor as a result of handling and installation of a cable.

[0032] The present invention relates to improvements in insulated solid and stranded cables. An electrical cable and a method for manufacturing the electrical cable are provided in which a plurality of insulated conductors have an inner protective layer extruded thereabout, the inner protective layer having an exterior ribbed or finned surface which includes a plurality of longitudinally extending ribs or fins are at least partially filled with a sealant and an outer insulation layer formed thereabout, all applied to the conductor as a combined multilayer flow. The inner layer can be polyethylene, pvc, or another suitable plastic material. The inner layer can be cross-linked while it is being applied or batch cross-linked after it is applied. The sealant is applied using a hot melt pumping system. Known methods of pumping sealant, applying sealant, and sizing the sealant layer can be used depending on process or product requirements. The sealant can be applied over a wide range of temperatures. Good results are obtained by applying the sealant above a pressure of from about 600 to about 3000 psi and preferably from about 600 to about 1200 psi and above about 100 degrees Fahrenheit. The outer layer can be polyethylene, pvc or another suitable plastic material. The outer layer can be cross-linked while it is being applied or afterwards in a batch process.

[0033] In the present manufacturing process, the conductor is fed into a head that consists of a single multilayer flow zone. The inner layer material, the sealant layer material and the outer encapsulating layer material are merged in the head and applied in a single zone. This process requires close control of the sealant and outer material flow rates and temperatures.

[0034] In one embodiment of the invention, during manufacture of the self-sealing cable, a material which provides the cable with puncture, crack, and void self-sealing properties is included between the ribs or fins and the outer insulation. The channel regions are at least partly filled by the material which will flow into any void, puncture, or crack formed in the insulation, thus preventing migration of moisture. The self-sealing material is present in the channel regions between the ribs or fins and the outer insulation, therefore the self-sealing material does not contact the conductor.

[0035] This patent contains at least one drawing executed in color. Copies of this patent with colored drawings will be provided by the office upon request and payment of the necessary fee.

[0036] The objects and advantages of the invention will be apparent from the following detailed description of the preferred embodiments thereof in conjunction with the accompanying drawings in which:

[0037] FIG. 1 is a perspective view of a cable of the invention showing a stranded conductor, the finned inner layer surrounding the conductor, the insulation, and the area between the fins containing the material which provides the self-sealing effect;

[0038] FIG. 2 is an end view of the embodiment of the cable shown in FIG. 1;

[0039] FIG. 3 is a side view of the cable shown in FIG. 1;

[0040] FIG. 4 is a side-view showing the size and shape of the damage to the cable.

[0041] FIG. 5 shows the wooden box test layout for the damaged cable.

[0042] FIG. 6 is a graph showing the damaged cable leakage current measurements.

[0043] FIG. 7 is a graph showing the damaged cable resistance measurements.

[0044] FIG. 8 is a graph showing the damaged cable temperature measurements.

[0045] FIG. 9 shows the damaged cable and control cable after ninety-one (91) days in the test.

[0046] FIG. 10 is a close-up showing the corrosion of the control cable after ninety-one (91) days in the test.

[0047] FIG. 11 is a close-up of the cable of the present invention after ninety-one days in the test.

[0048] FIG. 12 is a cut-away side view of the multilayer flow extrusion head of the present invention; and

[0049] FIG. 13 is an end view of an embodiment of the cable of the present invention.

[0050] Although the principles of the present invention are applicable to different types of electric cables, the invention will be described in connection with a known cable structure, such as a 600 volt cable, which normally comprises, as a minimum:

[0051] (1) A central conductor of stranded wires of a good conductivity metal such as copper, aluminum, copper alloys or aluminum alloys; and

[0052] (2) A layer of insulation around the stranded conductors which has been extruded thereover.

[0053] The same numbers for like elements are used in all figures for clarity.

[0054] FIG. 1 shows a cable 11 comprising a conductor 12 of stranded wires of copper or aluminum or alloys thereof. An inner layer 14 encircles cable 11 and has a plurality of longitudinally extending fins or ribs 15 between which extend a plurality of channel regions 16. A layer 10 of material which provides the self-sealing effect at least partly fills channel regions 16 between ribs 15, inner layer 14, and an outer insulation jacket 13. Insulation jacket 13 is of known material and is preferably an extruded polymeric material.

[0055] Preferred material 10 comprises a polymer which can be readily pumped at temperatures at least as low as 25° C. Preferably, the polymer will be a low molecular weight polymer such as low molecular weight isomer. Other materials, or combinations of materials, with or without such polymers, having such characteristics may also be useful in the present invention. A polymer which has been found to be particularly suitable is polyisobutene.

