Implantable medical device having flat electrolytic capacitor with porous gas vent within electrolyte fill tube

[0001] This application claims priority and other benefits from U.S. Provisional Patent Application Ser. No. 60/080,564, filed Apr. 3, 1998, entitled FLAT ALUMINUM ELECTROLYTIC CAPACITOR.

[0002] This application is a continuation-in-part of U.S. patent application Ser. No. 09/103,843 now U.S. Pat. No. 6,157,531 filed Jun. 24, 1998, in the names of Mark K. Breyen et al. and entitled IMPLANTABLE MEDICAL DEVICE HAVING FLAT ELECTROLYTIC CAPACITOR WITH LIQUID ELECTROLYTE FILL TUBE.

[0003] This invention relates to implantable medical devices (IMDs) and their various components, including flat electrolytic capacitors for same, and methods of making and using same and providing for outgassing of gases released during capacitor charge and discharge cycles.

[0004] As described in the above-referenced parent application Ser. No. 09/103,843, and the provisional application that it claims priority from, a wide variety of IMDs are known in the art. Of particular interest are implantable cardioverter-defibrillators (ICDs) that deliver relatively high energy cardioversion and/or defibrillation shocks to a patient's heart when a malignant tachyarrhythmia, e.g., atrial or ventricular fibrillation, is detected. Current ICDs typically possess single or dual chamber pacing capabilities for treating specified chronic or episodic atrial and/or ventricular bradycardia and tachycardia and were referred to previously as pacemaker/cardioverter/defibrillators (PCDs). Earlier developed automatic implantable defibrillators (AIDs) did not have cardioversion or pacing capabilities. For purposes of the present invention ICDs are understood to encompass all such IMDs having at least high voltage cardioversion and/or defibrillation capabilities.

[0005] Generally speaking, it is necessary to employ a DC—DC converter within an ICD implantable pulse generator (IPG) to convert electrical energy from a low voltage, low current, electrochemical cell or battery enclosed within the IPG housing to a high voltage energy level stored in one or more high energy storage capacitor, as shown for example, in commonly assigned U.S. Pat. No. 4,548,209. The conversion is effected upon confirmation of a tachyarrhythmia by a DC—DC “flyback” converter which includes a transformer having a primary winding in series with the battery and a secondary winding in series with the high energy capacitor(s) and an interrupting circuit or switch in series with the primary coil and battery that is periodically opened and closed during a charging cycle. Charging of the high energy capacitor is accomplished by inducing a voltage in the primary winding of the transformer creating a magnetic field in the secondary winding when the switch is closed. The field collapses when the current in the primary winding is interrupted by opening the switch, and the collapsing field develops a current in the secondary winding which is applied to the high energy capacitor to charge it. The repeated interruption of the supply current charges the high energy capacitor to a desired level of several hundred volts over a charging time of the charge cycle. Then, the energy is rapidly discharged from the high voltage capacitor(s) through cardioversion/defibrillation electrodes coupled to the IPG through ICD leads and arranged about or in a heart chamber or vessel if the tachyarrhythmia is confirmed as continuing at the end of the charge time. The cardioversion/defibrillation shocks effected by discharge of such capacitors are typically in the range of about 25 to 40 Joules. The process of delivering cardioversion/defibrillation shocks in this way may be repeated if an earlier delivered cardioversion/defibrillation shock does not convert the tachyarrhythmia to a normal heart rhythm.

[0006] Energy, volume, thickness and mass are critical features in the design of ICD pulse generators that are coupled to the ICD leads. The battery(s) and high voltage capacitor(s) used to provide and accumulate the energy required for the cardioversion/defibrillation shocks have historically been relatively bulky and expensive. Presently, ICD IPGs typically have a volume of about 40 to about 60 cc, a thickness of about 13 mm to about 16 mm and a mass of approximately 100 grams.

[0007] It is desirable to reduce the volume, thickness and mass of such capacitors and ICD IPGs without reducing deliverable energy. Doing so is beneficial to patient comfort and minimizes complications due to erosion of tissue around the ICD IPG. Reductions in size of the capacitors may also allow for the balanced addition of volume to the battery, thereby increasing longevity of the ICD IPG, or balanced addition of new components, thereby adding functionality to the ICD IPG. It is also desirable to provide such ICD IPGs at low cost while retaining the highest level of performance. At the same time, reliability of the capacitors cannot be compromised.

[0008] Various types of flat and spiral-wound capacitors are known in the art, some examples of which are described as follows and/or may be found in the patents listed in Table 1 of the above-referenced parent patent application Ser. No. 09/103,843.

[0009] Prior art high voltage electrolytic capacitors used in ICDs have two or more anode and cathode layers (or “electrodes”) and operate at room or body temperature. Typically, the capacitor is formed with a capacitor case enclosing an etched aluminum foil anode, an aluminum foil or film cathode, and a Kraft paper or fabric gauze spacer or separator impregnated with a solvent based liquid electrolyte interposed therebetween. A layer of aluminum oxide that functions as a dielectric layer is formed on the etched aluminum anode, preferably during passage of electrical current through the anode. The electrolyte comprises an ion producing salt that is dissolved in a solvent and provides ionic electrical conductivity between the cathode and the aluminum oxide dielectric. The energy of the capacitor is stored in the electrostatic field generated by opposing electrical charges separated by the aluminum oxide layer disposed on the surface of the anode and is proportional to the surface area of the aluminum anode. Thus, to minimize the overall volume of the capacitor one must maximize anode surface area per unit volume without increasing the capacitor's overall (i.e., external) dimensions. The separator material, anode and cathode layer terminals, internal packaging, electrical interconnections, and alignment features and cathode material further increase the thickness and volume of a capacitor. Consequently, these and other components in a capacitor and the desired capacitance limit the extent to which its physical dimensions may be reduced.

[0010] Some ICD IPGs employ commercial photoflash capacitors similar to those described by Troup in “Implantable Cardioverters and Defibrillators,” Current Problems in Cardiology, Volume XIV, Number 12, December 1989, Year Book Medical Publishers, Chicago, and as described in U.S. Pat. No. 4,254,775. The electrodes or anode and cathodes are wound spiral into anode and cathode layers separated by separator layers of the spiral. Anode layers employed in such photoflash capacitors typically comprise one or two sheets of a high purity (99.99%), porous, highly etched, anodized aluminum foil. Cathode layers in such capacitors are formed of a non-porous, highly etched aluminum foil sheet which may be somewhat less pure (99.7%) respecting aluminum content than the anode layers. The separator formed of one or more sheet or layer of Kraft paper saturated and impregnated with a solvent based liquid electrolyte is located between adjacent anode and cathode layers. The anode foil thickness and cathode foil thickness are on the order of 100 micrometers and 20 micrometers, respectively. Most commercial photoflash capacitors contain a core of separator paper intended to prevent brittle, highly etched aluminum anode foils from fracturing during winding of the anode, cathode and separator layers into a coiled configuration. The cylindrical shape and paper core of commercial photoflash capacitors limits the volumetric packaging efficiency and thickness of an ICD IPG housing made using same.

[0011] The aluminum anodes and cathodes of aluminum electrolytic capacitors generally each have at least one tab extending beyond their perimeters to facilitate electrical connection of all (or sets of) the anode and cathode layers electrically in parallel to form one or more capacitor and to make electrical connections to the exterior of the capacitor case. Tab terminal connections for a wound electrolytic capacitor are described in U.S. Pat. No. 4,663,824 that are laser welded to feedthrough pin terminals of feedthroughs extending through the case. Wound capacitors usually contain two or more tabs joined together by crimping or riveting.

[0012] Flat electrolytic capacitors have also been disclosed in the prior art for general applications as well as for use in ICDs. More recently developed ICD IPGs employ one or more flat high voltage capacitor to overcome some of the packaging and volume disadvantages associated with cylindrical photoflash capacitors. For example, U.S. Pat. No. 5,131,388 discloses a flat capacitor having a plurality of stacked capacitor layers each comprising an “electrode stack sub-assembly”. Each capacitor layer contains one or more anode sheet forming an anode layer having an anode tab, a cathode sheet or layer having a cathode tab and a separator for separating the anode layer from the cathode layer. In the '388 patent, the electrode stack assembly of stacked capacitor layers is encased within a non-conductive, polymer envelope that is sealed at its seams and fitted into a chamber of a conductive metal, capacitor case or into a compartment of the ICD IPG housing, and electrical connections with the capacitor anode(s) and cathode(s) are made through feedthroughs extending through the case or compartment wall. The tabs of the anode layers and the cathode layers of all of the capacitor layers of the stack are electrically connected in parallel to form a single capacitor or grouped to form a plurality of capacitors. The aluminum anode layer tabs are gathered together and electrically connected to a feedthrough pin of an anode feedthrough extending through the case or compartment wall. The aluminum cathode layer tabs are gathered together and electrically connected to a feedthrough pin of a cathode feedthrough extending through the case or compartment wall or connected to the electrically conductive capacitor case wall.

[0013] Many improvements in the design of flat aluminum electrolytic capacitors for use in ICD IPGs have been disclosed, e.g., those improvements described in “High Energy Density Capacitors for Implantable Defibrillators” presented by P. Lunsmann and D. MacFarlane at CARTS 96: 16th Capacitor and Resistor Technology Symposium 11–15, Mar. 1996, and at CARTS-EUROPE 96: 10th European Passive Components Symposium., 7–11, Oct. 1996, pp. 35–39. Further features of flat electrolytic capacitors for use in ICD IPGs are disclosed in U.S. Pat. Nos. 4,942,501; 5,086,374; 5,146,391; 5,153,820; 5,562,801; 5,584,890; 5,628,801; and 5,748,439, all issued to MacFarlane et al.

[0014] A number of recent patents including U.S. Pat. No. 5,660,737 and U.S. Pat. Nos. 5,522,851; 5,801,917; 5,808,857; 5,814,082; 5,908,151; 5,922,215; 5,926,357; 5,930,109; 5,968,210 and 5,983,472, all assigned to the same assignee, disclose related flat electrolytic capacitor designs for use in ICDs. In several of these patents, internal alignment elements are employed as a means for controlling the relative edge spacing of the anode and cathode layers from the conductive capacitor case. In these patents, each anode layer and cathode layer is provided with an outwardly extending tab, and the anode and cathode tabs are electrically connected in common to a feedthrough pin and a step feature of the conductive capacitor case, respectively. The cathode tabs are gathered together against the step feature and ultrasonically welded together and to the step feature. In the '357 patent, the anode tabs are laser welded to one end of an aluminum ribbon that is ultrasonically welded at its other end to an aluminum layer that is ultrasonically welded to the terminal pin. The feedthrough terminal pin is electrically isolated from and extends outside and away from the case to provide an anode connection pin. A cathode connection pin is attached to the case and extends outwardly therefrom. The anode and cathode connection pins are electrically connected into the DC—DC converter circuitry, but the attachment mechanism is not described in any detail.

[0015] The above-described electrolytic capacitors typically employ a liquid electrolyte that is injected into the case through a fill port or tube after the electrode stack assembly is nested within the case chamber, the electrical connections are made to the case or the feedthrough terminal pin, and the case cover or lid is hermetically sealed over the case opening to hermetically enclose the chamber and its contents. Such liquid electrolyte fill tubes or ports are described in U.S. Pat. Nos. 5,862,035 and 3,918,474, for example. Similarly, certain lead acid automotive batteries, have employed fill ports in their top surfaces that are normally upright for initially introducing and replenishing evaporation of acid electrolyte. In certain cases, the fill ports allow for the venting of hydrogen gases released during charging and discharging cycles, as disclosed in U.S. Pat. No. 5,449,574, for example. However, leakage can occur if such batteries are not kept upright.

[0016] Gases are also released during the charging and discharging cycles of various types of electrolytic capacitors including the above-described wound and flat aluminum anode and cathode layer electrolytic capacitors. In particular, hydrogen is released from the liquid electrolyte, and the gas builds up within the sealed capacitor case. Severe hydrogen accumulation can cause the capacitor case walls to swell outward and conceivably could cause the capacitor case to rupture at a laser weld seam or feedthrough. A variety of vent mechanisms have been provided through the capacitor case wall of various types of electrolytic capacitors that rupture or open when gas pressure reaches a threshold level as disclosed, for example, in U.S. Pat. Nos. 4,004,199; 4,074,417; 4,617,611; and 5,591,540. These approaches may be suitable in environments where the electrolyte can cause no harm and the capacitor can be readily replaced, but are inappropriate in the close confines of an IPG housing where the capacitor cannot be replaced and the electrolyte could harm other components. Therefore a need exists for an appropriate way to relieve gas pressure within the capacitor interior case chamber without loss of electrolyte.

[0017] Accordingly, the present invention is directed to the venting of gas from an flat electrolytic capacitor sealed within the hermetically sealed housing of an IMD, particularly an ICD. These problems are solved in accordance with the present invention by incorporating a gas vent and liquid electrolyte barrier into the electrolyte fill tube lumen that is used to fill the interior case chamber with electrolyte and then needs to be closed to prevent leakage of electrolyte.

