Method and apparatus for electrolytic production of printed circuits

March 8, 1966 s. l. marosi 3,239,441

METHOD AND APPARATUS FOR ELECTROLYTIC

PRODUCTION OF PRINTED CIRCUITS

Filed March 19, 1964 ? Sheets-Sheet 1

<img file="US3239441A_D0001.tif" />

ATTORNEYS

March 8, 1966

Filed March 19, 1964

S. L. MAROSI

METHOD AND APPARATUS FOR ELECTROLYTIC PRODUCTION OF PRINTED CIRCUITS

3,239,441

Sheets-Sheet S

<img file="US3239441A_D0002.tif" />

<img file="US3239441A_D0003.tif" />

INVENTOR.

STEPHEN L. MAROSI

AIR UNDER PRESSURE

<img file="US3239441A_D0004.tif" />

BY

ATTORNEYS

March 8, 1966 s. l. marosi 3,239,441

METHOD AND APPARATUS FOR ELECTROLYTIC

PRODUCTION OF PRINTED CIRCUITS

Filed March 19, 1964 7 Sheets-Sheet 3

<img file="US3239441A_D0005.tif" />

INVENTOR

STEPHEN L MAROSI

<img file="US3239441A_D0006.tif" />

ATTORNEYS

March 8, 1966 s. l. marosi 3,239,441

METHOD AND APPARATUS FOR ELECTROLYTIC

PRODUCTION OF PRINTED CIRCUITS

Filed March 19, 1964 7 Sheets-Sheet 4

<img file="US3239441A_D0007.tif" />

<img file="US3239441A_D0008.tif" />

ATTORNEYS

March 8, 1966 s. i_. marosi 3,239,441

METHOD AND APPARATUS FOR ELECTROLYTIC

PRODUCTION OF PRINTED CIRCUITS

Filed March 19, 1964 7 Sheets-Sheet 5

<img file="US3239441A_D0009.tif" />

ATTORNEYS

March 8, 1966 s. ι_. marosi 3,239,441

METHOD AND APPARATUS FOR ELECTROLYTIC

PRODUCTION OF PRINTED CIRCUITS

Filed March 19, 1964 ⁷ Sheets-Sheet 6

<img file="US3239441A_D0010.tif" />

<img file="US3239441A_D0011.tif" />

INVENTOR

STEPHEN L MAROS!

<img file="US3239441A_D0012.tif" />

ATTORNEYS

March 8, 1966 s. l. marosi 3,239,441

METHOD AND APPARATUS FOR ELECTROLYTIC

PRODUCTION OF PRINTED CIRCUITS

Filed March 19, 1964 ⁷ Sheets-Sheet 7

<img file="US3239441A_D0013.tif" />

3,239,441

Patented Mar. 8, 1966

United States Patent Office ever, involved pressing the master and the work sheet together (see Prest Patents Nos. 2,305,990, 2,306,082, and 2,421,735) either with no intervening material or with a porous member positioned therebetween (see Schinske Patent No. 2,491,910). As a result, the electrolyte is trapped between the master and the work sheet and is contaminated by the dissolved metal of the work sheet and/or other products formed during the deplating process. Accordingly, such processes are not satisfactory for deplating through a substantial layer of metal. Even where a complete deplating is possible, uniform results are not obtained.

British Patent No. 840,544 to Telefunken does suggest the use of the deplating technique, without printing, to make printed circuits, but the master member of this suggested system has protruding pattern segments corresponding to the metal to be left on the work piece, such segments being in contact with the metal work piece during the deplating operation. It is disclosed that the electrolyte 20 is forced through a number of channel holes in the master member to flow through the spaces between the pattern segments, but it is not disclosed how electrical connections could be made to the master and work piece so that electrolytic deplating could occur. Moreover, it will be apparent that, at least with some patterns of electric circuits, free and unimpeded flow of electrolyte would not be possible with the master pattern in direct contact with the work piece.

It has been found that it is extremely important to satisfactory use of an electrolytic deplating technique for manufacture of electroconductive patterns, and particularly to such pattern incorporating a plurality of isolated areas on an insulating sheet, that a continuous supply of electrolyte be passed between the master and the work sheet. It has 35 been found that in the static, trapped electrolyte techniques, the concentration polarization of the electrolyte renders the process self-limiting. Moreover, where the desired chelating agents are added to the electrolyte to sequester the deplated metal, static conditions often re40 suit in formation of conductive collodial systems causing short circuits.

It is an object of this invention to provide an improved method and apparatus for production of electroconductive patterns such as printed circuits, which method and ap45 paratus employ a repetitively usable master, rather than an etchant or deplating-resistant pattern on each work sheet, so that the above recited disadvantages of use of a resist pattern on each work sheet may be avoided. As indicated, the use of such master has the further advantage 50 of elimination of the problems incident application of resist material to holes through the work sheet in the platedthrough-hole technique.

Another object of the invention is the avoidance of the use of concentrated etchants which corrode the materials 55 contacted thereby.

A further object is the achievement of a better bond between the metal pattern of a printed circuit and the insulating laminate.

These and other objects of the invention are obtained 60 by a process in which a master is formed by placing insulating material in the configuration of the desired circuit pattern on a metal member, the metal-clad insulator forming the work sheet is disposed adjacent but narrowly spaced from the master, a direct current source is con65 nected between the master and the metal surface of the work sheet with the master the cathode, and a stream of electrolyte is continuously forced through the passage between the master and the work sheet.