[0056] The preferred polymer of the present invention has very little or no significant Shore A hardness. A test of determining whether or not the polymer has acceptable properties is the Penetrometer Test incorporated in ASTM D5 Penetration of Bituminous Materials. The 100 grams needle penetration value at 25° C. should be greater than about 100 tenths of a millimeter.

[0057] The material used to provide the self-sealing effect to the electric cable of the present invention has the following properties:

[0058] (a) The material is substantially insoluble in water;

[0059] (b) The material is a dielectric, i.e., it is non-conductive and is not a semi-conductor;

[0060] (c) The material causes the cable to be self-sealing, i.e., it will flow, at ambient temperature, into insulation voids and/or cracks and prevent contact between the conductor and moisture which could cause cable failure; and

[0061] (d) The material does not absorb moisture or swell upon contact with moisture.

[0062] In the preferred embodiment of the present invention, the material used to at least partly fill channel regions 16 is a compound of a low molecular weight isomer or a low molecular weight copolymer of an isomer. Preferably, the material is polyisobutene. Advantageously there is little or no air present between channel regions 16 and insulation jacket 13.

[0063] The material of the present invention may optionally contain filler material, but is essentially free of any solvents or oils.

[0064] The cable 11 described in connection with FIG. 1 can be used without further layers encircling the insulation jacket 13.

[0065] Also, in other embodiments of the present invention described herein, the conductor and layers of insulation can be the same as those described in connection with FIG. 1.

[0066] The cable 11 illustrated in FIG. 2 is an end view of the cable illustrated in FIG. 1.

[0067] FIG. 3 is a side view of cable 11 shown in FIG. 1 and illustrates channel regions 16 and ribs or fins 15.

[0068] FIG. 12 is a cut-away side view of the multi-layer extrusion head of the present invention showing the flow of outer layer 13, sealant layer 10 and inner layer 14 and the formation of the multi-layer thereof on conductor 12.

[0069] FIG. 13 is an end view of an embodiment of the cable of the present invention similar to FIG. 2, but showing the channel regions 16 as having curved or radius corners instead of the angled corners that are shown in FIG. 2. In FIG. 2 the angles of the corners of each channel region 16 may be the same or may be different. Likewise the curved corners 17 of each channel region 16 shown in FIG. 13 may have the same or different radius.

[0070] The ratio for the height of fins 15 to the width of channel regions 16 can vary. Advantageously, the height to width ratio ranges from about 0.1 to about 2.00. Preferably the height to width ratio ranges from about 0.25 to about 1.00. The fins do not have to be equally spaced but it is generally desirable to equally space the fins to achieve equal distribution of the medium that is in channel regions 16 and improve cable concentricity. The number of fins can range from a minimum of 2 up to any practical number that is needed based on the size of the cable, structural needs of the cable, the material being used in the channel regions, the delivery rate needed if applicable for the material, or the physical size of the material being delivered. The base thickness of the fins should not be less than about 10 percent of the width of the fins. The base thickness can vary based on thickness requirements of industry specifications, structural needs of the cable, or other specific cable needs.

[0071] The retaining mechanism between the outer encapsulating jacket or insulation and the inner finned layer is advantageously a polymeric bond between outer layer 13 fins 15, and inner layer 14. The polymeric bond should constitute bonding of at least 50% of the exposed surface area of fins 15, i.e., the upper portion of the fins that contact the interior surface of the outer extended layer 13. The upper portion of fins may intersect or contact the outer extended layer 13, may overlap into layer 13 as shown in FIG. 2, or layer 13 material may form a portion of fins 15.

[0072] Materials that can be delivered in the channel regions in addition to sealing materials are fiber optics, heat transfer fluids to enhance cable heat transfer properties, other desirable materials that would provide a beneficial cable property or use the cable as a messenger to connect a beginning and/or end point.

[0073] The most desirable materials for use as the inner layer 14, fins 15, and outer encapsulating layer 13 are plastics that can be either thermoset or thermoplastic. Known plastic materials can be used in order to achieve desired cable properties. Inner layer 14, fins 15, and outer layer 13 may be of the same or different materials depending upon the desired cable characteristics.

[0074] Advantageously fin 15 material is the same as either inner layer 14 material or outer layer 13 material. Where the inner layer 14 and outer layer 13 materials are different, fins 15 can comprise from 0 to 100 percent of either material. Preferably fins 15 are formed by partly blocking the flow of layer 10 material as the material(s) flow through the extrusion head 20.

[0075] The colors of the inner layer 14, fins 15, and outer layer 13 materials can be the same or they may differ. Different colors may be used to allow easier identification of the product in the field or for other desirable cable properties. The fins or ribs may be straight, may spiral, may oscillate about the axis of the cable, or may form different patterns depending on the desired cable characteristics and efficiency and flowability of the sealing material used.