[0018] Preferably, the fill port is formed through one of the case side wall and the cover wall for filling the capacitor with the liquid electrolyte upon welding of the cover to the side wall edge. The fill port is shaped to comprise a fill port tube having interior and exterior tube ends and a fill port ferrule intermediate the ends of the fill port tube and comprising a fill port ferrule flange extending transversely to and away from the fill port tube. The fill port ferrule is mounted in an opening disposed in one of the case wall and the cover wall with the ferrule flange in sealing engagement therewith to locate the exterior tube end extending outwardly away from the fill port ferrule flange and the interior tube end within the interior case chamber. Preferably the fill port tube and ferrule are formed integrally of a the same material, wherein the fill port ferrule comprises a fill port flange extending radially outwardly transversely from the fill port tube.

[0019] A microporous plug is injected into and fills the fill port lumen, the plug formed of a microporous material allowing the escape of gas released from the liquid electrolyte during capacitor charging while preventing escape of liquid or vaporized electrolyte. Suitable plug materials include some silicones, polyimides, polyphenylene oxides, cellulose acetates and triacetates and polysulfones. Several potting materials, such as epoxy or silicone rubber cement have the foregoing chemical resistance and hydrogen permeability properties and thus are suitable for use in the present invention.

[0020] The microporous plug is injected in a liquid state into the tube lumen from the exterior tube end and preferably extends through the tube lumen and into the interior case chamber in a region substantially covering the interior tube end such that when the plug solidifies, it adheres thereto with sufficient force to inhibit expulsion of the plug outward from the tube lumen under gas pressure.

[0021] Those of ordinary skill in the art will understand immediately upon referring to the drawings, detailed description of the preferred embodiments and claims hereof that many objects, features and advantages of the capacitors and methods of the present invention will find application in the fields other than the field of IMDs.

[0058] FIG. 1 illustrates one embodiment of ICD IPG 10 in which the flat electrolytic capacitor of the present invention is advantageously incorporated, the associated ICD electrical leads 14, 16 and 18, and their relationship to a human heart 12. The leads are coupled to ICD IPG 10 by means of multi-port connector block 20, which contains separate connector ports for each of the three leads illustrated. Lead 14 is coupled to subcutaneous electrode 30, which is intended to be mounted subcutaneously in the region of the left chest. Lead 16 is a coronary sinus lead employing an elongated coil electrode which is located in the coronary sinus and great vein region of the heart. The location of the electrode is illustrated in broken line format at 32, and extends around the heart from a point within the opening of the coronary sinus to a point in the vicinity of the left atrial appendage.

[0059] Lead 18 is provided with elongated electrode coil 28 which is located in the right ventricle of the heart. Lead 18 also includes stimulation electrode 34 which takes the form of a helical coil which is screwed into the myocardial tissue of the right ventricle. Lead 18 may also include one or more additional electrodes for near and far field electrogram sensing.

[0060] In the system illustrated, cardiac pacing pulses are delivered between helical electrode 34 and elongated electrode 28. Electrodes 28 and 34 are also employed to sense electrical signals indicative of ventricular contractions. As illustrated, it is anticipated that the right ventricular electrode 28 will serve as the common electrode during sequential and simultaneous pulse multiple electrode defibrillation regimens. For example, during a simultaneous pulse defibrillation regimen, pulses would simultaneously be delivered between electrode 28 and electrode 30 and between electrode 28 and electrode 32. During sequential pulse defibrillation, it is envisioned that pulses would be delivered sequentially between subcutaneous electrode 30 and electrode 28 and between coronary sinus electrode 32 and right ventricular electrode 28. Single pulse, two electrode defibrillation shock regimens may be also provided, typically between electrode 28 and coronary sinus electrode 32. Alternatively, single pulses may be delivered between electrodes 28 and 30. The particular interconnection of the electrodes to an ICD will depend somewhat on which specific single electrode pair defibrillation shock regimen is believed more likely to be employed.

[0061] FIG. 2 is a block diagram illustrating the interconnection of high voltage output circuit 40, high voltage charging circuit 64 and capacitors 265 according to one example of the microcomputer based operating system of the ICD IPG of FIG. 1. As illustrated, the ICD operations are controlled by means of a stored program in microprocessor 42, which performs all necessary computational functions within the ICD. Microprocessor 42 is linked to control circuitry 44 by means of bi-directional data/control bus 46, and thereby controls operation of the output circuitry 40 and the high voltage charging circuitry 64. Pace/sense circuitry 78 awakens microprocessor 42 to perform any necessary mathematical calculations, to perform tachycardia and fibrillation detection procedures and to update the time intervals controlled by the timers in pace/sense circuitry 78 on reprogramming of the ICD operating modes or parameter values or on the occurrence of signals indicative of delivery of cardiac pacing pulses or of the occurrence of cardiac contractions.

[0062] The basic operation and particular structure or components of the exemplary ICD of FIGS. 1 and 2 may correspond to any of the systems known in the art, and the present invention is not dependent upon any particular configuration thereof. The flat aluminum electrolytic capacitor of the present invention may be employed generally in conjunction with the various systems illustrated in the aforementioned '209 patent, or in conjunction with the various systems or components disclosed in the various U.S. patents listed in the above-referenced parent patent application Ser. No. 09/103,843.

[0063] Control circuitry 44 provides three signals of primary importance to output circuitry 40. Those signals include the first and second control signals discussed above, labeled here as ENAB, line 48, and ENBA, line 50. Also of importance is DUMP line 52 which initiates discharge of the output capacitors and VCAP line 54 which provides a signal indicative of the voltage stored on the output capacitors C1, C2, to control circuitry 44. Defibrillation electrodes 28, 30 and 32 illustrated in FIG. 1, above, are shown coupled to output circuitry 40 by means of conductors 22, 24 and 26. For ease of understanding, those conductors are also labeled as “COMMON”, “HVA” and “HVB”. However, other configurations are also possible. For example, subcutaneous electrode 30 may be coupled to HVB conductor 26, to allow for a single pulse regimen to be delivered between electrodes 28 and 30. During a logic signal on ENAB, line 48, a cardioversion/defibrillation shock is delivered between electrode 30 and electrode 28. During a logic signal on ENBA, line 50, a cardioversion/defibrillation shock is delivered between electrode 32 and electrode 28.

[0064] The output circuitry includes a capacitor bank, including capacitors C1 and C2 and diodes 121 and 123, used for delivering defibrillation shocks to the electrodes. Alternatively, the capacitor bank may include a further set of capacitors as depicted in the above referenced '758 application. In FIG. 2, capacitors 265 are illustrated in conjunction with high voltage charging circuitry 64, controlled by the control/timing circuitry 44 by means of CHDR line 66. As illustrated, capacitors 265 are charged by means of a high frequency, high voltage transformer 65. Proper charging polarities are maintained by means of the diodes 121 and 123. VCAP line 54 provides a signal indicative of the voltage on the capacitor bank, and allows for control of the high voltage charging circuitry and for termination of the charging function when the measured voltage equals the programmed charging level.

[0065] Pace/sense circuitry 78 includes an R-wave sense amplifier and a pulse generator for generating cardiac pacing pulses, which may also correspond to any known cardiac pacemaker output circuitry and includes timing circuitry for defining ventricular pacing intervals, refractory intervals and blanking intervals, under control of microprocessor 42 via control/data bus 80.

[0066] Control signals triggering generation of cardiac pacing pulses by pace/sense circuitry 78 and signals indicative of the occurrence of R-waves, from pace/sense circuitry 78 are communicated to control circuitry 44 by means of a bidirectional data bus 81. Pace/sense circuitry 78 is coupled to helical electrode 34 illustrated in FIG. 1 by means of a conductor 36. Pace/sense circuitry 78 is also coupled to ventricular electrode 28, illustrated in FIG. 1, by means of a conductor 82, allowing for bipolar sensing of R-waves between electrodes 34 and 28 and for delivery of bipolar pacing pulses between electrodes 34 and 28, as discussed above.

[0067] FIGS. 3(a) through 3(g) show perspective views of various components of ICD IPG 10, including one embodiment of a pair of the capacitors 265, as those components are placed successively within the housing of ICD IPG 10. In FIG. 3(a), electronics module 360 is placed in right-hand shield 340 of ICD IPG 10. FIG. 3(b) shows ICD IPG 10 once electronics module 360 has been seated in right-hand shield 340.

[0068] FIG. 3(c) shows a pair of capacitors 265 formed as described herein prior to being placed within right-hand shield 340, the capacitors 265 being connected electrically in series by interconnections in electronics module 340. FIG. 3(d) shows ICD IPG 10 once the pair of capacitors 265 has been placed within right-hand shield 340.

[0069] FIG. 3(e) shows insulator cup 370 prior to its placing atop capacitors 265 in right-hand shield 340. FIG. 3(f) shows electrochemical cell or battery 380 having insulator 382 disposed around the battery 380 prior to placing it in shield 340. Battery 380 provides the electrical energy required to charge and re-charge capacitors 265, and also powers electronics module 360. Battery 380 may take any of the forms employed in the prior art to provide cardioversion/defibrillation energy, some of which are identified in parent patent application Ser. No. 09/103,843.

[0070] FIG. 3(g) shows ICD IPG 10 having left-hand shield 350 connected to right-hand shield 340 and feedthrough 390 projecting upwardly from both shield halves. Activity sensor 400 and patient alert apparatus 410 are shown disposed on the side lower portion of left-hand shield 350. Left-hand shield 350 and right-hand shield 340 are subsequently closed and hermetically sealed (not shown in the figures).

[0071] FIG. 4 shows an exploded view of one embodiment of a capacitor layer or single anode/cathode sub-assembly 227 of capacitor 265. The capacitor design described herein employs a stacked configuration of a plurality of capacitor layers or single anode/cathode sub-assemblies 227 as further described below with respect to FIG. 6. Each anode/cathode sub-assembly 227 comprises alternating substantially rectangular-shaped anode layers 185 and cathode layers 175, with a substantially rectangular-shaped separator layer 180 being interposed therebetween. The shapes of anode layers 185, cathode layers 175 and separator layers 180 are primarily a matter of design choice, and are dictated largely by the shape or configuration of case 90 within which those layers are ultimately disposed. Anode layers 185, cathode layers 175 and separator layers 180 may assume any arbitrary shape to optimize packaging efficiency.

[0072] Anode sub-assembly 170d most preferably comprises a plurality of non-notched anode layers 185a, 185b, 185c, notched anode layer 190 including anode tab notch 200, and anode tab 195 coupled to anode layer 185a. It will be understood that anode sub-assembly 170d shown in FIG. 4 is but one possible embodiment of an anode sub-assembly 170. Cathode layer 175d most preferably is formed of a single sheet and has cathode tab 176 formed integral thereto and projecting from the periphery thereof.

[0073] In one preferred embodiment of the sub-assembly 227 as depicted in the figures, two individual separator layer sheets 180a and 180b form the separator layer 180 that is disposed between each anode sub-assembly 170 and cathode layer 175. Further single separator layer sheets 180a and 180b are disposed against the outer surfaces of the anode layer 185c and the cathode layer 175d. When the sub-assemblies are stacked, the outermost single separator layer sheets 180a and 180b bear against adjacent outermost single separator layer sheets 180b and 180a, respectively, of adjacent capacitor layers so that two sheet separator layers 180 separate all adjacent cathode and anode layers of an electrode stack assembly 225.

[0074] It will be understood by those skilled in the art that the precise number of sub-assemblies 227 selected for use in a electrode stack assembly 225 will depend upon the energy density, volume, voltage, current, energy output and other requirements placed upon capacitor 265. Similarly, it will be understood by those skilled in the art that the precise number of notched and un-notched anode layers 185, anode tabs 195, anode sub-assemblies 170, cathode layers 175 and separator layers 180 selected for use in a given embodiment of anode/cathode sub-assembly 227 will depend upon the energy density, volume, voltage, current, energy output and other requirements placed upon capacitor 265. It will now become apparent that a virtually unlimited number of combinations and permutations respecting the number of anode/cathode sub-assemblies 227, and the number of un-notched and notched anode layers 185 forming anode sub-assembly 170, anode sub-assemblies 170, anode tabs 195, cathode layers 175 and separator layers 180 disposed within each anode/cathode sub-assembly 227, may be selected according to the particular requirements of capacitor 265. Anode layers 185, cathode layers 175 and separator layers 180 are most preferably formed of materials typically used in high quality aluminum electrolytic capacitors.

[0075] Anode layers 185 and 190 are formed of anode foil that is most preferably through-etched, has a high specific capacitance (at least about 0.3, at least about 0.5 or most preferably at least about 0.8 microfarads/cm²), has a dielectric withstand parameter of at least 425 Volts DC, a thickness ranging between about 50 and about 200 micrometers, more preferably between about 75 and 150 micrometers, more preferably yet between about 90 and about 125 micrometers, and most preferably being about 100 micrometers thick, and a cleanliness of about 1.0 mg/m² respecting projected area maximum chloride contamination. The anode foil preferably has a rated surge voltage of 390 Volts, an initial purity of about 99.99% aluminum, a final thickness of about 104 micrometers, plus or minus about five micrometers, and a specific capacitance of about 0.8 microfarads per square centimeter. Suitable anode foils are commercially available on a widespread basis.