This process is preferably performed, particularly when 70 the desired circuit pattern is relatively long in the direction of flow of the electrolyte, by an apparatus in which a cylindrical master is rotatably mounted adjacent a suitMETHOD AND APPARATUS FOR ELECTROLYTIC PRODUCTION OF PRINTED CIRCUITS

Stephen L. Marosi, El Cajon, Calif., assignor to Marosi Precision Products Co., Inc., El Cajon, Calif., a corporation of California

Filed Mar. 19, 1964, Ser. No. 354,817

Claims. (CI. 204—143)

This application is a continuation-in-part of my applica- io tion Serial No. 198,992, filed May 31, 1962, and now abandoned and which in turn is a continuation-in-part of my application Serial No. 116,180, filed June 9,1961, and now abandoned.

This invention relates to the manufacture of electro- 15 conductive patterns, such as the so-called “printed circuits,” and more particularly, to a method and apparatus for forming such products by electrolytic deplating techniques.

The term “printed circuits” as used in industry, and herein, describes any portion of an electrical circuit, such as the wiring and/or the impedance and switching elements, finished in prefabricated form on an insulating base. Although capacitors, inductors, and even resistors have been produced in this form, most printed circuit production presently is of wiring and switches.

There are several techniques now in commercial use for production of printed circuits. The most popular of these currently is the print and etch method in which the desired pattern is first formed on the metal work sheet 30 (usually a metal-clad plastic laminate) in an etchant-resistant material, and the sheet is then exposed to a chemical etchant which removes all of the metal not protected by the resist.

While widely used, the print and etch technique has many undesirable features. First, the resist pattern must be formed on each work sheet before the sheet is etched, this usually being done through a photomechanical technique, or by application of the resist through a so-called “silk screen” bearing the pattern. Also, the patterns so formed are often imperfect in failing to completely cover all desired areas of the sheet to protect them against etching, with the result that the etchant attacks metal which should remain on the insulation, forming pin holes, creates areas of higher resistance than desired, and even forms discontinuities in the conductors themselves. Particular difficulty is encountered in assuring complete coverage of the interior surfaces of holes in the widely-practiced “plating-through-the-hole” techniques. A further disadvantage of the chemical etching method is that the etchant “undercuts” the desired pattern, a feature undesirable in many applications of printed circuitry but particularly so in switching or commutating circuits. An ever present problem also is found in employing and handling the necessarily strong etching solutions.

It has been suggested in various prior publications that electrolytic deplating, rather than chemical etching, be employed for the removal of unwanted metal in the manufacture of printed circuits. However, such prior suggestions generally have included retention of the “printing step through which a protective pattern is formed on each metal sheet prior to deplating. It is an object of this invention to employ electrolytic deplating in the production of printed circuits, but to eliminate this repetitive printing step through the use of a master bearing the desired circuit pattern and cooperable with the metal-clad insulating sheets to protect against electrolytic removal of the areas of those sheets corresponding to the pattern.

It has been previously suggested that such patterns as aircraft part templates be formed by electrolytic deplating through use of a metal master, rather than a printed pattern on each work sheet. Such prior suggestions, how3,239,441 able translating device supporting the work sheet, and the master is rotated in synchronism with linear movement of the translating device perpendicular to the axis of the master.

The master for use in the present invention is of conducting material, with the desired circuit pattern formed thereon in insulating material. While not essential to the invention, it is preferred that the insulating pattern be flush with the metal surface of the master. The exposed master surface should be spaced as closely as possible to the metal surface of the work sheet, over the deplating area, consistent with the requirement that electrolyte continuously flow through the passage therebetween, and the evident requirement that no physical contact occur between the metal surface of the master and the work sheet. Close spacing is desirable since the deplating action spreads out over the work sheet as the spacing between it and the master increased, thus causing a decrease in the .width of the metal surfaces remaining on the work sheet. It has been determined that a distance in the order of about 0.005 inch is the optimum spacing with commercially available-metal-clad insulating sheets.

It is possible to employ a planiform master in the deplating apparatus of the invention, particularly if the length of the master and the work sheet in the direction of movement of the electrolyte is quite short. However, most printed circuit patterns are of substantial length, and because of the variation in thickness of commercially available metal-clad laminates along their length, difficulty often is encountered in maintaining a uniform spacing of the order indicated between a planar master and a planar work sheet. Diminution in the size of the passage between these two elements impedes the necessary continuous flow of electrolyte and is undesirable. Further, deplating does not necessarily proceed at a uniform speed along the entire length of the work sheet, so that some surfaces might be removed all the way down to the insulating material while other surfaces still remained. Particularly when large isolated areas of metal must be removed from . the work sheet (as is true in most printed wiring production) it often is difficult to maintain the necessary electrical contact between some of these isolated areas and the direct current source, so that some unwanted metal areas would likely be left on the work sheet.

•For all these reasons it is preferably that the necessary narrow spacing between the master and the work sheet, and the deplating action, be over only a relatively narrow length of the work sheet. To obtain these effects, it is preferable for most printed circuit production that the master be of non-planar form and be moved in synchronism with a linear movement of the work sheet, so that the area of closest spacing between the master and the work sheet, and the deplating action, progresses along the length of the sheet, to provide what might be termed a moving line of electrolytic deplating. Preferably the master is of cylindrical form and is rotated in synchronism with linear movement of the work sheet along a path closely spaced from the master. With such a configuration, the deplating action may be confined to a relatively narrow area of the work sheet adjacent the line of closest approach between work sheet and master. It is preferred that a planiform master be employed only for relatively small printed circuits and a curvilinear master be employed at least for the larger more complicated configurations.

