This invention is related to surge protection devices, and more particularly, overvoltage protection devices, that employ metal oxide varistors (MOVs) in station protectors, central office overvoltage protection devices, on-line overvoltage protection devices, and remote terminal overvoltage protection devices. In particular, this invention is related to the use of MOVs in hybrid overvoltage protection devices designed for higher frequency digital subscriber line (DSL) transmissions.
Conventional overvoltage protection devices typically use a gas discharge surge arrestor, or “gas tube,” as a primary means for diverting voltage surges from a signal line to ground. Examples of such devices are shown in U.S. Pat. Nos. 5,388,023, 5,500,782 and 5,880,919. Gas tubes dissipate energy by causing electrical arcing to ground. A gas of known dielectric strength is ionized when subjected to an electrical surge. One drawback of gas tubes, however, is that they typically exhibit a relatively slow response time, and thus, may not be able to safely suppress fast rise time voltage surges. Accordingly, metal oxide varistors (MOVs) have been employed as secondary protectors in back-up and interacting overvoltage protection devices. For example, in a conventional hybrid station protector, an MOV is electrically connected in parallel with the gas tube between each signal line and electrical ground. Although the gas tube can repeatedly dissipate voltage surges without damage, the response time of the MOV is substantially faster than the response time of the gas tube. Therefore, the MOV can be relied upon to shunt fast rise time voltage surges to ground, while the gas tube is relied upon to shunt sustained voltage surges, which might otherwise damage the MOV.
Overvoltage protection devices utilizing MOVs as secondary protectors have been successfully employed to protect conventional twisted-pair (i.e., “tip” and “ring”) telephone lines. Broadband communications, such as digital subscriber line (DSL) transmissions, which are generically referred to herein as “xDSL”, operate at transmission frequencies that are substantially higher than the frequencies traditionally employed over twisted-pair telephone lines. Presently, frequencies of at least 1 megahertz, and generally about 30 megahertz, are utilized for xDSL communications transmitted over twisted-pair telephone lines. Existing twisted-pair telephone lines, also referred to as outside plant wire, are typically CAT-3 grade or less and were not intended for high frequency performance when originally manufactured or installed. In many instances, conventional overvoltage protection devices are inadequate for higher frequency digital transmissions, for example VDSL. Even if only a small number of overvoltage protection devices perform inadequately, the cost of identifying and replacing the overvoltage protection devices that may be adequate for lower frequency xDSL communications, but inadequate for higher frequency xDSL communications, is significant.
The inadequate performance of some conventional overvoltage protection devices, such as station protectors utilized at customer premises for higher frequency xDSL communications, has been traced to the relatively high capacitance and the variability of the capacitance of the MOVs that are employed in the station protector. At higher frequencies, the capacitance and the variability of the capacitance results in unacceptable insertion loss, return loss, and longitudinal imbalance. It is well known that the capacitance can be reduced by utilizing MOVs of the same thickness, but having a smaller diameter. Many conventional station protectors employ 5 mm diameter MOVs with 3.8 mm electrodes on either side of the varistor material instead of smaller diameter MOVs because the larger diameter MOVs absorb additional energy without permanent damage. MOVs of this size with symmetrical electrodes have a capacitance of about 60 picofarads with a tolerance of about 20% (i.e., 60 picofarads ±12 picofarads). This relatively large tolerance is believed to be due to variability in the varistor material and thickness, and/or to the relative placement and size of the electrodes on opposite sides of the varistor material. Electrodes, which are intended to be aligned on opposite sides of the varistor material, can in practice be laterally displaced relative to each other. The concentricity of the two electrodes can also vary. Lateral displacement and varying concentricity of the electrodes on opposite sides of the varistor material means that the overlapped surface area of the electrodes can vary significantly between MOVs that are intended to be identical, thereby generating dissimilar electric fields that result in relatively high capacitive tolerance. The difference in the capacitances of the MOVs utilized between the tip conductor and ground and between the ring conductor and ground results in significant capacitance mismatch, referred to herein as “capacitive imbalance.” In turn, excessive capacitive imbalance can cause unacceptable signal loss (e.g., insertion loss and return loss) and longitudinal imbalance at the higher frequencies utilized for xDSL communications transmitted over twisted-pair telephone lines.