[0076] Individual anode layers 185 are typically somewhat stiff and formed of high-purity aluminum processed by etching to achieve high capacitance per unit area. Thin anode foils are preferred, especially if they substantially maintain or increase specific capacitance while reducing the thickness of the electrode stack assembly 225, or maintain the thickness of electrode stack assembly 225 while increasing overall capacitance. For example, it is contemplated that individual anode layers 185 have a thickness of about 10 micrometers, about 20 micrometers, about 30 micrometers, about 40 micrometers, about 50 micrometers, about 60 micrometers, about 70 micrometers, about 80 micrometers, about 90 micrometers, about 100 micrometers, about 110 micrometers, about 120 micrometers, about 130 micrometers, about 140 micrometers and about 150 micrometers.

[0077] Cathode layers 175 are preferably high purity and are comparatively flexible. Cathode layers 175 are most preferably formed from cathode foil having high surface area (i.e., highly etched cathode foil), high specific capacitance (preferably at least 200 microfarads/cm², and at least 250 microfarads/cm² when fresh), a thickness of about 30 micrometers, a cleanliness of about 1.0 mg/m² respecting projected area maximum chloride contamination, and a purity which may be less than corresponding to the starting foil material from which anode foil is made. The cathode foil preferably has an initial purity of at least 99% aluminum, and more preferably yet of about 99.4% aluminum, a final thickness of about 30 micrometers, and an initial specific capacitance of about 250 microfarads per square centimeter. In other embodiments, cathode foil has a specific capacitance ranging between about 100 and about 500 microfarads/cm², about 200 and about 400 microfarads/cm², or about 250 and about 350 microfarads/cm², a thickness ranging between about 10 and about 150 micrometers, about 15 and about 100 micrometers, about 20 and about 50 micrometers, or about 25 and about 40 micrometers.

[0078] It is generally preferred that the specific capacitance of the cathode foil be as high as possible, and that cathode layer 175 be as thin as possible. For example, it is contemplated that individual cathode layers 175 have specific capacitances of about 100 microfarads/cm², about 200 microfarads/cm², about 300 microfarads/cm², about 400 microfarads/cm², about 500 microfarads/cm², about 600 microfarads/cm², about 700 microfarads/cm², about 800 microfarads/cm², about 900 microfarads/cm², or about 1,000 microfarads/cm². Suitable cathode foils are commercially available on a widespread basis. In still other embodiments, cathode foil is formed of materials or metals in addition to aluminum, aluminum alloys and “pure” aluminum.

[0079] Separator layer sheets 180a and 180b outer separator layers 165a and 165b are most preferably made from a roll or sheet of separator material. Separator layers 180 are preferably cut slightly larger than anode sub-assemblies 170 and cathode layers 175 to accommodate misalignment during the stacking of layers, to prevent subsequent shorting between anode and cathode layers, and to otherwise ensure that a physical barrier is disposed between the anodes and the cathodes of the finished capacitor.

[0080] It is preferred that separator layer sheets 180a and 180b and exterior separator layers 165a and 165b (shown in FIG. 9) be formed of a material that: (a) is chemically inert; (b) is chemically compatible with the selected electrolyte; (c) may be impregnated with the electrolyte to produce a low resistance path between adjoining anode and cathode layers, and (d) physically separates adjoining anode and cathode layers. In one preferred embodiment, separator material is a pure cellulose, very low halide or chloride content Kraft paper having a thickness of about 0.0005 inches (0.0013 mm), a density of about 1.06 grams/cm3, a dielectric strength of 1,400 Volts AC per 0.001 inch (0.025 mm) thickness, and a low number of conducting paths (about 0.4/ft2 or less). Separator layer sheets 180a and 180b and outer separator layers 165a and 165b may also be formed of materials other than Kraft paper, such as Manila paper, porous polymeric materials or fabric gauze materials. For example, porous polymeric materials may be disposed between anode and cathode layers like those disclosed in U.S. Pat. Nos. 3,555,369 and 3,883,784 in some embodiments of the capacitor layers

[0081] Continuing to refer to FIG. 4, a first preferred step in assembling a flat aluminum electrolytic capacitor is to cut anode layers 185 and 190, anode tabs 195, cathode layers 175 and separator layers 180. Those components are most preferably cut to shape using dies having low wall-to-wall clearance, where inter-wall spacing between the substantially vertically-oriented corresponding walls of the punch and die is most preferably on the order of about 6 millionths of an inch per side. Larger or smaller inter-wall spacings between the substantially vertically-oriented corresponding walls of the punch and cavity, such as about 2, about 4, about 5, about 7, about 8, about 10 and about 12 millionths of an inch may also be employed but are less preferred.

[0082] Such low clearance results in smooth, burr free edges being formed along the peripheries of anode layers 185 and 190, anode tabs 195, cathode layers 175 and separator layers 180. Smooth, burr free edges on the walls of the dies have been discovered to be critical respecting reliable performance of a capacitor. The presence of burrs along the peripheries of anode layers 185 and 190, anode tabs 195, cathode layers 175 and separator layers 180 may result in short circuit and failure of the capacitor. The means by which anode foil, cathode foil and separator materials are cut or formed may have a significant impact on the lack or presence of burrs and other cutting debris disposed about the peripheries of the formed or cut members. Low clearance dies produce an edge superior to that achieved by other cutting methods, such as steel rule dies. The shape, flexibility and speed of a low clearance die are be superior to those of laser or blade cutting. Other methods of cutting or forming anode layers 185 and 190, anode tabs 195, cathode layers 175 and separator layers 180 include, but are not limited to, steel rule die cutting, laser cutting, water jet cutting and blade cutting.

[0083] The preferred low clearance of the die apparatus is especially important for cutting thin ductile materials such as the cathode foil. In addition to improving reliability, burr and debris reduction permits reductions in the thickness of separator layer 180, thereby improving energy density of the capacitor. Angle cutting, where the face of the punch is not held parallel to the opposing floor of the die during the cutting step, is another less preferred method of cutting or forming anode layers 185 and 190, anode tabs 195, cathode layers 175 and separator layers 180.

[0084] It is preferred to cut or otherwise form separator layer 180 such that its outer periphery conforms closely to that of the corresponding side walls of the interior of case 90. In preferred embodiments, the periphery of separator layer is disposed within plus or minus 0.009 inches of the corresponding side walls of case 90. Such close conformity between the periphery of separator layer 180 and the corresponding internal side walls of case 90 has been discovered to provide the advantage of permitting separator layers 180 to immobilize or secure firmly in place electrode stack assembly 225 in case 90. This immobilization occurs because the separator paper forming separator layers 180 swells after electrolyte is added through the lumen of fill port 107 into otherwise assembled and sealed capacitor 265.

[0085] In a preferred method, foil or separator materials are drawn between the punch and cavity portions of a die having appropriate clearances on a roll. An air or hydraulically actuated press is then most preferably employed to actuate the punch or cavity portion of the die. The punch portion of the die is most preferably formed of hardened tool steel, or has other suitable wear resistant materials or coatings disposed on the cutting surfaces thereof When the cavity of the die is aligned vertically, the punch portion of the die may travel either upwards or downwards towards the die cavity during a cutting cycle. In the former case, components are cut and drop downwardly into a container for use in subsequent assembly operations. In the latter case, components are cut and may be presented directly to automated assembly equipment, such as robots equipped with vacuum or other pick-up tooling, for subsequent processing. Low clearance dies of the type described herein may be supplied by Top Tool, Inc. of Minneapolis, Minn.

[0086] Anode sub-assembly 170 most preferably includes one notched anode layer 190, which facilitates appropriate placing and positioning of anode tab 195 within anode sub-assembly 170. More than one notched anode layer 190 may also be included in anode sub-assembly 170. It is preferred that the remaining anode layers of anode sub-assembly 170 be non-notched anode layers 185. Anode tab 195 is most preferably formed of aluminum strip material. In one preferred embodiment, the aluminum strip has a purity of about 99.99% aluminum and a lesser degree of anodization than the anode foil or sheet. When anode tab 195 is formed of a non-anodized material, cold welding of anode tab 195 to non-notched anode layers 185 may be accomplished with less force and deflection, more about which we say below. It is preferred that the thickness of anode tab 195 be about equal to that of notched anode layer 190. If more than one notched anode layer 190 is employed in anode sub-assembly 170, a thicker anode tab 195 may be employed.

[0087] FIG. 13 shows a flow chart that describes generally one method, from beginning to end, of making flat aluminum electrolytic capacitor 265. FIGS. 14 through 20, on the other hand, show specific portions of the method or process described generally in FIG. 13.

[0088] FIG. 14 shows a flow chart of one method for making anode layer 170 wherein non-notched anode layers 185, notched anode layer 190 and anode tab 195 are provided and assembled within cold welder 202 to form anode sub-assembly 170.

[0089] Referring now to FIGS. 5(a) through 5(c), two non-notched anode layers 185a and 185b are placed on cold welding fixture base layer 207 of cold welding apparatus 202. The various structural members of cold welding apparatus 202 are most preferably formed of precision machined stainless steel or a high strength aluminum alloy. Layers 185a and 185b are next aligned and positioned appropriately on cold welding fixture base layer 207 using spring loaded alignment pins 209a through 209e. Pins 209a through 209e react upon top layer 208 being pressed downwardly upon layers 185a and 185b disposed within cold welding cavity 220. See also FIG. 5(c), where a cross-sectional view of cold welding apparatus 202 is shown.

[0090] Anode layer 190 is similarly disposed within cavity 220, followed by placing anode tab 195 within anode tab notch 200 in notched anode layer 190. Anode tab 195 is most preferably positioned along the periphery of notched anode layer 190 with the aid of additional spring loaded alignment pins 209f and 209g disposed along the periphery of anode tab 195. Non-notched anode layer 185c is then placed atop anode layer 190. Stacked anode sub-assembly 170 is then clamped between top plate 208 and base plate 207. Disposed within base plate 207 are anode layer cold welding pins 206a and anode tab cold welding pin 211a. Disposed within top plate 208 are anode layer cold welding pin 206b and anode tab cold welding pin 211b. Base plate 207 and top plate 208 are aligned such that the axes of cold welding pins 206a and 206b coincide with and are aligned respecting corresponding cold welding pins 211a and 211b.

[0091] Upper actuation apparatus 214 of cold welding apparatus 202 displaces cold welding pins 206b and 211b downwardly. Lower actuation apparatus 215 displaces cold welding pins 206a and 211a upwardly. In one embodiment of upper actuation apparatus 214 and lower actuation apparatus 215, pneumatic cylinders are employed to move pins 206a, 206b, 211a and 211b. In another embodiment of apparatus 214 and apparatus 215, a pair of rolling wheels is provided that move simultaneously and perpendicularly to the axes of pins 206a, 206b, 211a, and 211b. Still other embodiments of apparatus 214 and apparatus 215 may employ hydraulic actuators, cantilever beams, dead weights, springs, servomotors electromechanical solenoids, and the like for moving pins 206a, 206b, 211a and 211b. Control of actuation apparatus 214 and apparatus 215 respecting pin displacement force magnitude and timing may be accomplished using any one or combination of constant load, constant displacement, solenoid controller, direct or indirect means.

[0092] Following clamping with top plate 208, cold welding pins 206a, 206b, 211a and 211b are actuated. Cold welds 205 and 210 in anode sub-assembly 170 are formed by compression forces generated when cold weld pins 206a, 206b, 211a and 211b are compressed against anode sub-assembly 170. See FIG. 6(a), where the preferred regions in which cold welds 205 and 210 are formed are shown. Cold welds 205 and 210 may be described as not only cold welds, but forged welds. This is because the interfacial boundaries between anode layers 185 are deformed in the region of welds 205 and 210, thereby disrupting oxide layers and bringing base metals into direct contact with one another where metallic bonding occurs. Metallic bonding increases the strength of the welds.

[0093] In one embodiment of the method, a plurality of pneumatic cylinders function simultaneously in upper actuation apparatus 214 and lower actuation apparatus 215 to drive pins 206a, 206b, 211a and 211b against anode sub-assembly 170. Anode layer cold weld 205 and anode tab cold weld 210 are most preferably formed under direct constant load conditions, where pneumatic cylinders are pressurized to a predetermined fixed pressure. Anode layer cold weld 205 and anode tab cold weld 210 may also be formed under indirect constant displacement conditions, where pneumatic cylinders are pressurized until a displacement sensor placed across cold welding pins 206a, 206b, 211a or 211b generates a signal having a predetermined value, whereupon those pins are disengaged from anode/cathode sub-assembly 227.

[0094] In another embodiment of the method, a cantilever beam mechanism is incorporated into upper actuation apparatus 214 and lower actuation apparatus 215. Anode layer cold weld 205 and anode tab cold weld 210 are formed under direct constant displacement conditions, where cantilever beams are actuated and cause upper and lower members 208 and 207 to engage anode(cathode sub-assembly 227 until a hard stop point is reached. An indirect load controlled system may also be employed in apparatus 214 and apparatus 215, where cantilever or other means include a load measuring sensor for controlling the stop point of the cantilever beam, for example, when a predetermined load is measured by the sensor.