The prior art suggests another method for formation of printed circuits which does not require the use of a resist pattern on each work sheet. In this process, however, spark erosion is employed for removal of the unwanted metal from the work sheets, and the liquid disposed between the work piece and master is an insulating material which is periodically penetrated by sparks generated by buildup of the voltage between the master

4!

and the work sheet. (See Eisler Patent No. 2,785,280.) In contrast to the operation of this and other prior art suggestions, the method and apparatus of the present invention forms a well and accurately defined pattern 5 on the work sheet, without forming pin holes, without pitting of the retained metal, but while effecting removal of all of the undesired metal without any damage to the master or the insulation backing.

A further feature of advantage with respect to the ]₀ product of the present invention is the character of the· bond between the metal surface and the insulating material. That bond is actually increased during the deplating process, especially along the edges of the metal pattern, thus diminishing the possibility of the metal 15 pattern stripping away from the insulator during subsequent processing or use. The increased bond is believed due to the localized heating which occurs in deplating areas of the metal and especially in the areas immediately adjacent the edges of the retrained pattern. 20 It is postulated that such heating effects cause super polymerization of the resin in the laminated insulating base. While the increased bond is particularly significant with phenolic laminates such as XXXP where extreme difficulty has been encountered in obtaining a satisfactory bond be25 tween the laminate and metal, the phenomenon has been observed in working with all types and grades of laminates.

As indicated above, the thickness of the commercially available metal-clad laminates varies. Since the deplat30 ing time is directly proportional to the metal thickness, the apparatus and method of the invention desirably include adjustment of the speed of operation of the process in accordance with the thickness of the metal covering of the metal-clad sheet being treated. In accordance 35 with one embodiment of the invention the time required for deplating an initial portion of the sheet automatically is measured, and the speed of movement of the deplating area automatically adjusted. It: will be appreciated, however, that speed of movement can: be . maintained. ⁴⁰ constant and the voltage varied.

An important advantage of the process and apparatus of the invention resides in the .circumstance that excellent plated-through holes are .obtained without use of a resist.

When employing the acid etch process in conjunction with the plated-through hole technique, a laminate clad on both sides with metal is? drilled or punched at the desired points. The interior surfaces of the holes are made conductive by chemically depositing a? film or 50 thin foil of copper over the entire surface of the piece.

An additional· layer of metal then is electrodeposited over the “electroless” copper. . A resist pattern then is applied not only to the upper and lower surfaces of the clad laminate but also to the interior surfaces of the 55 holes. Unwanted metal then is removed by acid etching and the resist removed. Application and removal of resist from the hole surfaces is obviously essential where acid etching is employed and is a tedious and time consuming operation.

In the present invention, however, the electroplated, punched metal-clad work sheet is treated without resist. The master is merely provided with insulatory material in the regions coinciding with the holes and during passage of the work sheet under the drum—or between the drums 65 where the two sides of the work sheet simultaneously are deplated—no deplating occurs in the holes and an excellent product is produced.

The method and apparatus of the invention will now be more fully described in conjunction with a preferred 70 embodiment thereof, shown in the accompanying drawings. Such embodiment includes a reciprocable table which supports a metal-clad insulating sheet beneath a cylindrical master which is spaced above the work sheet, moves the sheet under the master until the unwanted metal 75 is removed therefrom over its entire length, and then

3.239,441 returns the sheet to the initial position for removal thereof and replacement by the next sheet. However, it will be apparent that the sheets could be disposed on a different type of translating member than a reciprocable table, such as a continuously-moving belt. In such 5 case, with appropriate synchronization of continuous rotation of the master with linear movement of the belt, there would be no necessity for the time delay occasioned in return of the translating member to its original position between deplating operations. Further, it will jq be evident that both sides of a doubly clad insulating sheet could be simultaneously deplated by use of a pair of masters closely spaced with respect to opposite sides of the sheet, with electrolyte forced through the passages between both of the metal layers and the adjacent mas- 15 ters. In such case the metal-clad sheets could be supported from their edges rather than along one side of a belt or table.

In the drawings,

FIG. 1 is a perspective view of an embodiment of the 20 apparatus of the invention;

FIG. 1A is a block diagram of an alternate system for removal of metal from electrolyte before recirculation thereof to the deplating apparatus;

FIG. 2 is a partial perspective view of a portion of 25 the master and the work sheet, showing the configuration of the elements of the pattern obtained with the apparatus of FIG. 1;

FIG. 3 is a partial sectional view taken along line 3—3 of FIG. 1, showing the open passage between the master 30 and the work sheet, and the side flaps for supply and reception of the electrolyte which is forced through the passage;

FIG. 4 is a vertical elevational view of a portion of the apparatus of FIG. 1 showing the flap members and 35 the drum, with the mechanism for elevating these members for placement of a work sheet thereunder;

FIG. 4A is a partial perspective view showing an alternative system for raising and lowering the drum and the flap members;

FIG. 5 is a plan view of a portion of the apparatus of FIG. 1 showing the table member and several cooperating switches employed for automatic operation of the apparatus;

FIG. 6 is a partial detail view, partly in section, show- 45 ing the mechanism for translating the table of FIG. 1;

FIG. 7 is a partial elevational view of the apparatus of FIG. 6, on a larger scale, showing various elements of the translating means;

FIGS. 8A and 8B are schematic views of cooperating 50 portions of the electrical control apparatus of the invention;

FIG. 9 is a schematic diagram of a control system for the deplating current source of the apparatus;

FIG. 9A is a diagrammatic elevational view of one of 55 the magnetic amplifier units of FIG. 9; and,

FIG. 9B is a diagrammatic plan view of all three magnetic amplifier units of FIG. 9.