As previously mentioned, it would be possible to reduce the capacitance between a signal line and ground in a station protector by utilizing an MOV of the same thickness having a smaller diameter. Because the electrodes of the smaller diameter MOV inherently have a smaller overlapped surface area, the smaller diameter MOV also has less capacitance. However, a smaller diameter MOV is not able to withstand the same sustained current as a larger diameter MOV. Furthermore, substitution of the smaller diameter MOV would result in significant engineering, re-tooling and testing expense. Even if the desired reduction in capacitance could be achieved by substituting a smaller diameter MOV for the 5 mm MOV presently in use, there could still be an excessive capacitive imbalance between the tip conductor and ground and the ring conductor and ground. Accordingly, it would be preferable if a reduction in the capacitive imbalance could be achieved without the need for extensive modifications to the present design of existing station protectors.
The number of station protectors and other overvoltage protection devices manufactured that are incapable of adequate performance at higher frequencies is significant, at least in the aggregate. In particular, when MOVs having relatively high capacitance and large capacitive tolerance are employed in twisted-pair telephone lines, an excessive imbalance in the capacitance between the tip conductor to ground signal line and the ring conductor to ground signal line will be present in a significant number of station protectors. In fact, a final inspection rejection rate as much as 10% is not uncommon. The capacitive imbalance for such station protectors has been found to be up to about 5 picofarads. For xDSL communications, a capacitive imbalance of less than about 1.3 picofarads is desired. Accordingly, what is needed is an overvoltage protection device in which the capacitive imbalance due to the capacitive tolerance of the MOVs utilized in the device is reduced, but in which the same current can be sustained without permanent damage to the MOVs. Such overvoltage protection devices, for example station protectors, could then be utilized without introducing an excessive capacitive imbalance in twisted-pair telephone lines that transmit higher frequency xDSL communications.
According to this invention, overvoltage protection devices for use with twisted-pair telephone lines can be manufactured such that the capacitive imbalance between the signal lines and electrical ground introduced by the overvoltage protection device does not adversely affect higher frequency transmissions, such as xDSL communications. These overvoltage protection devices, such as station protectors for interconnecting subscriber equipment at customer premises to a telephone network, can be assembled by a method including the following steps. A group of MOVs, each protecting against a surge of the same value, will exhibit levels of capacitance differing from other MOVs in the group by an amount greater than a predetermined first value of capacitive imbalance to be maintained by the overvoltage protection device. This large capacitive tolerance is due to manufacturing limitations and MOVs of the same type with the same rating will exhibit this variability. These MOVs are then sorted into subgroups, each subgroup consisting only of MOVs with capacitances differing by a predetermined second value, no greater than the predetermined first value. Individual overvoltage protection devices, such as station protectors, are then assembled using MOVs selected only from the same subgroup so that the capacitive imbalance of the station protector will not exceed the predetermined first value. A plurality of overvoltage protection devices assembled from MOVs selected from the same subgroup will still have a capacitive imbalance that does not exceed the predetermined first value, even though the capacitances of the MOVs in different overvoltage protection devices may differ significantly. Using this method, the remaining MOVs manufactured in the group of MOVs can be combined with MOVs manufactured in other groups to assemble station protectors despite the large capacitive tolerance exhibited by each group of MOVs.
According to another aspect of this invention, asymmetrical MOVs, each with electrodes having different surface areas, are also sorted. Such asymmetrical MOVs will necessarily have a lower capacitance and will exhibit less variability in capacitance. Thus, asymmetrical MOVs manufactured as part of the same production run can be sorted more efficiently than symmetrical MOVs of the same production run because of their smaller capacitive tolerance. Station protectors can then be manufactured using MOVs only from the same sorted subgroup and from the same production run. The capacitive imbalance of station protectors manufactured in this way will have both a reduced capacitance and a reduced capacitive imbalance.
Preferred embodiments of the invention shown and described herein include surge protection devices, and in particular overvoltage protection devices, that are used at the interface between a telecommunications network and customer premises on twisted-pair telephone signal lines comprising conventional tip and ring conductors. Overvoltage protection devices, commonly referred to as station protectors, protect personnel and telecommunications equipment from voltage surges and overvoltage transients by shunting the voltage surges and transients to electrical ground. However, overvoltage protection devices according to the invention are not limited to station protectors, which are shown and described herein only as exemplary embodiments. Furthermore, other surge protection devices, such as central office overvoltage protection devices, on-line overvoltage protection devices, and remote terminal overvoltage protection devices can also benefit from the reduced capacitance and reduced capacitive imbalance achieved by this invention.