[0095] The cross-sectional shape of cold weld pins 206a, 206b, 211a and 211b may be square, circular, oval or any other suitable shape. The shape of the ends of cold weld pins 206a, 206b, 211a and 211b may be flat, rounded, domed or any other suitable shape appropriate for selectively controlling the properties of the cold welds produced therein. Likewise, more or fewer than four cold weld pins may be employed. The ends of cold weld pins 206a, 206b, 211a and 211b are most preferably rounded or domed and circular in cross-section. Cold weld pins 206a, 206b, 211a and 211b preferably have a diameter of about 0.060 inches (0.174 mm) and further have a beveled or radiused end. Cold weld pins 206a, 206b, 211a and 211b are preferably made from a high strength material that does not readily deform under the pressures obtained during welding, such as stainless steel, titanium tool steel or HSLA steel. The ends or side walls of cold welding pins 206a, 206b, 211a and 211b may be coated, clad or otherwise modified to increase wear resistance, deformation resistance or other desirable tribilogical attributes of the pins.

[0096] The primary function of cold welds 205 and 210 is to provide electrical interconnections between layers 185a, 185b, 185c and 190 and anode tab 195, while minimizing the overall thickness of anode sub-assembly 170 in the regions of welds 205 and 210. Typical prior art commercial cylindrical capacitors exhibit a significant increase in the thickness of the anode layer in the regions of the cold welds. This increase in thickness is typically on the order of about two times the thickness of the tab, or about 0.008 inch (0.020 mm). In the case of cylindrical capacitors where only one or two non-coincident tab connections are present, the overall effect on anode layer thickness may be minimal. In a stacked layer design having many more interconnections and welds, however, increases in weld zone thickness have been found to significantly increase the overall thickness of the anode layer and the electrode stack assembly as a whole.

[0097] In one cold welding method and corresponding apparatus, no or an inappreciable net increase in anode sub-assembly 170 thickness results when cold weld geometries and formation processes are appropriately optimized. Several embodiments of anode-assembly 170 have been found to have no more than about a 20% increase in layer thickness due to the presence of cold welds, as compared to about a 200% increase in thickness resulting from cold welds found in some commercial cylindrical capacitors. Two, three, four, five, six or more anode layers 185 and 190 may be cold-welded to form anode sub-assembly 170 as described herein.

[0098] FIG. 6(b) shows a cross-sectional view of a portion of one embodiment of a cold-welded anode assembly formed in accordance with the preferred cold welding method. Anode layers 185a, 190, 185b and 185c having anode layer thicknesses t_a, t_N, t_b and t_c, respectively, are cold-welded together at weld 205 through the compressive action of pins 206a and 206b mounted in bottom plate 207 and top plate 208, respectively. Pins 206a and 206b form central depressions 293 and 294, respectively, in anode sub-assembly 170d, and further result in the formation of rims 295 and 296, respectively. Rims 295 and 296 project downwardly and upwardly, respectively, from the surrounding surfaces of anode sub-assembly 170d, thereby increasing the overall thickness T of anode sub-assembly 170d by ΔT (T measured in respect of the non-cold-welded surrounding regions or portions of anode sub-assembly 170d).

[0099] FIG. 6(c) shows a cross-sectional view of another portion of one embodiment of a cold-welded anode assembly wherein anode layers 185a, 185b and 185c and anode tab 195, having anode layer/tab thicknesses t_a, t_b, t_c and t_tab, respectively, are cold-welded together at weld 210 through the compressive action of pins 211a and 211b mounted in bottom plate 207 and top plate 208, respectively. Pins 211a and 211b form central depressions 297 and 298, respectively, in anode sub-assembly 170d, and further result in the formation of rims 299 and 301, respectively. Rims 299 and 301 project downwardly and upwardly, respectively, from the surface of anode sub-assembly 170d, thereby increasing overall thickness T of anode sub-assembly 170d by ΔT (T measured in respect of the non-cold-welded surrounding regions or portions of anode sub-assembly 170d).

[0100] The overall thickness T of anode sub-assembly 170d is therefore defined by the equation:
T=nt
The maximum overall thickness T+ΔT of anode sub-assembly 170d in the region of cold welds 205 or 210 is then defined by the equation:
T+ΔT=nt+ΔT
where T_as is the overall thickness of anode sub-assembly 170d in non-cold-welded regions, n is the number of anode layers 185 and/or 190 in anode sub-assembly 170d, and t is the thickness of individual anode layers 185 and/or 190 or anode tab 195 where the thicknesses t_n, t_a, t_b, t_c and t_tab, are assumed to be the same.

[0101] It is highly desirable to form anode sub-assembly such that the ratio ΔT/T is less than or equal to 0.05, 0.1, 0.15, 0.20, 0.25, 0.30, 0.35, 0.40, 0.45 or 0.50. The lower the value of the ratio ΔT/T, the greater the volumetric efficiency of capacitor 265. Additionally, the overall thickness of capacitor 265 may be reduced when the value of the ratio ΔT/T is made smaller.

[0102] Referring now to FIG. 6(a), the overall thickness of electrode stack assembly 225 may be reduced further by staggering or offsetting horizontally the respective vertical locations of tabs 195a through 195h (and corresponding cold welds 210). In this embodiment, tabs 195a 195b, for example, are not aligned vertically in respect of one another. Such staggering or offsetting of tabs 195 permits the increases in thickness ΔT corresponding to each of anode subassemblies 170a through 170h to be spread out horizontally over the perimeter or other portion of electrode stack assembly 225 such that increases in thickness ΔT do not accumulate or add constructively, thereby decreasing the overall thickness of electrode stack assembly 225. Cold welds 205 may similarly be staggered or offset horizontally respecting one another and cold weld 210 to achieve a reduction in overall thickness of electrode stack assembly 225.

[0103] In another preferred embodiment, the anode sub-assembly 170 of each capacitor layer or electrode sub-assembly comprises a plurality of three, four, five or more anode sheets or layers 185 and 190, each sub-assembly most preferably having at least one anode layer having a corresponding anode tab 195 attached thereto or forming a portion thereof the layers being cold welded together to form anode sub-assembly 170. For example, an anode sub-assembly 170 may comprise six anode layers 185 constructed by cold-welding two separate triple anode layers 185 that were previously and separately cold-welded or otherwise joined together. Alternatively, anode sub-assembly 170 layer may comprise seven anode layers constructed by cold-welding together one triple anode layer 185 and one quadruple anode layer 185 that were previously and separately cold-welded or otherwise joined together. In another preferred embodiment, multiple notched anode layers 190 may employed in anode sub-assembly 170, thereby permitting the use of a thicker anode tab material.

[0104] The geometry of base plate 207 and top plate 208 in the regions surrounding cold welding pins 206a, 206b, 211a and 211b has been discovered to affect the properties of cold welds 205 and 210. In a preferred method, the mating surfaces of plates 207 and 208 surfaces have no radiused break formed in the perimeters of the pin holes. The presence of radiused breaks or chamfers in those regions may cause undesired deformation of cold welds 205 and 210 therein. Such deformation may result in an increase in the thickness of anode sub-assembly 170, which may translate directly into an increase in the thickness of capacitor 265. Note further that the increase in thickness so resulting is a multiple of the number of anode sub-assemblies 170 present in electrode stack assembly 225. Alternatively, radiused breaks or chamfers may be employed in the region of the pin holes in base plate 207 and top plate 208, but appropriate capacitor design accommodations are most preferably made, such as staggering the positions of adjoining stacked cold welds.

[0105] As shown in FIG. 14, once cold welding pins 206a, 206b, 211a and 211b have been actuated against anode sub-assembly 170, top plate 208 is removed and cold-welded anode sub-assembly 170 is provided for further stacking of anode/cathode sub-assembly 227. FIG. 15 shows a flow chart corresponding to one preferred method for making electrode stack assembly 225. See also FIG. 6(a), where an exploded top perspective view of one embodiment of an electrode stack assembly 225 of capacitor 265 is shown. As illustrated in FIGS. 4, 6(a) and 15, electrode stack assembly 225 most preferably comprises a plurality of cold-welded anode sub-assemblies 175a through 175h, a plurality of cathode layers 175a through 175i, a plurality of separator layers 180, outer separator layers 165a and 165b, outer wrap 115 and wrapping tape 245.

[0106] Outer wrap 115 is most preferably die cut from separator material described supra, but may be formed from a wide range of other suitable materials such as polymeric materials, aluminum, suitable heat shrink materials, suitable rubberized materials and synthetic equivalents or derivatives thereof, and the like. Wrapping tape 245 is most preferably cut from a polypropylene-backed acrylic adhesive tape, but may also be replaced by a staple, an ultrasonic paper joint or weld, suitable adhesives other than acrylic adhesive, suitable tape other than polypropylene-backed tape, a hook and corresponding clasp and so on.

[0107] Outer wrap 115 and wrapping tape 245 together comprise an electrode stack assembly wrap which has been discovered to help prevent undesired movement or shifting of electrode stack assembly 225 during subsequent processing. It will now become apparent to one skilled in the art that many means other than those disclosed explicitly herein exist for immobilizing and securing electrode stack assembly 225 during subsequent processing which accomplish substantially the same function as the electrode stack assembly wrap comprising outer wrap 115 and wrapping tape 245. Alternative means for immobilizing and securing electrode stack assembly 225 other than those described hereinabove exist. Such alternative means include, but are not limited to, robotic or other mechanical clamping and securing means not necessarily forming a portion of electrode stack assembly 225, adhesive electrolytes for forming separator layers 180, and so on.

[0108] The stacking process by which electrode stack assembly 225 is most preferably made begins by placing outer wrap 115 into a stacking fixture followed by placing outer paper or separator layer 165a thereon. Next, cathode layer 175a is placed atop separator layer 165a, followed by separator layers 180a and 180b being disposed thereon. Cold-welded anode sub-assembly 170a is then placed atop separator layer 180b, followed by placing separator layers 180a and 180b thereon, and so on. The placing of alternating cathode layers 175 and anode sub-assemblies 170 with separator layers 180a and 180b interposed therebetween continues in the stacking fixture until final cathode layer 175h has been placed thereon.

[0109] In the embodiment of electrode stack assembly 225 shown in FIG. 6(a), eight anode sub-assemblies (anode sub-assemblies 170a through 170h) and nine cathode layers (cathode layers 175a through 175i) are illustrated. The voltage developed across each combined anode sub-assembly/separator layer/cathode layer assembly disposed within electrode stack assembly 225 most preferably ranges between about 360 and about 390 Volts DC. As described below, the various anode sub-assemblies of electrode stack assembly 225 are typically connected in parallel electrically, as are the various cathode layers of electrode stack assembly 225.

[0110] Consistent with the discussion hereinabove concerning FIG. 4, it will now be understood by one skilled in the art that electrode stack assembly 225 shown in FIG. 6(a) is merely illustrative, and does not limit the scope of the present invention in any way respecting the number or combination of anode sub-assemblies 170, cathode layers 175, separator layers 180, anode tabs 195, cathode tabs 176, and so on. The number of electrode components is instead determined according to the total capacitance required, the total area of each layer, the specific capacitance of the foil employed and other factors.

[0111] In another embodiment of electrode stack assembly 225, the number of anode layers 185 employed in each anode sub-assembly 170 is varied in the stack. Such a design permits the fabrication of capacitors having the same layer area but nearly continuously varying different and selectable total capacitances that a user may determine by increasing or decreasing the number of anode layers 185/190 included in selected anode sub-assemblies 170 (as opposed to adding or subtracting full anode/cathode sub-assemblies 227 from electrode stack assembly 225 to thereby change the total capacitance). Following placing of cathode layer 175i in the stack, outer paper layer 165b is placed thereon, and outer wrap 115 is folded over the top of electrode stack assembly 225. Wrapping tape 245 then holds outer wrap 115 in place and secures the various components of electrode stack assembly 225 together.

[0112] The physical dimensions of separator layers 165 and 180 are most preferably somewhat larger than those of anode sub-assemblies 170 and cathode layers 175 to prevent contact of the electrodes with the case wall or electrical shorting between opposing polarity electrode layers due to the presence of burrs, stray or particulate material, debris or imperfections occurring therein. The reliability and functionality of capacitor 265 may be compromised if a portion of anode sub-assembly 170 comes into contact with a conducting case wall, if a burr on the periphery of anode sub-assembly 170 or cathode layer 175 comes into contact with an adjoining layer of opposing polarity, or if separator layer 180a or 180b does not provide sufficient electrical insulation between adjoining opposite-polarity electrode layers and conducting particulate matter bridges the gap therebetween.

[0113] The additional separator material most preferably disposed about the periphery of electrode stack assembly 225 is referred to herein as separator overhang. Decreasing the amount of separator overhang increases the energy density of capacitor 265. It is beneficial from an energy density optimization perspective, therefore, to decrease the amount or degree of separator overhang. The amount of separator overhang required has been discovered to be primarily a function of the stack-up tolerance characteristic of the stacking method employed.

[0114] In commercial cylindrical capacitors, the amount of separator overhang is typically on the order of 0.050 to 0.100 inches (0.127 to 0.254 mm). The above-referenced '851 patent describes a flat aluminum electrolytic capacitor wherein the housing of the capacitor has at least two internal alignment members. Those alignment members necessarily add volume to the capacitor while taking away from the total amount of “active” electrode material available, thereby decreasing the energy density of the capacitor.

[0115] Consistent registration of separator layers 165 and 180, anode sub-assemblies 170 and cathode layers 175 in electrode stack assembly 225 is advantageously achieved by stacking the various elements of electrode stack assembly 225 using robotic assembly techniques. More particularly, the various electrode and separator layers of electrode stack assembly 225 are stacked and aligned using an assembly work cell comprising four Seiko 4-axis SCARA Model No. TT8800 and TT8500, or equivalent, to pick up and place the various electrode and separator elements in an appropriate stacking fixture. Other suitable methods for stacking and registering electrode and separator layers include cam driven walking beam assembly machine techniques, rotary table machine techniques, multiple station single stacking machine techniques, and the like.