Referring first to FIG. 1, the illustrative embodiment of the invention, generally speaking, includes a rotatable 60 drum 10 carrying a master sheet 11 bearing a pattern of the type to be duplicated, and a reciprocable table 12 supporting a work sheet 13 upon which the pattern is to be formed. The drum 10 and the reciprocable table 12 are of the same width. The drum 10 is mounted on 65 a rotatable shaft 14 which is supported in trunnions 15 at opposite sides of a cabinet 16. The table 12 is in turn supported on cylindrical guide rods 17 and 18 which themselves are mounted in a bed 19 positioned within the cabinet 16. The table is reciprocable along the guide 70 rods through rotation of a lead screw 20 driven by an electric motor 21 through a speed reducer 22 and a shiftable transmission 23.

The driving mechanism for the table is more fully shown in FIGS. 6 and 7 from which it will be seen 75 that the lead screw is supported by the bed 19 and engages a nut 25 fixed to one end member 26 of the table 12. For purposes that will be explained hereinafter, the transmission 23 is shiftable between a high speed condition and a low speed condition, through operation of a solenoid 27 which may appropriately operate when energized to shift a conventional gear train in the transmission from one condition to another. The solenoid and transmission, as well as the speed reducer 22, may appropriately be of well-known and conventional type so that they need not be more fully described herein.

The work sheet, which is appropriately a metal-clad insulating material, such as the well-known copper-clad laminate, is fixed in position on the table 12 by a clamp assembly generally shown at 28. The clamp assembly is controlled by a handle 29 which is operable when shifted from one position to another to contact the upper metal surface of the metal-clad sheet and hold it in position on the table. Electrical connection may be made to the metal surface of the work sheet 13 by connection to the clamp 28, as shown schematically at 30. Connection to the master 11 may in turn be made through connection with the shaft 14 through a lead schematically shown at 31.

While a manually-operable clamp control handle 28 may be employed as indicated, it is preferable that the clamp position be controlled by a cam automatically operative to operate the clamp when the apparatus begins operation.

Referring to FIG. 2, the master sheet 11 is therein shown as a metal member having insulating material segments 32 on its surface defining the configuration of the desired pattern. The pattern may be formed by use of a fairly thick sheet, or foil, of metal, such as copper, with portions of the outer surface of the metal member removed down to an appropriate depth less than the thickness of the member along its radius, and with such slots filled with an appropriate electrical insulating material such as an insulating resin. One convenient way in which the master pattern may be formed is by covering a sheet of copper material preferably of the order of 12-20 mils in thickness, with a photosensitive coating, exposing the coating through a negative of the desired pattern to a source of light, developing the resultant image to remove the unexposed photosensitive material over the areas defining the pattern, chemically etching the copper metal member down to an appropriate depth, and then filling the resultant slots in the outer surface of the master member with the insulating material. Of course the remaining photosensitive material may then be removed, and the master member is thereupon ready for placement upon the drum.

The drum 10 is desirably of conductive material, but may be constructed of insulating material and with current being uniformly supplied to the master 11 by suitable connecting means.

Referring to FIG. 3, as well as FIG. 1, the master 11 is disposed in spaced relationship with the metal-clad work piece, preferably at a distance in the order of about .005 inch. Such spacing provides a passage for electrolyte and for avoidance of damaging short circuits which might be occasioned by reasonably expected surface irregularities of the metal-clad laminate. Such spacing is conveniently furnished by mounting a pair of bands 33 and 34 on the drum 10 at opposite sides of the master, and near the opposite ends of the drum. The bands 33 and 34 may be of insulating material or may be of conductive material which is suitably insulated from the drum. These bands function as spacer rings and are spaced apart a distance substantially equal to the width of the metal-clad work sheet 13, so as to bear against the upper surface of the work sheet. Since the bands 33-34 position the drum 10 in parallel contact with the work sheet 13 at all times and because the drum axles 14 are vertically movable in the slots of the trunnions 15, a uniform gap is maintained at

3,239,441 all times between the master 11 and the work piece 13. With the described construction, excellent reproduction of the design in the master 11 may be obtained, despite variations in thickness of the work sheet along its length.

Liquid electrolyte is supplied to the passage between the drum and the work sheet through a pipe 35 mounted on a rear flap member 36. The flap member 36 has a passage 37 aligned with the pipe 35 and communicating with a longitudinal chamber 38 extending from one side to the other of the flap. That chamber in turn communicates with a slot 39 opening out from the chamber into the passage 40 between the drum and the work sheet.

At the other side of the drum 10 there is preferably mounted a second flap member 41 which has a passage 42 : extending longitudinally thereof and communicating with the passage 40 between the drum and the work sheet. The other end of passage 42 may communicate with the space above table 12 so that electrolyte which has been forced through the passage into passage 42 may drain down into the bed 19.