[0076] Optionally a branched flow of material may be applied to the conductor before the combined multilayer flow is applied. This may be used when it is desired to fill the interstices of the conductor to provide a smoother cable surface.

[0077] It is to be understood that additional embodiments may include additional layers of protective material between the conductor and the insulation jacket as additional layers between the conductor and the self-sealing material and/or between the self-sealing material and the outer layer, including an additional water barrier of a polymer sheet or film, in which case it is not essential that the jacket tightly enclose the layers there within or enter into the spaces between the wires and protective materials, i.e., the interior size of the jacket can be essentially equal to the exterior size of the elongated elements so that compression of the elongated elements, and hence, indentation of the layers there within including the insulation, is prevented.

[0078] The cable of the present invention is of particular advantage in that not only does the material fill the space between the inner layer and the insulation as the cable is manufactured, but after the cable is placed in service the material will flow into any cuts or punctures formed as a result of damage during handling and installation of the cable or its use in service. The stresses placed on the conductor and the insulation during handling and installation of the cable, such as bending, stretching, reeling and unreeling, striking with digging and installation equipment can form cuts or punctures in the insulation and between the insulation and the conductor. Such cuts or punctures can also be formed after the cable has been placed in service as a result of damage from adjacent utilities, homer owners, or lightening strikes.

[0079] The cable of the present invention can provide acceptable service even after the insulation has been cut or punctured, exposing the conductor. In order to determine the efficiency of the present inventive cable defects were made in the insulation layer of two 600 V cable samples. On one of the cable samples, a layer of polyisobutene polymer was applied in accordance with the present invention. The other cable sample did not have a polyisobutene layer. Both cable samples were placed inside separate 1 liter glass beakers containing tap water. Each cable sample was energized at 110 V to ground with AC current. The sample which did not have a polyisobutene layer exhibited severe corrosion overnight. The sample containing the polyisobutene layer exhibited no corrosion after being energized and submerged for 4 weeks in tap water in the glass beaker.

[0080] This test was designed to evaluate the performance of the present invention's self sealing, 600 V underground cable. The test program was patterned after a previously developed procedure to evaluate self-sealing or self-repairing cable designs.

[0081] To conduct the test damaged cables were placed in a specially mixed, moist soil. The cables were then energized with 120 V ac to ground. Measurements made included changes in leakage current to earth and cable conductor resistance. The temperature of each cable near the damage point was also monitored.

[0082] Four control sample replicates and eight self-sealing sample replicates were evaluated. All four control samples failed the test relatively early in the test program. All eight self-sealing samples performed well, with no significant increase in conductor resistance and low leakage current values throughout the 60-day test period.

[0083] Conventional and self-sealing 600 volt underground cable with a 2/0 AWG combination unilay aluminum conductor were tested in 10-foot lengths.

[0084] The soil used in the test was a mixture of Ottawa Sand, Wyoming Bentonite and fertilizer. The combination of the three materials provides a sandy-silt type soil, which is very conductive. The sand serves as the basic soil structure while the silt provides small particles that can work their way into the damaged areas of the cable. The silt also helps to keep water evenly dispersed throughout the soil. The fertilizer enhances the conductivity of the soil and may enhance corrosion as well. The goal was to achieve a soil electrical resistivity of <50 ohmmeters.

[0085] Tap water was used to achieve a moisture content near saturation. This combination of soil materials provides a worst case condition for the ac corrosion of the aluminum conductor in 600 V underground cables and is also repeatable from lab to lab.

[0086] The soil mixture was:

[0087] 100 lbs. Ottawa Sand

[0088] 3.33 lbs. Bentonite

[0089] 23.33 lbs. Tap Water

[0090] 1.26 lbs. of Peters 20-20-20 Plant Fertilizer (mixed with the water before added to the sand and clay ingredients)

[0091] The amount of water added achieved near saturation conditions. The wet density was approximately 127 lbs./ft.

[0092] The aging box was made of wood and lined with polyethylene to hold moisture. The approximate inside dimensions were 6.5 feet long by 1.3 feet wide by 1 foot high. A wide, copper tape ground electrode covered the bottom and sides of the box on top of the polyethylene. A wire connected this electrode to ground.

[0093] After moist soil was packed in the bottom of the box (approximately 6 inches), four control samples and eight self-sealing samples were installed, approximately six inches apart.