[0116] In a preferred method, a pre-formed or cut separator, electrode layer or sub-assembly is presented to a robot arm, which then picks the part up with end-of-arm tooling. A Venturi system produces a vacuum in the end-of-arm tooling. The system creates a vacuum at an appropriate time such that the part is sucked up onto the end-of-arm tooling. The vacuum is next released when the part is placed in the stacking fixture. A direct vacuum system, such as rubber suction cups, or other contact or non-contact pick up robotic or manual assembly methods may also be employed. The position of the part is robotically translated from the pickup point into the stacking fixture by the robot arm with an accuracy of 0.005 inch (0.013 mm) or less. After placing the part in the stacking fixture, part alignment is most preferably verified electronically with a SEIKO COGNEX 5400 VISION System, or equivalent, in combination with a SONY XC-75 camera, or equivalent. The camera is mounted on the robot arm to permit the accuracy of part placing to be verified. This system can accurately determine the position of each part or element in electrode stack assembly 225 to within 0.01 millimeters. Once all layers have been placed in the stacking fixture by the robot arm, the stack is presented for wrapping.

[0117] The foregoing methods permit precise alignment and stacking of separator layers 165 and 180, anode sub-assemblies 170 and cathode layers 175 in electrode stack assembly 225, while minimizing the addition of undesirable unused volume to capacitor 265.

[0118] Another method for assuring registration of separator layers 165 and 180, anode sub-assembly 170 and cathode layer 175 in electrode stack assembly 225, involves alignment elements disposed within the stacking fixture are employed in a manual process which utilizes fixture registration points. In such a method, the stacking fixture has several alignment elements such as posts or side walls disposed about its periphery for positioning separator layers 165 and 180. Because cathode layers 175 and anode sub-assemblies 170 do not extend to the periphery of the separator, an alternative means for accurately positioning those electrodes becomes necessary.

[0119] Positioning of alternating cathode layers 175 and anode sub-assemblies 170 is most preferably accomplished using alignment elements such as posts or side walls disposed about the periphery of cathode tab 176 and anode tab 195. It has been discovered that the accuracy of layer placing and positioning is primarily a function of the length of the electrode tabs. The longer the tab, the less significant the alignment error becomes. Electrode tab length must typically be balanced against the loss of electrode material which occurs during die cutting, which in turn results primarily due to the longer length of cathode tab 176 in respect of the length of anode tab 195. Tabs 176 and 195 may include or contain alignment features therein having any suitable geometry for facilitating registration and positioning in respect of alignment elements. Any additional tab length utilized for registration of the electrode layers is most preferably trimmed from electrode stack assembly 225 during the process of electrode tab interconnection (more about which we say below).

[0120] Another method for ensuring registration of separator layers 165 and 180, anode sub-assembly 170 and cathode layer 175 in electrode stack assembly 225 which does not require the use of internal alignment elements within capacitor 265 is enveloping or covering anode sub-assembly 170 and cathode layer 175 with separator material. In this method, separator layers 180a and 180b are combined into a single die cut piece part that is folded around either anode sub-assembly 170 or cathode layer 175. The free edges of the separator are then secured by doubled-sided transfer tape, another adhesive, stitching or ultrasonic paper welding. Construction of an electrode sub-assembly in this manner secures and registers anode sub-assembly 170 and cathode layer 175 in respect of the periphery of the separator envelope so formed. The resulting anode/cathode sub-assembly 227 is then presented for stacking in electrode stack assembly 225.

[0121] Preferably, separator layer 165 or 180 is pressed into a surface of anode sub-assembly 170 or anode layer 185 over a localized region thereof with sufficient force to rigidly affix the separator paper to anode sub-assembly 170, but not with such great force that a portion of underlying anode sub-assembly 170 is fractured. Other methods of securing all or portions of separator layer 165 or 180 to anode sub-assembly 170 or anode layer 185 include, but are not limited to, stitching, adhesive bonding and ultrasonic paper welding techniques.

[0122] FIG. 7 shows a top perspective view of one embodiment of an electrode stack assembly 225 of capacitor 265. FIG. 8 shows an enlarged view of a portion of the electrode stack assembly 225 of FIG. 7. After wrapping electrode stack assembly 225 with outer wrap 115 and wrapping tape 245, interconnection of gathered anode tabs 232 and gathered cathode tabs 233 with their respective external terminals is most preferably made.

[0123] FIG. 9 shows an exploded top perspective view of one embodiment of a capacitor 265 employing the electrode stack assembly of FIGS. 6, 7 and 8 therein. This embodiment includes anode feedthrough 120 and cathode feedthrough 125 most preferably having coiled basal portions 121 and 126, respectively. Feedthroughs 120 and 125 provide electrical feedthrough terminals for capacitor 265 and gather gathered anode tabs 232 and gathered cathode tabs 233 within basal portions 121 and 126 for electrical and mechanical interconnection.

[0124] FIG. 16 shows a flow chart corresponding to one method of making tab interconnections and feedthrough terminal connections. Feedthrough wire is first provided for construction of feedthroughs 120 and 125, as shown in FIGS. 9 and 10. A preferred feedthrough wire is aluminum having a purity greater than or equal to 99.99% and a diameter of 0.020 inch (0.510 mm). Wire is trimmed to predetermined lengths for use in anode feedthrough 120 or cathode feedthrough 125. One end of the trimmed wire is coiled such that its inside diameter or dimension is slightly larger than the diameter or dimension required to encircle gathered anode tabs 232 or gathered cathode tabs 233.

[0125] Gathered anode tabs 232 are next gathered, or brought together in a bundle by crimping, and inside diameter 131 of anode feedthrough coil assembly 120 is placed over gathered anode tabs 232 such that anode feedthrough pin 130 extends outwardly away from the base of gathered anode tabs 232. Similarly, gathered cathode tabs 233 are gathered and inside diameter 136 of cathode feedthrough coil assembly 125 is placed over gathered cathode tabs 233 such that cathode feedthrough pin 135 extends outwardly away from the base of cathode tab 233. Coiled basal portions 121 and 126 of anode and cathode feedthroughs 120 and 125 are then most preferably crimped onto anode and cathode tabs 232 and 233, followed by trimming the distal ends thereof, most preferably such that the crimps so formed are oriented substantially perpendicular to imaginary axes 234 and 235 of gathered anode and cathode tabs 232 and 233. Trimming the distal ends may also, but less preferably, be accomplished at other non-perpendicular angles respecting imaginary axes 234 and 235.

[0126] In some preferred methods, a crimping force is applied to feedthrough coils 121 and 126 and tabs 232 and 233 throughout a subsequent preferred welding step. In one method, it is preferred that the crimped anode and cathode feedthroughs be laser or ultrasonically welded along the top portion of the trimmed edge of the distal ends to anode and cathode tabs 232 and 233. Following welding of feedthroughs 120 and 125 to gathered anode tabs 232 and gathered cathode tabs 233, respectively, pins 130 and 135 are bent for insertion through feedthrough holes 142 and 143 of case 90.

[0127] Many different embodiments of the feedthroughs and means for connecting the feedthrough pins to anode and cathode tabs exist other than those shown explicitly in the figures. For example, feedthrough embodiments can comprise basal portions having open sides, forming “U” or “T” shapes in cross-section, forming a coil having a single turn of wire, forming a coil having three or more turns of wire, formed from flattened wire, or basal portions formed from crimping sleeves or layers of metal for connecting feedthrough pins 130 and 135 to anode and cathode tabs 232 and 233.

[0128] FIG. 17 shows a flow chart corresponding to one method for making tab interconnections and feedthrough connections. In this method, anode feedthrough 120 and cathode feedthrough 125 have no coiled portions. Gathered anode tabs 232 and gathered cathode tabs 233 are gathered and trimmed, followed by the basal portions of anode and cathode feedthroughs 120 and 125 being placed near to gathered anode tabs 232 and gathered cathode tabs 233, respectively. The basal portions of feedthroughs 120 and 125 are then joined to gathered anode tabs 232 and gathered cathode tabs 233, respectively, most preferably by ultrasonic welding means.

[0129] In yet another method, the basal portions of feedthroughs 120 and 125 are flattened to facilitate welding to anode and cathode tabs 232 and 233. In still another method, the basal portions of feedthrough pins 130 and 135 are formed such that they engage gathered anode tabs 232 or gathered cathode tabs 233 around the periphery of the tabs by means other than coiling. For example, basal portions 121 and 126 of feedthroughs 120 and 125 may be “flag shaped,” and the flag portions thereof may be wrapped around tabs 232 and 233. Feedthrough pins 130 and 135 may alternatively be attached to anode and cathode tabs 232 and 233 with resistance welds, cold welds, brazing, friction welds, or an additional feedthrough component such as a crimping sleeve may capture and join tabs 232 and 233 for providing electrical and mechanical connections thereto.

[0130] It has been discovered that the processes of forming electrical connections between tabs 232 and 233 and feedthrough coil assemblies 120 and 125 can introduce undesirable stress on tabs 176 and 195. The resultant strain induced in those tabs has further been found to manifest itself as tears in cathode layer 175 at the base of cathode tab 176, or as fractures in relatively low strength cold welds 205 or 210 within anode sub-assembly 170. One advantage of the coiled portions of feedthroughs 120 and 125 is that they can provide strain relief between feedthrough pins 130 and 135 and tabs 232 and 233. Thus, the strain relief features of feedthroughs 120 and 125 help minimize or eliminate undesirable stress in feedthrough connections.

[0131] As employed in the specification and claims hereof, the term “laser welding” means, but is not necessarily limited to, a method of welding wherein coherent light beam processing is employed. Other means of coherent light beam processing include electron beam or laser welding methods (e.g., Nd:YAG, CO₂ processes) having hard or fiber optic beam delivery in pulsed, continuous, or q-switched modes. Still other welding means and processes, such as micro metal inert gas welding and micro plasma welding processes, can be employed.

[0132] Table 2 sets forth optimized, preferred processing parameters under which various components of capacitor 265 are laser welded to one another. Table 3 sets forth a range of parameters under which the same type of laser welding system provides acceptable weld characteristics in accordance with other methods. The parameters set forth in Tables 2 and 3 correspond to those for a Model No. JK702H pulsed Nd:YAG laser welding system having hard optic beam delivery manufactured by Lumonics Laserdyne of Eden Prairie, Minn.

[0133]

TABLE 2
Optimized Nd: YAG Laser Welding Parameters
	Optimized Laser Welding Parameters*
	Energy		Feed		Argon
	per Pulse	Pulse	Rate	Pulse	Cover
	(Joules/	Frequency	(inches/	Width	Gas
Weld Type	pulse	(Hertz)	min)	(msec)	(SCFH)
Feedthrough Ferrule	13.5	4.5	3	5	35
to Case Tack 1
Feedthrough Ferrule	9.75	20	2	4.5	35
to Case Weld
Fill port Ferrule to	13.5	4.5	3	5	35
Case Tack 1
Fill port Ferrule to	15	15	2	6	35
Case Weld
Anode Feedthrough	8	10	2	5	35
Tabs
Cathode Feed-	4	10	2	5	35
through Tabs
Cover to Case	7.5	40	6	5.4	60
*Lumonics JK702H Nd: YAG laser having an initial beam diameter of approximately 1.0 inch (2.54 cm) passing through a final focusing lens with a 146 mm focal length (purchased having “160 mm lens”, actual fine focal point measured was 146 mm) and a spot size at the joint surface of 0.022 inch (0.5060 mm) diameter. The cover gas was coaxial. It will be understood that variations respecting the manufacturer of the laser, beam delivery optics, the initial beam

[0134]

TABLE 3
Generalized Nd: YAG Laser Welding Parameters
	Optimized Laser Welding Parameters*
	Energy		Feed		Argon
	per Pulse	Pulse	Rate	Pulse	Cover
	(Joules/	Frequency	(inches/	Width	Gas
Weld Type	pulse	(Hertz)	min)	(msec)	(SCFH)
Feedthrough Ferrule	2–15	3–30	1–5	3.5–8	30–60
to Case
Fill port Ferrule	2–15	3–30	1–5	3.5–8	30–60
to Case
Feddthrough	1–10	1–10	1–7	3.5–8	30–60
Tabs
Cover to Case	5–25	10–40	1–7	3.5–8	30–60
*Lumonics JK702H Nd: YAG laser having an initial beam diameter of approximately 1.0 inch (2.54 cm) passing through a final focusing lens with a 146 mm focal length (purchased having “160 mm lens”, actual fine focal point measured was 146 mm) and a spot size at the joint surface of 0.022 inch (0.560 mm) diameter. The cover gas was coaxial. It will be understood that variations respecting the manufacturer of the laser, beam delivery optics, the initial beam size, final focusing lens, spot size ofthe beam and the like may be employed in the welding processes.