Liquid electrolyte is supplied under pressure to the inlet pipe 35 from a pump 43, through a solenoid-controlled valve 44 and an appropriate filter 45. With the valve in its normal position, however, the electrolyte is merely pumped from the supply tank 46 through the valve back to the tank. The pump may be of any appropriate wellknown type suitable to the character of electrolyte, which will be hereinafter described. The liquid electrolyte, after flow through the passage 40 between the drum and the work sheet is drained from the bed 19 through a drain pipe 47. If desired, the electrolyte may then be mixed in a mixing chamber 48 with appropriate constituents to adjust its characteristics to the desired level. The electrolyte may also be passed through a centrifugal separator 49 or a filtering means for separating out any particles of material that may have become entrained therein during the course of its passage through the machine.

If desired, an appropriate precipitating agent may be introduced into the mixing chamber 48 to precipitate the de-plated metal from the electrolyte. Such precipitant may be removed from the electrolyte in an appropriate separating apparatus as generally shown at 50.

Alternatively, or even in conjunction with the above described apparatus, there may be a plating tank 51 between the drain pipe 47 and the tank 46 for electrolytic removal of the metal ions dissolved in the electrolyte, before the electrolyte is recirculated by the pump 44 into the passage between the master and the work sheet. The plating tank 51/ like the remaining elements shown schematically between the drain pipe 47 and supply pipe 35, may be of well known type, appropriate to the particular electrolyte employed, as well as to the particular metal being electrolyticaly removed from the work sheet 13.

An advantage of the method of the present invention is that the metal removed from the work sheet is in a form such that it may be plated out in the tank 51 to decontaminate the electrolyte, and yet to permit recovery of the metal in form such as to allow some other use thereof. Unlike the copper compounds in the spent chemical etchant of the acid etch processes, the copper compounds in the spent electrolyte of the process of the invention need only be chemically reduced and plated out on suitable electrodes. Alternatively, the . copper may be removed and the electrolyte regenerated by passage through a bed of suitable ion exchange resin followed by elution of the bed with dilute nitric acid or other suitable eluent to recover the copper (or other metal) values.

A system of the latter type is shown in FIG. 1A, wherein the electrolyte, after passage through the “deplater” 62, is directed through a control valve 63 to one of two ion 7Q exchange columns 68 and 69. These columns may contain ion exchange resins, of the chelating type, in the form of small beads over which the copper-containing electrolyte passes. The copper ions are removed from the electrolyte by these beads, so that the electrolyte can be 75 supplied once again to the storage tank 46 through valves 70 and 71.

The valve 70 controls which of the ion exchange columns 68 or 69 is connected to the valve 71. The com5 bination of valves 63 and 70 may therefore be controlled to circulate electrolyte through one ion exchange column while the copper (or other metal) values are recovered from the resin beads in the other column by appropriate means (not shown).

jq It will be evident that appropriate additions may be made to the electrolyte supplied to valve 71 (for the purpose of adjusting the pH to the desired valve for example) before it is supplied to the storage tank 46. Such control, as well as the control of the valves such as 63, 70 and 71 15 may of course be automatic, under the direction of an appropriate detecting device.

In order that the conductivity of the electrolyte may remain constant the storage tank may be provided with an appropriate thermostatically-controlled heater. While 20 only a single pump 43 has been shown, it will be evident; that a separate pump may be desirable in the regeneration system. .

Referring again to FIG. 1, it will be apparent that it isr. necessary to rotate the drum 10, and hence the metal mas25 ter 11, in synchronism with translation of the work sheet 13. While synchronous rotation may be accomplished by a variety of mechanical means, it. is essential that no slippage occur between these members. The apparatus of the illustrated embodiment of the invention is supplied 30 with a preferred positive drive mechanism for maintaining synchronism between the two members.. As shown in FIG. 1, this rotating means for the drum 10 may appropriately consist of four straps 52-55. Each of these straps is fixed at one end to the drum and at the other end to the 35 table through an appropriate clamp mechanism shown at 56. It will be seen that each strap is wound partially about the drum, so that as the table moves, the.straps cause the drum to rotate in synchronization with such linear movement. The straps may be of insulating material but 40 for better wearing characteristics preferably are of metal suitably insulated from the master 11.

It will be evident that many other forms of drive mechanisms could be employed for rotating the drum in synchronism with translation of the metal-clad work sheet, 45 and the strap-driving mechanism is not at all critical to the invention.

As indicated in FIG. 1, the negative terminal of an appropriate source of direct current voltage is connected 'through, the shaft 14 to the master 11, _twhile the positive 50 terminal , of the source is connected through the clamp 28 to upper metal surface of the work sheet 13.' The amplitude of this voltage is not critical, as long as it is not sufficiently great as to cause “breakdown” of the electrolyte. It has been found that a voltage of the ⁵⁵ order of 3-5 volts is satisfactory in the. deplating operation. The specific conductivity of )the electrolyte of course limits the amount of electrical current that will flow between the master and the work sheet, so that as θθ the voltage is increased, a level of voltage is reached beyond which the de-plating current does not increase significantly. The voltage amplitude may appropriately be selected by experiment to be the best level for removal of metal from the work sheet 13 at a desirably fast rate,. ₆₅ and in an appropriate form for later removal of the metal ions from the electrolyte.

Various types of electrolyte systems may be used in this process. However, it has been found that acid electrolyte systems having a pH of about 3 or below and alkaline electrolyte systems having a pH of about 13 or above are preferred for utilization in. the invention. Those systems which are on the acid side suitably may contain any acid which is not converted to objectionable substances under the conditions encountered during electrolytic de-plating. Suitable acids, without limitations,

3,239,441 ⁹ .

include nitric, sulfuric, hydrochloric, sulfamic, formic and fluoboric, with nitric acid constituting the preferred embodiment.