The present invention can be utilized on twisted-pair telephone lines in hybrid overvoltage protection devices that employ gas discharge tubes and MOVs electrically connected in parallel between the tip and ring conductors and a common electrical ground. Each gas tube and MOV provides an alternative electrical path to ground. The gas tube is often considered to be the primary electrical path between the corresponding signal line (e.g., tip or ring conductor in the case of twisted-pair telephone lines) and ground, because the gas tube is able to repeatedly withstand high current surges. However, because of the time required to ionize the gas in the gas tube, the response time of the MOV is significantly faster than the relatively slow response time of the gas tube. On the other hand, sustained or repeated currents tend to damage the MOV. In hybrid overvoltage protection devices employing interacting varistors, the parallel combination of the gas tube and the MOV permit the overvoltage protection device to respond to fast rise time surges or transients because the MOV will divert voltage surges to ground until the gas tube fires. The gas discharge tube provides the primary electrical path to ground and protects the MOV because the MOV will not be subjected to sustained or repeated high currents. In other hybrid overvoltage protection devices, an MOV having a much larger DC breakdown voltage than the gas tube provides back-up surge protection in case of damage to the gas tube (e.g., venting). This invention can be employed with either interacting or back-up type hybrid overvoltage protection devices.
An example of a hybrid overvoltage protection device in the form of a station protector comprising a protector assembly employing a gas discharge tube and an MOV is shown in FIG. 1. The gas tube and the MOV are electrically connected in parallel between a signal line and ground. The protector assembly shown in
As shown in
The primary protector in the protector assembly 4 shown in
An MOV spring 18 and a failsafe contact 22 formed from conductive materials retain the gas discharge tube 14 and the MOV 20. The MOV 20 is positioned between a top portion 19 of the MOV spring 18 and a central web 21 on the failsafe contact 22. The gas discharge tube 14 is positioned between the central web 21 and the bottom portion 17 of the MOV spring 18. The MOV spring 18 urges the MOV 20 toward the central web 21 of the failsafe contact 22. The electrodes on the opposite sides of the MOV 20 engage the lower surface of the top portion 19 of MOV spring 18 and the upper surface of the central web 21 of failsafe contact 22. However, the composition of the varistor material prevents conduction through the MOV 20 until the signal line is subjected to a voltage surge of sufficient energy for the varistor material to conduct and shunt the voltage surge to ground. A fusible member 16 is positioned between the central web 21 of the failsafe contact 22 and the gas discharge tube 14. The fusible member 16 is normally formed of a eutectic solder so that it will rapidly soften and flow when subjected to a predetermined temperature. However, the fusible member 16 may be made of any other suitable material that softens and flows sufficiently to permit the central web 21 of failsafe contact 22 to move in the direction of the gas discharge tube 14. Once the fusible member 16 softens and flows, the MOV spring 18 urges the failsafe contact 22 in the direction of the gas discharge tube 14 so that at least one leg 23 of failsafe contact 22 electrically contacts the terminal 6 and shorts the protector assembly 4 to the ground terminal 24 through ground springs 8 and pin 10, thereby diverting the voltage surge to electrical ground.
Since the MOV 20 is not conductive under normal circumstances, there will be a capacitance introduced primarily by the MOV 20 between the tip conductor and ground and between the ring conductor and ground. If these capacitances differ, there will be a capacitive imbalance that exists between the electrical path from the tip conductor to ground and the electrical path from the ring conductor to ground. Capacitive imbalance is typically not problematic at lower frequencies, but results in unacceptable signal loss (e.g., insertion loss and return loss) and longitudinal imbalance at higher frequencies. In particular, excessive capacitive imbalance results in inadequate performance at higher frequencies in a number of existing station protectors.
Overvoltage protection devices according to this invention are not limited to the station protector shown in
When the overvoltage protection device is to be used to form separate electrical paths to ground for two conductors, such as the tip and ring conductors of a twisted-pair telephone line, capacitive imbalance becomes an issue. In station protectors used to shunt voltage surges from the tip conductor to ground and/or from the ring conductor to ground at the network interface to the customer premises, separate MOVs are used between tip and ground and between ring and ground. If the capacitances of these two MOVs differ, capacitive imbalance is introduced between the tip and ring conductors.