[0135] FIG. 10 shows an exploded top perspective view of capacitor 265 of FIG. 9 in a partially assembled state. FIG. 18 shows a flow chart of one method of making case sub-assembly 108. Case 90, anode ferrule 95, cathode ferrule 100, and fill port ferrule 105 are first provided. Case 90 contains a means for accepting anode ferrule 95 therein, shown in FIGS. 9 and 10 as anode feedthrough hole or opening 142. Case 90 further contains a means for accepting cathode ferrule 100, shown in FIGS. 9 and 10 as cathode feedthrough hole or opening 143. Case 90 also includes a means for accepting fill port ferrule 105, shown in FIGS. 9 and 10 as fill port hole 139.

[0136] In a preferred embodiment of capacitor 265, the case 90 and cover 110 are formed of aluminum. In other embodiments, case 90 or cover 110 may be formed of any other suitable corrosion-resistant metal such as titanium or stainless steel, or may alternatively be formed of a suitable plastic, polymeric material or ceramic.

[0137] Case 90, cover 110 and capacitor 265 may additionally form a case negative capacitor (where case 90 and cover 110 are electrically connected to the cathode layers and are therefore at the same electrical potential as the cathode layers, i.e., at negative potential), or a floating case capacitor (where case 90 and cover 110 are electrically connected neither to the cathode layers nor to the anode sub-assemblies, and where case 90 and cover 110 are at substantially no electrical potential or at an electrical potential that floats with respect to the respective potentials of the cathode layers and the anode sub-assemblies). In some embodiments, case 90 or cover 110 may be formed of an electrically non-conductive material or substantially electrically non-conductive material such as a suitable plastic, polymeric or ceramic material.

[0138] Ferrules 95, 100 and 105 are most preferably welded to case 90 (or otherwise attached thereto such as by a suitable epoxy, adhesive, solder, glue or the like), and together comprise case sub-assembly 108. Radial flanges in anode ferrule 95 and cathode ferrule 100 provide a region for making a lap joint between the side wall of case 90 and around the perimeters of feedthrough ferrule holes 142 and 143. In preferred methods, a circumferential laser weld is made in the circumferential joint between the ferrules and the case side wall 92, and welding is carried out in two primary steps. First, a series of tack welds is made around the circumference of the joint. The tack welds are most preferably made either by making adjoining, successive tack welds around the perimeter or by making a first tack weld at a first location along the perimeter, making a second weld diametrically opposed from the first weld along the perimeter, making a third weld adjacent to the first weld, making a fourth weld adjacent to the second weld, and so on. Finally, a final closing weld is made around the hole perimeter to hermetically seal tack welded joint 93.

[0139] Table 2 sets forth an optimized set of parameters under which anode ferrule 95 and cathode ferrule 100 are joined to case 90. Table 3 sets forth a range of general parameters under which the same laser welding system provides acceptable weld characteristics for joining anode ferrule 95 and cathode ferrule 100 to case 90.

[0140] FIG. 18 shows steps for inserting anode wire guide 140 into the inside diameter of anode ferrule 95, and inserting cathode wire guide 141 into the inside diameter of cathode ferrule 100. Wire guides 140 and 141 center pins within the inside diameter of the ferrules to permit anode and cathode pins 130 and 135 to be electrically insulated from the inside surface of case 90, anode ferrule 95, and cathode ferrule 100. Wire guides 140 and 141 may themselves be electrically insulating, and electrical insulation of pins 130 and 135 from case 90 and other components is most preferably enhanced by means of potting adhesive 160. FIG. 20 shows further details concerning one method for forming electrical insulation between pins 130 and 135 and anode ferrule 95 and cathode ferrule 100.

[0141] Wire guides 140 and 141 most preferably contain annular, ramped, or “snap-in” features formed integrally therein. Those features prevent wire guides 140 and 141 from being pushed out of their respective ferrules during handling, but are most preferably formed such that insertion of wire guides 140 and 141 in their corresponding ferrules may occur using forces sufficiently low so as not to damage case 90 or ferrules 95 or 100 during the inserting step.

[0142] Wire guides 140 and 141 may be formed from any of a wide variety of electrically insulating materials that are stable in the environment of an electrolytic capacitor. In one preferred embodiment, the material from which wire guides 140 and 141 is made is an injection molded polysulfone known as AMOCO UDEL supplied by Amoco Performance Products of Atlanta, Ga. In other embodiments, wire guides 140 and 141 may be formed from other chemically resistant polymers such as fluoroplastics (e.g., ETFE, PTFE, ECTFE, PCTFE, FEP, PFA or PVDF), fluoroelastomers, polyesters, polyamides, polyethylenes, polypropylenes, polyacetals, polyetherketones, polyarylketones, polyether sulfones, polyphenyl sulfones, polysulfones, polyarylsulfones, polyetherimides, polyimides, poly(amideimides), PVC, PVDC-PVC copolymers, CPVC, polyfurans, poly(phenylene sulfides), epoxy resins, silicone elastomers, nitrile rubbers, chloroprene polymers, chlorosulfonated rubbers, polysulfide rubbers, ethylene-polypropylene elastomers, butyl rubbers, polyacrylic rubbers, fiber-reinforced plastics, glass, ceramic and other suitable electrically insulating, chemically compatible materials.

[0143] As used in the specification and claims hereof, the foregoing acronyms have the following meanings: the acronym “ETFE” means poly(ethylene-co-tetrafluoroethylene); the acronym “PTFF” means polytetrafluoroethylene, the acronym “CTFE” means poly(ethylene-co-chlorotrifluoroethylene); the acronym “PCTFE” means polychlorotrifluoroethylene; the acronym “FEP” means fluorinated ethylene-propylene copolymer; the acronym “PFA” perfluoroalkoxy fluoropolymer; the acronym “PVDF” means polyvinylidene fluoride; the acronym “PVC” means polyvinyl chloride; the acronym “PVDC-PVC” means polyvinylidene chloride-polyvinyl chloride copolymer; and the acronym “CPVC” means chlorinated polyvinyl chloride.

[0144] FIG. 11 shows a top view of one embodiment of assembled capacitor 265 with cover 110 not present. In one embodiment, the head space portion of electrode stack assembly 225 (referred to herein as head space 230) is insulated from case 90 and cover 110. The means by which head space insulation may be provided include molded, thermally-formed, die cut, or mechanically formed insulating materials and means, where the materials and means are stable in the environment of an electrolytic capacitor. Suitable materials from which head space insulators may be formed include all those listed hereinabove respecting materials for forming wire guides 140 and 141. Another means of providing head space insulation is to wrap electrically insulating tape, similar to wrapping tape 245, around head space 230 to prevent the anode or cathode terminals from contacting case 90 or cover 110 or each other.

[0145] FIGS. 26(a) through 26(p) show different embodiments of joint 93 and the crimp and joint configurations for joining the cover 110 to case 90. The particular structural configuration of joint 93 is of the utmost importance in respect of the suitable laser weldability thereof. More particularly, joints for covers of prior art flat capacitors having metal cases and covers and conventional joint structures generally permit laser energy to enter the interior case chamber of capacitor 265 through the joints formed between the covers and cases thereof thereby damaging or heating up components disposed inside case 90. Laser energy penetrates into the interior case chamber of capacitor 265 due to a joint geometry wherein a straight or substantially straight line of sight or portion existed or was disposed through the joint between the interior of the capacitor and the exterior of the capacitor. Joints having no such straight line of sight or portion through the joint between the exterior and interior of the capacitor were found to eliminate or at least diminish substantially the ill effects attending laser energy penetration to the interior of the capacitor.

[0146] In one preferred embodiment to avoid this problem, case 90, cover 110, joint 93, upper edge 94, raised portion 95, stepped portion 96, groove 97, stepped portion 98 and outer edge 111 cooperate with one another to cause laser energy entering joint 93 from the exterior of capacitor 265 to be reflected or scattered to the outside of capacitor 265, and further to be contained or absorbed within joint 93 in such a manner that no or substantially no laser energy penetrates joint 93 and enters the interior of capacitor 265 while simultaneously forming a suitable weld in joint 93 between case 90 and cover 110. This absorption, containment, backscattering or reflecting of laser energy by joint 93 results at least partially from the multiple orientations of joint 93 as it wends its way from the exterior of capacitor 265 to the interior thereof. In other words, and as illustrated in FIGS. 26(a) through 26(p), joint 93 has multiple portions that are bent, non-parallel or serpentine respecting one another.

[0147] In one preferred method for laser welding joint 93, an axis of a laser beam is directed inwardly along or parallel to the surfaces defining a first portion of joint 93 (e.g., parallel to imaginary axis 102 or imaginary axis 101, depending on the particular embodiment). Upon entering the first portion of joint 93 or a region near thereto, the laser beam encounters at least a second portion of joint 93 defined by surfaces that are bent or not parallel respecting the surfaces defining the first portion of joint 93. As shown in FIGS. 26(e) through 26(h) and 26(m) through 26(p), joint 93 may also have a third portion defined by surfaces that are bent or non-parallel respecting the surfaces defining the second portion of joint 93. Consequently (providing appropriate parameters are selected by a user for operating the laser welding system), no portion of the laser beam impinging upon the first portion of joint 93 may penetrate joint 93 sufficiently far such that the laser beam reaches the interior case chamber of capacitor 265 without first being absorbed, reflected or scattered.

[0148] In another preferred method for laser welding joint 93, an axis of a laser beam is directed inwardly along or parallel to the surfaces defining a second portion of joint 93 (e.g., parallel to imaginary axis 102 or imaginary axis 101, depending on the particular embodiment). Upon entering the second portion of joint 93 or a region near thereto, the laser beam encounters at least a first or third portion of joint 93 defined by surfaces that are bent or not parallel respecting the surfaces defining the second portion of joint 93.

[0149] FIGS. 26(a) through 26(d) show a first embodiment of joint 93 and the crimp, wherein case 90 has inner and outer side walls 91 and 92 extending upwardly from a flat planar base of case 90 to form an open end that terminates in upper edge 94 disposed between inner and outer side walls 91 and 92. Upper edge 94 most preferably comprises at least one stepped portion 96 and at least one raised portion 95. Substantially planar cover 110 seals the open end of the case, cover 110 having upper and lower surfaces 112 and 113, respectively, separated by outer edge 111. At least portions of outer edge 111 are shaped to engage at least one stepped portion 96 of upper edge 94 such that cover 110 self-registers on case 90 when cover 110 is disposed over the open end of case 90, outer edge 111 is aligned approximately upper edge 94, and cover 90 is placed thereon.

[0150] As shown in FIGS. 26(a) and 26(c), at least one raised portion 95 of upper edge 94 initially extends above upper surface 112 of cover 110 when at least portions of outer edge 111 are placed on and engage at least one stepped portion 96. As shown in FIGS. 26(b) and 26(d), at least one raised portion 95 is crimped or folded inwardly over or along upper surface 112 of cover 110 to form joint 93 after at least portions of outer edge 111 are placed on and engage the least one stepped portion 96. Next joint 93 is laser welded to hermetically seal cover 110 to case 90.

[0151] In the laser welding step, the laser beam may be directed substantially parallel to axes 101 and 102 of FIG. 26(b) to form a weld in the first or second portions of joint 93. Alternatively, the laser beam may be directed substantially parallel to axis 101 of FIG. 26(a) (i.e., substantially parallel to upstanding side walls 91 and 92) after raised portion 95 is crimped over cover 110 such that at least portions of raised portion 95 melt and thereby weld first, second, third or other portions of joint 93 closed. The preferred laser welding methods also include laser welding steps where the laser beam is oriented in directions other than those set forth explicitly above.

[0152] In FIGS. 26(a) and 26(c), imaginary axes 101 and 102 are oriented at an angle φ of about 90 degrees respecting one another, where imaginary axis 101 defines the initial orientation of upper edge 94 and imaginary axis 102 defines the orientation of the plane within which cover 110 is disposed. In FIGS. 26(b) and 26(d), after upper edge 94 has been crimped or folded inwardly over or along upper surface, imaginary axis 101 is oriented at an angle φ of about 0 degrees respecting imaginary axis 102.

[0153] FIGS. 26(e) through 26(f) show a second embodiment of the crimp and joint 93, where case 90 has inner and outer side walls 91 and 92, respectively, extending upwardly from a flat planar base of case 90 to form an open end terminating in upper edge 94 disposed between inner and outer side walls 91 and 92. Substantially planar cover 110 seals the open end of case 90. Cover 110 comprises upper and lower surfaces 112 and 113, respectively, separated by outer edge 111. Lower surface 113 of cover 110 has disposed thereon at least one of groove 97 (see FIGS. 26(e) and 26(f)) and stepped portion 98 (see FIGS. 26(g) and 26(h)). Groove 97 or stepped portion 98 is disposed radially inward from outer edge 111.

[0154] At least portions of groove 97 or stepped portion 98 are shaped to engage corresponding portions of upper edge 94 such that groove 97 or stepped portion 98, in combination with upper edge 94, cause cover 110 to self-register on upper edge 94 when cover 110 is disposed over the open end of case 90. Groove 97 or stepped portion 98 is aligned approximately with upper edge 94 when cover 110 is placed on upper edge 94. Outer portions 117 of cover 110 extending between outer edge 111 and groove 97 or stepped portion 98 are crimped or folded downwardly over at least portions of outer side wall 92 of case 90 to form joint 93 after cover 110 is placed on the open end of case 90. Joint 93 is laser welded to hermetically seal cover 110 to case 90.