In order to obtain the conductivity of the electrolyte solution at the requisite level, while maintaining the corrosiveness of the solution at a minimum for a given concentration of solute, neutral salts of the particular acid employed may be substituted in part for the acid. The corrosivity of the electrolyte desirably is maintained at a level where apparatus damage is minimized.

While many salts, such as the ammonium, sodium or potassium salts, of the selected acids may be employed, the ammonium and potassium salts are preferred. In order to prolong the effective life of an electrolyte and to aid in adjustments of the pH thereof, it is also preferred to employ a chelating agent in the electroyte system. The preferred sequestering agent is the sodium salt of diethylenetetramine penta-ascetic acid. Other known chelating agents such as sodium tartrate, sodium dihydroxyethyl glycerate, and sodium ethylenediamine tetraacetic acid suitably may be employed.

Incorporation of a chelating agent provides a smoother deplating, provides buffering action, increases the anode current efficiency, and smooths out erratic current fluctuations which may be encountered during electrolytic metal removal.

The electrolyte is appropriately selected so that it is of sufficiently high conductivity to carry the desired current load, which may be, for instance, of the order of 15,000 amperes per square foot at 4.5 volts, between the master and the metal-clad work sheet. It is of course preferred that the components of the solution do not decompose during electrolysis into degradation products which adversely affect the desired performance of the process. The electrolyte must also tolerate relatively high concentrations of copper or other deplated metal, and large volumes of work-sheet metal removed into the electrolyte desirably cause relatively small changes in pH. The electrolyte is also preferably selected so that compounds of the removed metal which are formed at the cathode master are highly soluble so that under the desired flow condtions the removed metal is not deposited on the master. It is further desirable that the electrolyte solution be such that copper or other deplated metal may be plated out using appropriate anodes formed of such material as platinum plated titanium, without electrolyte decomposition. The preferred electrolytes are those which are safe to handle, fairly inexpensive, non-fuming, and relatively non-toxic. Further, it is preferable that variations in solution composition resulting from electrode reaction be readily adjustable by simple addition of inexpensive chemical agents.

The following electrolyte compositions have been found suitable:

(1)

NaNO₃ _________________________________g—104

HNO₃ _________________________________cc__2.7

H₂O ___________________________________1—1 pH 3 (2)

NaNO₃ ________________________________g—104

HN0₃ _________________________________cc__2.7

Sodium potassium tartrate__________________g—12

H₂O ____________________________________1-1 pH 3 (3)

H₂SO₄ (93%) ___________________________cc__130

DTPA (40% NH₂0) (Hampex 80)________cc__166

Na₂SO₄ ________________________________g—40

H₂O ____________________________________1__1 pH 2 ¹51 g./l.

(4)

HNO₃, sufficient to adjust system to pH 2.5

NH₄NO₃ ________________________________g—no

DTPA _________________________________cc__ ⁵ H₂O ____________________________________1—1 = 21 g.

(5)

HN0₃ (69%) ___________________________CC__28.7 ^υ NH₄NO₃ ________________________________g—no

DTPA _________________________________cc—80

H₂O _____________________________________1-1 pH 2.1 ¹⁵(6)

HN0₃ (69%)___________________________cc__29

KN0₃ __________________________________g—139

DTPA _________________________________cc__100

H₂O ___________________________________1—1 pH 2.05 (7)

HN0₃ (69%)___________________________cc—28.7

NH₄NO₃ ________________________________g—HO

DTPA ________________________________cc__100

H₂O ____________________________________1—1 pH 2.3

30(

HBF₄ (48%) ___________________________cc__ 75

H₂O ____________________________________1—1 pH 0.5

35(9)

HBF₄ __________________________________cc__100

DTPA _________________________________cc__40

H₂O ___________________________________1—1

PH 0.7 (10)

HBF₄ (48%)____________________________cc__ 100

Sodium dihydroxylethyl glycerate__________cc 40 h₂o 1__ 1

The examples above given are merely intended to illustrate various electrolytes which may be used in the present process. Additional suitable examples are also disclosed in applications Serial Nos. 328,475 and 328,509, 50 on “Electrolytic Solution and Processes,” in the name of Ernest H. Wake.

The operation of the process of the invention in removal of metal from the work sheet 13 may be readily seen by examination of FIG. 2. The work sheet 13 is 55 shown therein to include a layer of insulating material 57 covered by a layer of metal 58. The insulating material may be of any suitable type depending upon the desired end use for the product, such as a glass epoxy resin laminate, or a paper-phenolic resin laminate (such as XXXP). The metal layer may also be of any appropriate type, such as a copper, aluminum, or nickel bonded to the insulator to form the metal-clad materials now available for production of printed circuits.

_6g In most printed circuit production the thickness of the insulating material is much greater than that of the metal layer, and may be of the order of to % of an inch, while the thickness of the metal layer usually is of the order of one to five mils. The thickness of the metal ₇q layer, however, has been exaggerated in FIG. 2 for more easy visualization of the operation of the apparatus in the method of the invention. It will be understood that the invention is not limited to any particular metal forming the layer 58 of the work sheet, since any metal which 75 may be electrolytically removed such as aluminum,

3,239,44111 nickel and the like, may be used in the process of the invention.