MOVs of the type used in overvoltage protection devices, such as station protectors, statistically exhibit a large capacitive tolerance. One reason for the relatively large capacitive tolerance of MOVs is misalignment of the electrodes on opposite sides of the varistor material. Misalignment can occur when the electrodes are laterally displaced, or when one or both of the electrodes is misshapen (e.g., not concentric). The electric field generated between the electrodes will not be the same when the electrodes are misaligned as when the electrodes are properly aligned because the overlapped surface area of the electrodes will be different. Since capacitance is a function of the electric field generated in a nonconductive material between two spaced apart, conductive electrodes, the capacitance will also be a function of the relative alignment of the two electrodes. The varistor material will be nonconductive under normal conditions. Therefore, the capacitance introduced by an MOV is a function of the electric field generated between two electrodes that cannot be positioned relative to each other consistently with a level of precision that does not adversely affect higher frequency transmissions, such as xDSL communications. The exact placement of the electrodes on opposite sides of the MOV cannot be adequately controlled by conventional manufacturing techniques.
A 5 mm, 230V MOV with symmetrical electrodes has a capacitance of about 60 picofarads. However, the capacitive tolerance of these MOVs is typically 20%, meaning that the capacitance can range from about 48 picofarads to about 72 picofarads for MOVs that afford the same surge protection (i.e., 60 picofarads ±12 picofarads). This large tolerance is believed to be attributable primarily to manufacturing difficulties that lead to misalignment of the electrodes on opposite sides of the MOV. The difference in capacitance of an MOV between tip and ground and of an MOV between ring and ground, however, must be significantly less than the above range for the station protector to not adversely affect xDSL transmissions over a twisted-pair telephone line. It has been determined empirically that improved xDSL performance can be achieved if the capacitive imbalance introduced between tip to ground and ring to ground by a station protector does not exceed about 1.3 picofarads. It has also been determined that if MOVs are sorted into subgroups in which the capacitance of any two MOVs within that subgroup do not vary by more than 1.0 picofarad (i.e., the difference in capacitance between any two MOVs in the subgroup is no greater than 1.0 picofarad), and if both MOVs used in a station protector are selected from a common subgroup, than the capacitive imbalance introduced by the station protector will not exceed 1.3 picofarads. Even though this procedure will not necessarily reduce the amount of capacitance introduced between the tip conductor and ground or between the ring conductor and ground, nevertheless the reduction in capacitive tolerance (and therefore, capacitive imbalance) will result in a desirable improvement in the performance of the station protector for higher frequency xDSL communications.
Assembly of station protectors in accordance with a first aspect of the invention includes the steps illustrated in FIG. 9. The capacitances of a plurality of manufactured MOVs can be readily determined by known means prior to incorporation of the MOVs into station protectors. In accordance with a preferred embodiment of this invention, MOVs forming a group of otherwise identically manufactured MOVs are sorted into subgroups in which the capacitances of any two MOVs within the subgroup do not vary by more than 1.0 picofarad. In other words, the difference between the capacitance of the MOV within the subgroup having the greatest capacitance and the capacitance of the MOV within the subgroup having the least capacitance is less than or equal to 1.0 picofarad. In
Although improved performance can be achieved by sorting MOVs in the above manner, at least two factors limit the practicality of utilizing the method. First, even though sorting MOVs into subgroups leads to a reduction in the capacitive imbalance in an overvoltage protection device, there is typically no reduction in the magnitude of the capacitances introduced between the tip conductor and ground or between the ring conductor and ground. A reduction in the magnitude of the capacitances would lead to an even further improvement in the performance of a station protector for higher frequency transmissions, such as xDSL communications, and particularly for very high speed digital subscriber lines (VDSL) communications. Capacitances of about 30 picofarads instead of the current nominal value of about 60 picofarads for standard 5 mm MOVs would be very desirable for higher frequency transmissions, and especially VDSL communications.