[0155] In the laser welding step, the laser beam may be directed substantially parallel to axes 101 and 102 of FIG. 26(f) to form a weld in the first, second or other portions of joint 93. Alternatively, the laser beam may be directed substantially parallel to axis 102 of FIGS. 26(e) or 26(g) (i.e., substantially parallel to the plane forming cover 110) after outer portion of cover 110 is crimped over outer side wall 92 such that at least portions of outer portions of cover 110 melt and thereby weld first, second, third or other portions of joint 93 closed. The laser beam may be oriented in directions other than those set forth explicitly above.

[0156] In FIGS. 26(e) and 26(g), imaginary axes 101 and 102 are initially oriented at an angle φ of about 90 degrees respecting one another, where imaginary axis 101 defines the orientation of upper edge 94 and imaginary axis 102 defines the initial orientation of outer edge 111. In FIGS. 26(f) and 26(h), after outer edge 111 has been crimped or folded downwardly over at least portions of outer side wall 92, imaginary axis 102 is oriented at an angle φ of about 0 degrees respecting imaginary axis 102.

[0157] FIGS. 26(i) through 26(p) show yet other preferred embodiments of the crimp and joint, where the angle φ defining the orientations of imaginary axes 101 and 102 respecting one another after upper edge 94 has been crimped or folded inwardly, or outer edge 111 has been crimped or folded downwardly, is greater than or equal to 0 degrees but less than 90 degrees. The embodiments shown in FIGS. 26(i) through 26(p) have been discovered to be particularly efficacious for providing good access to joint 93 for a laser welding beam.

[0158] Note, however, that many variations of the particular cover, case and joint geometries disclosed explicitly herein are possible. For example, the case 90 and cover 110 may form two aluminum-containing half-cases having upwardly and downwardly extending side walls, the two half-cases forming two open ends that are subsequently laser welded together. Alternatively, the case 90 and cover 110 may form two substantially planar aluminum-containing members separated by a single or multiple side wall members, the two planar members being laser welded to the intervening side wall members.

[0159] FIGS. 26(a) through 26(p) also show registration marks or alignment features 99 disposed on case 90 or cover 110. Registration mark or alignment feature 99 is employed to establish a reference position in joint 93 for the welding apparatus after the case or cover has been crimped or folded, thereby ensuring precise position of the welding apparatus in respect of case 90, cover 110 and joint 93 when a weld is being formed in joint 93. It has been discovered that optimum results are obtained when registration mark 99 is disposed on upper surface 112 of cover 110.

[0160] FIG. 19 shows a flow chart according to one method for sealing case 90 and cover 110. Case sub-assembly 108 is provided with electrode stack assembly 225 inserted in the interior case chamber of case 90. Cover 110 is disposed atop upper edge 94 formed in the side wall of case 90. In one case side wall edge configuration, raised portion 95 of upper edge 94 extends about 0.014 inches (0.355 mm) above upper surface 112 of cover 110 when cover 110 is placed on upper edge 94. The assembly is placed within a crimping mechanism or nest, and a clamp is actuated to hold cover 110 against upper edge 94 and stepped portion 96. The crimping mechanism is actuated to crimp or fold over inwardly raised portion 95 onto, along or over upper surface 112 of cover 110.

[0161] In another preferred method, crimping of raised portion 95 is accomplished using a die cut to the shape of case 90 and further having angled or ramped side walls for engaging and pressing inwardly raised portion 95 over upper surface 112 of cover 110. A crimp may also be formed with a moving crimp apparatus that travels around the perimeter of case 90 while continuously crimping raised portion 95 over upper surface 112 of cover 110. The foregoing methods may be readily adapted to permit the crimping or folding of edge 111 of cover 110 downwardly over outer side wall 92.

[0162] Crimping of raised portion 95 onto cover 110 or outer edge 111 onto side wall 92 provides several advantages. First, laser welding of cover 110 to case 90 may be accomplished using relatively simple tooling, thereby resulting in short process times. Laser welding often provides a bottleneck in manufacturing process flow when components such as case 90 and cover 110 typically must be aligned precisely respecting one another. The elimination of such alignment steps during the laser welding process has been discovered to help eliminate manufacturing process bottlenecks. Folding or crimping raised portion 95 or outer edge 111 prevents a laser beam from entering the interior of capacitor 265. Instead, a laser beam is forced to couple with the material of case 90 and cover 110 to thereby induce melting. It was discovered that joints 93 not having crimps forming at least a portion thereof may permit a laser beam to damage components inside capacitor 265.

[0163] Another advantage of the crimped joint is that the crimp provides additional metal in the weld zone. Aluminum, having a high thermal expansion coefficient, is sensitive to cracking upon rapid cooling from the high temperatures characteristic of welding processes. The additional metal provided by the crimp decreases cracking sensitivity in joint 93. Joint 93 is formed such that imaginary axes 101 and 102 are oriented at an angle φ respecting one another where φ is less than 90 degrees but greater than or equal to 0 degrees. It is notable that crimping of case 90 and cover 110 to one another helps registration of case 90 and cover 110 in respect of one another prior to the welding of at least portions of joint 93 being undertaken.

[0164] Crimped case 90 and cover 110 are next removed from the crimp fixture and placed in a welding fixture. A laser weld is made in joint 93 to hermetically seal case 90 to cover 110. Table 2 sets forth an optimized set of parameters under which the crimped case/cover joint may be sealed using a pulsed Nd:YAG laser welding system. Table 3 sets forth a generalized range of conditions under which the same laser welding system provides acceptable results.

[0165] In a preferred method, machined, stamped, etched or otherwise-formed registration marks or alignment features 99 are disposed on cover 110 or case 90 to permit the relative positions of cover 110 and case 90 to be determined precisely for the laser welding step. Connectors are then attached to the welded case/electrode stack assembly.

[0166] FIG. 20 shows a flow chart according to one method for sealing anode feedthrough portion 236 and cathode feedthrough portion 240 of capacitor 265. See also FIG. 10. FIGS. 9 through 12 show various embodiments of the sealing and connector attachments that can be employed in capacitor 265.

[0167] FIG. 21 shows several top, perspective and cross-sectional views according to one embodiment of capacitor connector block 145. In one preferred embodiment, connector block 145 is disposed atop or otherwise connected to case 90 and/or cover 110, and has wire harness 155 attached thereto and potting adhesive disposed therein. However, the particular configuration of connector block 145 and its method of fabrication does not play a role in the practice of the present invention.

[0168] A preferred material for forming connector block 145 is an injection molded polysulfone known as AMOCO UDEL supplied by Amoco Performance Products of Atlanta, Ga. Connector block 140 may also be formed from any suitable chemically resistant thermoplastic polymers such as a flouroplastic (e.g., ETFE, PTFE, ECTFE, or PCTFE, FEP, PFA, PVDF), polyester, polyamide, polyethylene, polypropylene, polyacetal, polyarylketone, polyether sulfone, polyphenyl sulfone, polysulfone, polyarylsulfone, polyetherimides, polyimide, poly(amide-imide), PVC, PVDC-PVC copolymer, CPVC, polyfuran, poly(phenylene sulfide), epoxy resin and fiber reinforced plastic.

[0169] In one embodiment, connector block 145 is placed on anode ferrule 95 and cathode ferrule 100 by guiding anode feedthrough pin 130 through connector block anode feedthrough hole 300, and then guiding cathode feedthrough pin 135 through connector block cathode feedthrough hole 305. Connector block 145 is next seated flush against the exterior surface of case 90. Anode feedthrough pin 130 is then inserted into anode crimp tube 150b of wire harness 155. Cathode feedthrough pin 135 is then inserted into cathode crimp tube 150a of wire harness 155. Crimp tubes 150a and 150b are then crimped to feedthrough pins 130 and 135.

[0170] In other preferred embodiments, electrical connections in connector block 145 may be established using techniques such as ultrasonic welding, resistance welding and laser welding. In such joining techniques, the joint geometry may also be a cross-wire weld between feedthrough wire 130 or 135 and harness wire 151 or 152.

[0171] The distal or basal portions of crimp tubes 150a and 150b are crimped on insulated anode lead 151 and insulated cathode lead 152, respectively. Insulated leads 151 and 152 are likewise connected to terminal connector 153. Terminal connector 153 may most preferably be connected to electronics module 360. Standard methods of making aluminum electrolytic capacitors do not lend themselves readily to very small crimp connections, especially in miniaturized ICD designs. A preferred method permits small crimp connections and interconnection means to be formed, and further permits highly efficient packaging in ICD IPG 10.

[0172] In the preferred method described above, connector block 145 and epoxy adhesive provide strain relief to feedthrough pins 130 and 135 and to the feedthrough wire crimp connections, and further provide an epoxy seal between wire guides 140 and 141, case 90 and ferrules 95 and 100. The crimp tubes may also serve as a connection point for device level assembly. Alternatively, the crimp tubes may be integrated within wire harness 155 prior to capacitor assembly. The wire harness may then serve as a means of routing capacitor electrical connections as desired in, for example, device level assembly steps. In the embodiment shown in FIG. 11, terminal connector 153 forms the female end of a slide contact. In another embodiment, terminal connector 153 is connected to other modules by resistance spot welding, ultrasonic wire bonding, soldering, crimping, or other attachment means.

[0173] Referring again to FIG. 21, insulated anode lead 151 is inserted into anode block channel 310. Anode feedthrough pin 130 is centered in connector block anode feedthrough hole 300 by anode pin block guide 320. Insulated cathode lead 152 is inserted into cathode block channel 315. Cathode feedthrough pin 135 is centered in connector block cathode feedthrough hole 305 by cathode pin block guide 325. Centering of the pin through the ferrule assures that the pin does not contact the conducting wall of the ferrule, and also permits a more concentric epoxy seal to be formed around the pin. Centering of the pin may also be accomplished through means disposed in or on the epoxy dispensing or curing tools. Once the epoxy has hardened sufficiently, the centering tool is removed.

[0174] When employed, a potting adhesive is mixed and dispensed through connector block feedthrough holes 300 and 305 and block channels 310 and 315. Such an adhesive may also be dispensed through connector block hole 330 between connector block 145 and case 90. Adhesive bonding between block 145 and case 90 enhances structural stability of capacitor 265. The epoxy is then cured and capacitor 265 is filled with electrolyte.

[0175] The life of capacitor 265 may be appreciably shortened if solvent vapor or electrolyte fluid escapes from the interior of capacitor 265. Moreover, if capacitor 265 leaks electrolyte, the electrolyte may attack the circuits to which capacitor 265 is connected, or may even provide a conductive pathway between portions of that circuit. The present invention provides a beneficial means for preventing the escape of solvent and solvent vapor from capacitor 265. More particularly, capacitor 265 most preferably includes hermetic laser welded seams between joint case 90 and cover 10, and between ferrules 95, 100, and 105 and case 90. Additionally, anode feedthrough portion 236 and cathode feedthrough portion 240 most preferably have an adhesive seal disposed therein for sealing the ferrule walls and the feedthrough wires.

[0176] The epoxy adhesive or potting material is most preferably chemically resistant to the electrolyte employed in capacitor 265 and adheres well to surrounding surfaces. Adhesion promotion (such as by chemical deposition, etching, corona or plasma treatment of the polymeric wire guide of a polymeric case) may be employed to maximize the reliability of capacitor 265. In one preferred embodiment, an aliphatic epoxy such as CIBA-Geigy Araldite 2014 is employed. Other suitable potting adhesives include chemically resistant thermoplastic hot melt materials such as polyamides, polyesters, polyurethanes, epoxies, and polyethylenevinyl acetates, UV curable resins such as acrylates and methacrylates, and other thermosetting resins such as aliphatic and aromatic epoxies, silicones, polyamides, polyesters and polyurethanes. Many suitable potting adhesives may be thermally cured or cured with ultraviolet light. A focused IR procedure may be employed in some instances to minimize cure time and localize heat.

[0177] Since hermeticity is desirable in feedthrough assemblies, the method by which the feedthrough seals are made should be predictable, uniform, reliable and produce high quality hermetic seals. In a preferred method, an epoxy adhesive is employed which has few or no voids and cracks and completely or substantially completely adheres to the surrounding pin, ferrule wall and wire guide components. Filling of the ferrule hole with sealing adhesive may be accomplished in several ways, depending largely on the viscosity of the potting agent selected. A balance in viscosity characteristics of the sealing adhesive has been found to be desirable. More particularly, it is desired that the sealing adhesive be thin enough to fill without voids forming and to wet the surface, yet thick enough not to escape around or through the wire guide. The potting adhesive may be B-staged and inserted as a plug; likewise a hot melt adhesive may be applied in similar fashion. Subsequent heating completes curing of the sealing adhesive. In a preferred method, CIBA-Geigy Araldite 2014 epoxy is mixed with a static mix tube and dispensed within 45 minutes. The assembly is cured in an oven for 30 minutes at 90 degrees Celsius.

[0178] Connector block 145, ferrules 95 and 100, and wire guides 140 and 141 can also be formed from a single molded component formed of a suitable chemically resistant thermoplastic or thermoset material that is sealed to case 90 using a potting adhesive. Channels or voids may be included in the basal portions of connector block 145 to permit potting adhesive to flow between those basal portions and case 90. Such a seal between the case and connector block 145 may replace the aforementioned laser welded seal between the ferrule and the case. Such a sealing method eliminates the requirement for several components and removes several processing steps, leading perhaps to significant manufacturing cost reductions.