As the work sheet 13 moves in the direction indicated by the arrow in FIG. 2, current passes between the metal surface area of the master 11 through the electrolyte and the metal layer 58. The surface of the work sheet is then removed over the area not covered by the insulating pattern on the master, down to the upper surface of the insulating layer 57. The metal remaining on the insulating layer 57 after passage of the work sheet beneath the surface of the drum may be in a pattern of isolated seg-, ments, such as the commutating switch configuration shown in the drawings, yet electrical contact is nevertheless maintained between the metal being removed, until all of the metal is removed down to the insulating layer. The desirable effect is achieved by reason of the fact that the electrical connection to the anode work sheet is to the rear end of the metal layer 58, while electrolytic removal progresses along the work sheet from the forward end toward the rear end, during translation of the table beneath the drum.

While operation of the process of the invention results in removal of all undesired metal down to the insulating layer, there still may be some isolated particles of metal remaining embedded in the insulating layer, after deplating is completed. This may occur by reason of the characteristics of the metal-clad insulating material now available for production of printed circuits. Such material is usually made by electrolytic deposition of copper oh a metal drum, stripping it therefrom, and bonding the resultant strips, or foils, to an appropriate insulating sheet. The surfaces of the metal strips or foils which were outward of the drum during electrolytic deposition naturally have a rough texture in comparison with the smooth inward surfaces. In order to provide for a better bond between the insulating layer 57 and the metal layer 58, the rough or “toothy” surface is generally bonded against the insulating laminate under heat and pressure. Further, this same surface of the metal layer is usually oxidized before bonding to provide for a better bond. As indicated, the result of preparation of the metal clad insulating material in this manner may be that isolated particles of the metal are embedded in the surface of the insulating material. When the metal layer is removed down to the level of the insulating material, electrical contact will then be broken between the rear end of the metal surface of the work sheet and these isolated particles, so that the particles cannot be completely removed by the process hereinabove described. However, these particles are so minute that they may readily be removed by so called “bright dipping” in an appropriate etchant specific to the metal forming the layer 58 on the work sheet.

In contrast, if the smooth surface of the metal layer were placed against the insulating material during production of the metal clad insulator, these isolated particles of metal would not be present and all of the undesired metal could be removed from the work sheet by the process of the invention, without the additional bright dipping step. Therefore, the invention is not to be considered limited to this step, although it or an equivalent step may be necessary in production, depending upon the characteristics of the work sheet.

Some of the various steps identified above as being used in production of the work sheet, and particularly those used for the attainment of a satisfactory bond between the insulating material and the metal layer, may even be eliminated in the production of work sheets particularly designed for the process of the invention. This is for the reason that the process itself increases the bond between the edges of the retained metal and the insulating material, during the electrolytic removal step. As explained above, this apparently results by reason of localized heating at the edges of the retained metal areas during removal of the adjacent metal areas.

It will also be-noted from FIG. 2 that the retained’ metal areas have a rounded configuration,-which is somewhat distorted for the purpose of illustration in that figure. This rounded configuration is readily distinguishable from the undercut edges formed by the chemical etching process of manufacture of printed circuits, and results from the “fanning out” of the deplating current due to the spacing between the master and the work sheet.. As indicated, this configuration is generally desired for all types of printed circuits, since it reduces the danger, of accidental stripping of the retained metal from the insulating material, but it is particularly desirable in. switching circuits of the type specifically illustrated in the drawings.

It has been explained hereinabove that a . continuous flow of electrolyte through the passage between the master. and the work sheet is essential to practical operation of the invention. Hbwever, it is not essential, that the electrolyte be continuously regenerated during repetitive , operation of the invention and in fact, batch operation · without any regeneration during processing of the batch is contemplated by the invention. In such case regeneration might desirably be effected between batches or whenever the electrolyte becomes so contaminated that regeneration is necessary;

In the embodiment of the invention described the drum and the flap assembly may be moved away from the table when a new work sheet is to be placed in the machine in order to. prevent injury to the flaps 36 and 41. The mechanism for accomplishing this effect is shown iit some detail in FIG. 4. It will be seen therefrom that the drum shaft 14 has its lowermost surface bearing-agaiflst a roller 60 supported by a pivot arm assembly generally, shown at 61. The pivot arm assembly is movable fit opposite directions by a solenoid 64 and a tension spring. 65. The spring 65 returns the pivot arm assembly 61 tothe. position shown, thus lifting the drum and-keeping it in its upper position. In such position the drum is raised above the table, so that the metal clad work - sheet 13 · may be placed in position thereon. The two flap assemblies 36 and 41 are also preferably supported by the shaft 14, so that they are in their upiper positions when the pivot arm assembly 61 is in the position shown.

When the solenoid 64 is energized, it pulls the pivot arm assembly 61 to the right of FIG. 4 to permit the drum shaft 14 to lower into operative position. The drum is spring loaded to the lower position to counteract the electrolyte pressure tending to- force the drum up. When the solenoid 64 is de-energized, the spring 65 forces· the pivot arm assembly to the left, to raise the drum and flap' assemblies. The drum and flap lowering mechanism may also be appropriately provided with a suitable motion damping element, such as the fluid damper cylinder generally shown at 67, and whose shaft is fixed to the lower end of the pivot arm assembly 61.:

It will of course be understood that a pivot arm assembly, solenoid and damper cylinder are provided for each side of the drum, but since they are identical only one need be shown or described.

An alternative apparatus for raising and lowering the drum (and flaps) is shown in FIG. 4A, this apparatus being designed to insure proper spacing between drum and workpiece along the entire length of the drum, even if the thickness of the workpiece varies from side to side thereof.