A second limitation on the degree of improvement that can be achieved by sorting manufactured MOVs to reduce capacitive imbalance is economic in nature. Sorting MOVs into subgroups in which the difference in capacitance is no greater than 1.0 picofarad to achieve improved performance at higher frequencies is relatively costly when the tolerance in capacitance is as much as 24 picofarads, as is the case of 5 mm MOVs with a nominal capacitance of about 60 picofarads and a tolerance of about ±20%. Indeed, it has been suggested by some manufacturers that it is economically impractical to sort MOVs having such a large tolerance of capacitance in this manner. Even if the variation in capacitance for a given production run is considerably less than 24 picofarads, it is still difficult and costly to sort MOVs having a large variation in capacitance into a plurality of subgroups each confined to a range of only 1.0 picofarad. For a given production run, it is reasonable to anticipate that the MOVs fabricated during that same run will exhibit a capacitance predominately clustered about a mean value somewhere within the 20% tolerance range. However, it is quite difficult to anticipate that mean value in advance, and experience has shown that the mean value can be expected to vary significantly between production runs, even without intervening tooling changes. Furthermore, the distribution (i.e., the standard deviation of the capacitance) has been found to be so large that MOVs fabricated in the same production run will rarely, if ever, be limited to a single subgroup. For instance, it has been determined empirically that approximately 20% of the MOVs manufactured in a given production run cannot be included in the same subgroup as the remaining MOVs manufactured in that same production run. Accordingly, those MOVs must be discarded or put aside for use in subgroups compiled from other production runs Experience has also shown that station protectors containing two or more MOVs from the same production run that are not sorted as described above, result in a final inspection rejection rate of between about 7% and about 10%. This rejection rate of production station protectors is economically unacceptable and clearly exceeds the additional expense of sorting the MOVs from each production run.
The further limitations described above can be overcome in accordance with a second aspect of this invention. The magnitude of the capacitance of each MOV can be lowered by manufacturing MOVs with a smaller electrode on one side of the varistor material and a larger electrode on the opposite side of the varistor material. Such MOVs are referred to herein as “asymmetrical MOVs” or “MOVs with asymmetrical electrodes.” The distribution of the capacitance of asymmetrical MOVs is sufficiently reduced such that the MOVs manufactured in a given production run can be sorted into a practical number of subgroups in which the capacitances of any two MOVs vary by no more than 1.0 picofarad. Furthermore, the standard deviation of the capacitance for a group of asymmetrical MOVs is less than the standard deviation of the capacitance for a group of symmetrical MOVs. Thus, a greater percentage of the asymmetrical MOVs produced in a given production run will satisfy the 1.0 picofarad requirement. Assuming the material, thickness and overall diameter of the asymmetrical MOVs is the same as the symmetrical MOVs, the asymmetrical MOVs will be capable of handling a sustained voltage surge of the same electrical energy.
MOVs with asymmetrical electrodes will both reduce capacitance and will have less variability in capacitance (referred to herein as capacitive tolerance) than MOVs having symmetrical electrodes.
In one embodiment, an asymmetrical MOV 20 has a diameter of about 5 mm, and the larger electrode 46, which covers substantially the entire surface of the side 42, also has a diameter of about 5 mm. As previously mentioned, MOVs of the type used in conventional station protectors typically have a diameter of only about 3.8 mm. Since the larger electrode 46 covers substantially the entire side 42 of the varistor material, it will completely overlap the surface area of the smaller electrode 48, regardless of its size, shape or lateral placement on side 44.
Furthermore, the capacitance of the MOV can be varied by utilizing asymmetrical electrodes having different surface areas. For example, the capacitance of the MOV can be further reduced by reducing the surface area of the smaller electrode since the size and shape of the electric field generated between the two electrodes will be dependent upon the surface area of the smaller electrode, and will be relatively independent of the surface area of the larger electrode. Thus, use of an asymmetrical MOV with a smaller electrode overlapped by a larger electrode will further reduce both the capacitance of each MOV and the capacitive imbalance introduced in a station protector between the tip conductor and ground and between the ring conductor and ground as a result of the use of MOVs from the same subgroup.
Although the capacitive tolerance of MOVs can be reduced by manufacturing 5 mm, 230V MOVs with asymmetrical electrodes, it has been found that the variability of the capacitance is still too great to reliably produce station protectors that have a capacitive imbalance no greater than about 1.3 picofarads without sorting the MOVs prior to assembly. However, the capacitive tolerance for a given production run of asymmetrical MOVs is significantly less than the capacitive tolerance for a given production run of symmetrical MOVs, at least in part because of the absence of electrode misalignment. As a result of the smaller capacitive tolerance, it has been determined that it is both feasible and practical to sort asymmetrical MOVs into subgroups wherein the difference in the capacitances of any two MOVs is no greater than 1.0 picofarad.