[0094] The two sample sets were:

[0095] Samples 1-4: conventional 600 V UD wire (control samples) all with slot damage at the center of the sample

[0096] Samples 5-12: self-sealing cable—all with slot damage near the center of the sample

[0097] Immediately before the samples were placed in the box, they were damaged down to the conductor. One damage condition was used. It consisted of a slot cut into the insulation down to the conductor, perpendicular to the cable axis. A razor knife and an angle guide was used to control the slot size. The size and shape of the damage location is shown in FIG. 4. The damage locations were staggered so they were not adjacent to each other.

[0098] The 10-foot long self-sealing samples were first damaged in the middle. After 5 minutes, they were placed in the box with the damage facing up. They were then covered with soil.

[0099] The control samples were initially 2.5-foot long. They were also damaged in the middle, then installed in the box. There was no waiting period before they were covered with soil.

[0100] As each sample was installed, a type T thermocouple with a welded bead was attached to the cable surface, approximately one inch from the damage location. Once all samples were installed, the soil was compacted. After 24 hours, the ends were cut off of the self-sealing samples so they were the same length as the control samples. The test layout is shown in FIG. 5.

[0101] After the installation was complete, the soil was covered with polyethylene to minimize the evaporation of water from the soil. 120 V ac was applied continuously to all sample conductors. The soil was grounded via the copper ground mat lining the tank. The data collection was as follows:

[0102] 1) Measurements (Measured initially, then daily for first 5 workdays, then on Monday, Wednesday and Friday of each week thereafter.)

[0103] a) Conductor resistance, each sample individually—Biddle DLRO, CQ #1010 (Expected accuracy: ±3% of reading)

[0104] b) Leakage to ground @120 V, each sample individually—Fluke 87, CN 4007 (Expected accuracy: ±3% of reading)

[0105] c) Sample surface temperature—Yokaggawa DC100, CN 4015 (Expected accuracy: ±2 Deg. C.)

[0106] 2) The test ran for 91 days. When significant degradation occurred on a sample, it was disconnected from the voltage source. Significant degradation is defined as:

[0107] a) Several days with leakage current greater than 1 amp on an individual sample

[0108] b) Conductor resistance on an individual sample 10 times greater than starting resistance

[0109] 3) Final soil electrical resistivity and moisture content was measured when the test was completed

[0110] 4) All measurements were recorded and resistance, leakage and temperature data were plotted using an Excel spreadsheet

[0111] During the first 26 days of the test the conductor resistance and the leakage current into the soil increased significantly on all four control samples. They were each removed from the test (disconnected from the test voltage) as the conductor resistance exceeded 1,000 micro-ohms.

[0112] The conductor resistance and the leakage current to the soil for the eight self-sealing samples did not change significantly during the test.

[0113] The soil electrical resistivity was measured at the end of the test by placing a sample of the soil in a 17-inch long, 2-inch inside diameter PVC tube. It was packed to the same density used in the test tank. Two-inch diameter copper plate electrodes were pressed against the soil on each end of the tube. 120 volts ac was applied across the electrodes and the resulting current was measured. The current and voltage were used to calculate the sample resistance, which was then converted to resistivity.

[0114] Moisture content and density were measured at the beginning and end of the test. To make the measurement, a soil sample was taken using a {fraction (1/30)} cubic foot metal shelby tube. The sample was then oven dried to calculate moisture and density. The measured weights were used to calculate moisture content.

[0115] Soil resistivity, moisture and density measurements are summarized in Table 1.

[0116] The insulation resistance, conductor resistance and sample temperature measurements made during the test are shown in FIGS. 6-8. The samples are identified as S1, S2, S3, etc. The first four are control, the remaining eight are self-sealing. In addition, C=Control, SS=Self-Sealing.

[0117] During periods of relatively high leakage current on the control samples the temperature of these samples was also relatively high. Photos of the samples under test are shown in FIGS. 5, 6 and 7. From the photos it is obvious that the control samples experienced significant corrosion while the self-sealing samples experienced no noticeable corrosion.

[0118] A cyclic load test was run on the finned cable of the present invention and compared with similar non-finned prior art cables. 50 ft. samples were tested. The samples had a 50° C. conductor temperature, and were cycled on 8 hours a day and off 16 hours, 7 days a week. The cables were terminated with a mechanical connector. No duct seal, mastic tape, electrical tape, or the like was used. The tops of the samples were approx. 11 ft. above the floor. The samples gradually droop to the floor.

[0119] Over 80% of the total shrinkback of the prior art cable occurred in the first week of testing.

[0120] Comparative results with the present invention show a dramatic reduction in shrinkback after 1 week of testing. The reduction is more than 92% when compared with the prior art.

[0121] Although preferred embodiments of the present invention have been described and illustrated, it will be apparent to those skilled in the art that various modifications may be made without departing from the principles of the invention.