[0179] Referring again to FIG. 13, after the welding steps are completed, capacitor 265 is filled with electrolyte through a fill port welded into a hole in the side wall of the capacitor case, and the fill port is then closed. The filling is accomplished in a plurality of vacuum impregnation cycles described in reference to FIGS. 22–25. The electrolyte may be any suitable liquid electrolyte for high voltage electrolytic capacitors. In a preferred embodiment of the present invention, the electrolyte is an ethylene glycol based electrolyte having an adipic acid solute. It is contemplated that other liquid electrolytes suitable for use in high voltage capacitors may also be employed.

[0180] During capacitor charging, the ethylene glycol based electrolyte releases hydrogen gas which accumulates within the interior capacitor chamber and eventually can cause the base and cover to bulge outward. In accordance with the present invention, hydrogen gas is released through the fill port 107 while loss of liquid or vaporized electrolyte is prevented.

[0181] FIG. 27(a) shows a top view of capacitor 265 with a portion of cover 90 removed and a portion of electrode stack assembly 225 exposed within the interior capacitor chamber. Fill port 107 shown in FIG. 27(a) comprises a fill port tube shown projecting outwardly from an end of the side wall of capacitor case 90 to an exterior tube end 106 surrounded by a ferrule 105 welded to the case side wall as described above. FIG. 27(b) shows an end view of capacitor 265 of FIG. 27(a), and a corresponding end view of fill port 107 and its ferrule 105. FIGS. 28(a) through 28(c) show various views of one embodiment of liquid electrolyte fill port 107 of the present invention prior to closure or obstruction of fill port lumen 103. Preferably, the fill port 107 is shaped to comprise a fill port tube extending between an interior tube end 104 and an exterior tube end 106 and with a fill port ferrule 105 in the shape of a circular flange extending outwardly from and transversely to the fill port tube and the tube lumen 103. The interior tube end 104 fits through the preformed hole 139 in the side wall of case 90 (or the cover 110) and the ferrule flange 105 is welded to the exterior of the side wall 90.

[0182] It is preferred that fill port 107 be formed of a single integrated piece of metal, although the tube and ferrule 105 may be formed of separate non-integral components and may further be formed of materials other than metal, such as ceramic or plastic. The illustrated fill port ferrule 105 and the fill port tube are formed integrally of tubular metal stock having the tube lumen 103 and a circular flange extending outward of the tube body that is welded against the outer surface of the case or cover when the fill port tube interior end 104 is inserted into the hole 139 in the case side wall or cover

[0183] The height 109 of the fill port tube shown in FIG. 28(c) is most preferably about 0.200 inches with respect to the embodiment of capacitor 265 shown in the drawings hereof, although other heights 109 are contemplated in the present invention, such as 0.065 inches, 0.300 inches, and so on. It is preferred that height 109 be sufficiently great to accommodate a fitting of a helium leaktightness testing apparatus, the fitting being fitted in sealing engagement over the exterior tube end 106.

[0184] In another embodiment of a fill port 107 of the present invention, the side wall of case 90 is formed of a suitable metal, and a fill port tube is extruded from, punched in or otherwise integrally formed in a side wall or other portion of case 90. Such a design eliminates the need for fill port ferrule 105 disposed in a wall or surface of case 90. For example, a tapered punch may be employed to initially punch a small diameter hole in a side wall of case 90, followed by causing the punch to travel through the hole, causing metal from side wall 90 to be extruded outwardly from the side wall, and forming an outwardly projecting cylindrically or otherwise shaped fill port 107.

[0185] In preferred embodiments of the present invention, fill port 107 provides electrolyte filling and helium leak verification capabilities and is easy to hermetically seal when these functions are completed. The hermeticity of capacitor 265 is preferably measured using a helium leak test that. In one type of helium leak testing, a helium leak testing apparatus forms a seal around fill port 107. It is preferred that an O-ring be disposed between the fitting and the fill port 107 as a vacuum of about 50 Tor is pulled on the interior of capacitor 265 through the fill port tube lumen and the gas pulled from the interior of capacitor 265 is directed past a tuned mass spectrometer. Helium gas is then emitted about and around capacitor 265, cover 110, case 90, joint 93 between cover 110 and case 90, connector block 145, ferrule 105, fill port 107 and ferrule 105 and other components while the helium leaktightness testing apparatus tests gas and molecules evacuated from the interior of capacitor 265 for the presence of helium gas which has leaked from the exterior of capacitor 265 into the interior thereof. The leak rate for helium through the materials and joints within capacitor 265 is determined by the mass spectrometer. This measure of leaktightness or hermeticity provides a means of assuring the quality of the welded joint of the cover to the case opening, the feedthrough ferrules to the case side wall and the fill tube ferrule to the case side wall.

[0186] In another type of helium leak testing, “bombing” or filling of the interior chamber of capacitor 265 with helium gas is accomplished immediately prior to sealing of fill port 107. The exterior of the sealed capacitor 265 is then monitored under vacuum conditions with a tuned mass spectrometer to determine the rate of helium leakage past the materials and joints of capacitor 265.

[0187] A tuned mass spectrometer is most preferably included in the helium leaktightness testing apparatus. The spectrometer is sensitive to the presence of helium atoms or molecules.

[0188] An example of such an apparatus is a LEYBOLD INFICON Model No. UL-200 Helium Leaktester manufactured in East Syracuse, N.Y. An O-ring having a leaktightness rating of about 1×10⁻⁹ cm³/sec. is most preferably employed in conjunction with the fill tube and the fitting of the leaktightness testing apparatus. A typical fail point specification for the leaktightness testing apparatus when employed with the capacitor 265 is about 1×10⁻⁹ cm³/sec.

[0189] When hermeticity testing is completed, the fill tube 107 is employed to fill the capacitor case with electrolyte. The capacitor 265 and the electrolyte source are then placed in a vacuum chamber with the exterior tube end 106 of fill port 107 connected to a source of the electrolyte optionally using a temporary fill tube attached thereto. Multiple vacuum impregnation cycles are then performed at pressures exceeding the vapor pressure of the electrolyte described further below. In a less preferred method, capacitor 265 is filled with electrolyte by immersing capacitor 265 in the electrolyte or by vacuum-filing capacitor 265 with a metered filling machine.

[0190] Once capacitor 265 is filled with electrolyte, it is preferred that an aging process be undertaken to form the dielectric aluminum oxide layer. Aging is generally accomplished by applying a current through the capacitor terminals and gradually raising the voltage across those terminals from zero to the peak aging voltage of the capacitor (usually between about 360 and about 390 Volts DC). Once the aging voltage is attained, capacitor 265 is held at that voltage until the leakage current stabilizes at an acceptably low value. It is preferred that capacitor 265 be aged until a voltage of about 370 Volts is attained during a current limiting process.

[0191] In one preferred method, the aging process is carried out with the voltage set at 370 Volts and the current limited to about 1.5 mA (for capacitor 265 having a capacitance of 214 microfarads) while observing leakage current. It is beneficial to increase the temperature of the aging system at higher voltages. In one preferred method, the temperature is increased to about 70 degrees Celsius when the voltage reaches 230 Volts. After charging to 370 Volts, the capacitors are most preferably permitted to continue aging with the voltage held at 370 Volts until the leakage current decreases to a predetermined value, a predetermined time at 370 Volts has elapsed, or until a predetermined rate of decrease in leakage current has been obtained.

[0192] Following aging, post aging vacuum treatment or filling of the capacitor 265 contributes to significant improvements in capacitance and equivalent series resistance (ESR). FIG. 22 is a flow chart describing one method of vacuum treating the aged capacitor. The aged capacitor 265 is placed inside a vacuum chamber and held at 27 inches of mercury for three minutes. The chamber is vented and then held at 27 inches of mercury for three minutes for two additional cycles.

[0193] FIG. 23 shows the steps of a preferred method for a vacuum refilling operation after aging. Aged capacitor 265 is placed inside a vacuum chamber with a temporary fill tube connected to exterior tube end 106 of fill port 107 while being immersed in electrolyte. The chamber is then held at 27 inches of mercury for three minutes and vented. This step is repeated once with the temporary tube immersed in the electrolyte and a second time with the temporary tube out of the electrolyte. The third cycle is intended to draw excess electrolyte from capacitor 265.

[0194] FIG. 24 graphically shows the increase in capacitance of five capacitors following the vacuum refilling operation of FIG. 23. The noted increase in capacitance is on the order of about 1 to about 2 microfarads (˜0.3%). FIG. 25 graphically shows the decrease in ESR of the same five capacitors after the vacuum refilling operation of FIG. 23. The noted decrease in ESR is on the order of about 0.2 ohms (˜20%). The vacuum treatments are believed to remove entrapped gas that evolves during aging and refilling, and are also believed to replace electrolyte lost during aging, thereby permitting the microstructural pores of the anode and separator layers to be substantially fully filled and saturated with electrolyte. Excess electrolyte may also be removed through vacuum cycling with the fill tube pointing downwardly.

[0195] Returning to FIG. 13, after the aging and vacuum refilling cycles are completed, the fill port lumen 103 is sealed with a microporous plug to inhibit the loss of liquid electrolyte but to allow hydrogen gas pressures that build up within the interior capacitor chamber to escape. A hydrogen permeable membrane seal or vent plug is inserted into the fill port lumen 103 that does not permit electrolyte components to escape through fill port 107 but that does permit hydrogen gas evolved through charge and discharge of capacitor 265 to escape from the interior thereof. By sealing fill port 107 with a plug or barrier having sufficient chemical resistance, but that is selective to hydrogen gas (such as some silicones, polyimides, polyphenylene oxides, cellulose acetates and triacetates and polysulfones), no electrolyte is lost. Several potting adhesives (such as epoxy or silicone rubber or cement) have the foregoing chemical resistance and hydrogen permeability properties and thus are suitable for use in the present invention.

[0196] Wire guides 140 and 141 may be formed from any of a wide variety of electrically insulating materials that are stable in the environment of an electrolytic capacitor. In one preferred embodiment, the material from which wire guides 140 and 141 is made is an injection molded polysulfone known as AMOCO UDEL supplied by Amoco Performance Products of Atlanta, Ga. In other embodiments, wire guides 140 and 141 may be formed from other chemically resistant polymers such as fluoroplastics (e.g., ETFE, PTFE, ECTFE, PCTFE, FEP, PFA or PVDF), fluoroelastomers, polyesters, polyamides, polyethylenes, polypropylenes, polyacetals, polyetherketones, polyarylketones, polyether sulfones, polyphenyl sulfones, polysulfones, polyarylsulfones, polyetherimides, polyimides, poly(amideimides), PVC, PVDC-PVC copolymers, CPVC, polyfurans, poly(phenylene sulfides), epoxy resins, silicone elastomers, nitrile rubbers, chloroprene polymers, chlorosulfonated rubbers, polysulfide rubbers, ethylene-polypropylene elastomers, butyl rubbers, polyacrylic rubbers, fiber-reinforced plastics, glass, ceramic and other suitable electrically insulating, chemically compatible materials.

[0197] FIGS. 29(a) through 29(d) show various views of the liquid electrolyte fill port 107 incorporating one form of a microporous plug 119 of the present invention within the fill tube lumen 103 and extending over the interior tube end 104. In FIGS. 29(a)–29(c), the tube exterior end 106 is not trimmed, whereas, the tube exterior end 106 is trimmed or ground down in FIG. 29(d). The plug 119 is formed by applying or injecting the plug material in high viscosity liquid, uncured, state into the fill port tube lumen 103 sufficiently that it tends to expand in the interior case chamber and also extends over the tube exterior end 106 and to the surface of ferrule 105. The plug 119 is formed when the applied plug material cures.

[0198] A similar plug 119′ that does not extend over the exterior surface of the tube exterior end 106 can be formed in the same manner and is shown in FIGS. 30(a) through 30(d).

[0199] Once the fill port tube lumen is sealed by one of the means and methods described above, the capacitor 265 is electrically tested. Applications in implantable defibrillators may require two capacitors 265 to be connected in series. In this embodiment, an insulator is provided by a two sided adhesive being disposed between the capacitors 265 so that they are joined along opposing faces with the insulator/adhesive strip disposed therebetween. The pair of capacitors 265 is then provided for assembly in ICD IPG 10 as shown and described above with respect to FIGS. 3(a) through 3(g).

[0200] Although only a few exemplary embodiments of a capacitor 265 in which the fill port tube and gas vent of the present invention is advantageously implemented have been described in detail above, those skilled in the art will appreciate readily that many modifications are possible in the exemplary embodiments without materially departing from the novel teachings and advantages of the invention. Accordingly, all such modifications are intended to be included within the scope of the present invention as defined in the following claims.

[0201] The preceding specific embodiments are illustrative of the practice of the invention. It is to be understood, therefore, that other expedients known to those skilled in the art or disclosed herein, and existing prior to the filing date of this application or coming into existence at a later time may be employed without departing from the invention or the scope of the appended claims. In the claims, means-plus-function clauses are intended to cover the structures described herein as performing the recited function and not only structural equivalents but also equivalent structures.

[0202] All patents and printed publications disclosed herein are hereby incorporated by reference herein into the specification hereof, each in its respective entirety.