Station protectors 2 employing 5 mm, 230V asymmetrical MOVs that have a capacitance no greater than about 30 picofarads and a capacitive imbalance no greater than about 1.3 picofarads can be reliably and economically produced by another method according to the invention illustrated in FIG. 10. For each production run of asymmetrical MOVs, the MOVs are sorted into subgroups in which no MOV has a capacitance that differs by more than 1.0 picofarad from any other MOV within the same subgroup. The number of subgroups for each continuous production run will be relatively small because of the lower variability of capacitance exhibited by asymmetrical MOVs, and because relevant operational factors, including environmental factors and process variables, should remain substantially constant during the same run. These relevant operational factors can be expected to vary somewhat among different production runs, for example on subsequent days, even though no tooling changes may have been made between successive runs. Therefore, as used herein a continuous production run means a production cycle during which operational factors relevant to the capacitance of an MOV can be expected to not exhibit significant change. Typically the length of such a continuous production run is known from experience with the manufacture of conventional 5 mm, 230V symmetrical MOVs. It will also be possible to readily identify appropriate subgroups based on a small number of capacitance values. Statistically, virtually all of the asymmetrical MOVs manufactured in the same continuous production run can be divided into no more than two subgroups in which the difference in the capacitances of any two MOVs does not exceed 1.0 picofarad. If desired, the very few asymmetrical MOVs having capacitances that do not fit within this small number of subgroups can be discarded. At the end of each production run, the sorted subgroups of asymmetrical MOVs can be packaged as a unit and transported to the location where the station protectors 2 will be assembled.
Station protectors 2 are then assembled in a conventional manner, except that any two asymmetrical MOVs used in the same station protector 2 must be taken from a single sorted subgroup that contains only asymmetrical MOVs preferably manufactured in the same continuous production run. For the sake of consistency, MOVs manufactured during one production run generally are not sorted with MOVs manufactured during a different production run. This is primarily due to the fact that the 1.0 picofarad range of capacitances for the MOVs is separately determined for each production run. Thus, even though the capacitive tolerance (i.e., 1.0 picofarad) may be the same for all subgroups, the average (i.e., statistical mean) capacitance of the MOVs is likely to be somewhat different. Accordingly, the maximum and minimum capacitances within each subgroup can be unique for each production run. By segregating subgroups according to specific production runs, it becomes relatively simple to insure that the difference in capacitance between MOVs employed in the same station protector 2 will be less than 1.0 picofarad, and thus, that the capacitive imbalance between the tip conductor to ground and the ring conductor to ground circuits within the same station protector 2 will be no greater than 1.3 picofarads. The use of asymmetrical MOVs will also produce station protectors 2 with tip conductor and ring conductor capacitances no greater than about 30 picofarads. The resulting combination of reduced capacitance and reduced capacitive imbalance permits the fabrication of station protectors 2 with improved performance for higher frequency transmissions including xDSL communications, for example VDSL communications.
This application is a continuation-in-part of U.S. patent application Ser. No. 10/304,150, filed Nov. 26, 2002, and assigned to the assignee of the present invention.
Number | Name | Date | Kind |
---|---|---|---|
4802055 | Beckerman | Jan 1989 | A |
4931901 | Heron, Jr. | Jun 1990 | A |
5142430 | Anthony | Aug 1992 | A |
5557065 | Brower et al. | Sep 1996 | A |
5623388 | Chaudhry | Apr 1997 | A |
5854730 | Mitchell et al. | Dec 1998 | A |
5880919 | Napiorkowski et al. | Mar 1999 | A |
6188557 | Chaudhry | Feb 2001 | B1 |
6195245 | Kobsa | Feb 2001 | B1 |
6226166 | Gumley et al. | May 2001 | B1 |
6327129 | Oertel et al. | Dec 2001 | B1 |
6421218 | Vo et al. | Jul 2002 | B1 |
6421220 | Kobsa | Jul 2002 | B2 |
6680839 | Napiorkowski | Jan 2004 | B2 |
6785110 | Bartel et al. | Aug 2004 | B2 |
Number | Date | Country | |
---|---|---|---|
20040100747 A1 | May 2004 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 10304150 | Nov 2002 | US |
Child | 10322112 | US |