In the apparatus of FIG. 4A, the drum control solenoid 64 is replaced by a solenoid-controlled four way air valve 64 A. whose parts are respectively connected to a source of air under pressure (not shown), to atmosphere, and to the air chambers of a pair of control chambers or cylinders 66A and 66B. These control cylinders also have hydraulic, chambers which are respectively connected to opposite ends of a first hydraulic motor 87A and a second hydraulic motor 87B, the respective ends of the two motors being connected in parallel. The air and

3,239,441 hydraulic chambers of the control cylinders 66A and 66B are preferably separated only by the interface between the hydraulic fluid and the air under pressure in each cylinder.

Each of the hydraulic motors has a piston rod 88A and 88B, respectively, which are connected to lever arms 89A and 89B, respectively. These lever arms are in turn attached at their opposite ends to a torque bar 89C which is supported for rotation in the reciprocable table 12 (the table not being shown in FIG. 4A). At points offset from the torque bar 89C, the lever arms 89A and 89B are respectively connected, as by an appropriate clevis connection, to lift rods 66A and 66B, respectively. As will be seen, the lift rods control the vertical position of the shaft 14 and therefore of the master drum 10.

In operation of the apparatus of FIG. 4A, when the drum is to be lowered, the valve 64A is operated (as by energization of an appropriate solenoid) to connect air under pressure to control cylinder 66A and to vent control cylinder 66B to atmosphere. Hydraulic fluid will then be forced into the upper ends of hydraulic motors 87A and 87B, to cause their piston rods to lower rods 66A and 66B until the bands 33 and 34 on the drum 10 (FIG. 1) contact the metal surface of the work sheet 13. If the work sheet is considerable thicker at one side than at the other there would be a tendency for one end of the drum to remain at a greater distance above the work sheet than the other, with the band 33 or 34 out of contact with the work sheet. In such case, however, that band would not prevent further downward movement of the drum, of at least its end thereof. The hydraulic fluid from control cylinder 66A can then continue to urge the corresponding piston rod 88A or 88B downwardly, until both bands 33 and 34 are in contact with the work sheet 13.

Of course the difference in thickness of the work sheet from side to side will not be more than about 0.015 inch, so that the difference in height of the opposite ends of the drum 10 can readily be provided for by somewhat loose connections between the torque bar 89C and the link bars or level arms 89A and 89B.

When the drum 10 is to be raised, the valve 64A is reversed in position, as by de-energization of the solenoid which controls it, to force piston rods 88A and 88B (and therefore lift rods 66A and 66B) upwardly.

Before describing the electrical components which cause automatic operation of the apparatus of the invention, the placement of certain switches will first be described by reference to FIG. 5. A trip finger 72 is mounted adjacent the table and extends into the path of movement of the work sheet 13. The trip finger is attached to a link 73 which in turn is connected to a second link 74 movable about a pivot pin 75. The link 74 is connected to a rod 76 which is movable vertically (as shown in FIG. 5) by a solenoid 77. The rod 76 also carries a cam member, or switch operating member, 78 which is cooperabie with a roller 79 fixed to a switch lever 80. The switch lever forms part of a trip finger microswitch 81 which is operated when the cam 78 moves upwardly to move the roller 79 to the right of FIG. 5. - The trip finger assembly is shown in the position with the solenoid 77 energized to move the trip finger into the path of the work sheet 13. When the work sheet strikes the trip finger 72, the various link members cause movement of the rod 76 upwardly to cause the cam 78 to operate the switch 81.

A tension spring 82 is connected between the link 74 and the cabinet and is operative to move the trip finger assembly so that the finger is out of the way of table 12 and work sheet 13 when the solenoid 77 is de-energized.

(It will be evident that a cam-operated microswitch could be employed in place of the trip finger assembly, to be actuated by movement of the work sheet 13 under the drum 10.)

The apparatus further includes a pair of switches which are operated by the table movement, these switches including a forward limit microswitch 83 which is operated by a switch actuator member 84 mounted on the forward end of the table, when the table moves to its full forward 5 position. A similar rearward limit microswitch 85 is mounted at the rearward end of the cabinet and is operated by a switch operating member or actuator 86 mounted on the table, when the table moves to its full rearward position.

j₍j Referring now to FIG. 8A, power for operation of the machine may be supplied from a conventional threephase alternating current source shown in block form at 90. A transformer 91 has its primary 92 connected across two of the leads of the three-phase source and its 15 secondary 93 connected through fuses 94 and 95 to a control relay 96. One terminal of the operating coil 97 of the control relay is connected to fuse 95, while the other terminal is connected to one of the contacts 98 of a normally-open push-button “ON” switch 99. The other 20 terminal 100 of the “ON” switch is connected through a normally-closed emergency stop or “OFF” switch 101 and a short circuit detector switch 102 to the fuse 94. The purpose of the short circuit detector switch will be described hereinafter.

With the “OFF” switch 101 in its normal position and with the “ON” switch depressed, the coil 97 of the relay 96 is connected directly across the secondary 93 of the power transformer, so the relay 96 is operated. With the relay operated, each one of its contact sets is engaged. One 30 of these contact sets, labelled 103, forms a part of a holding circuit for the relay, bypassing the “ON” switch 99. One of the contacts is connected directly to the operating coil of the relay, while the other contact is connected through the “OFF” switch 101 and the short circuit de35 tector switch 102 to the fuse 94. With the other side of the operating coil 97 connected to the fuse 95, the controy relay will remain energized as long as neither the “OFF” switch nor the short circuit detector switch is operated.

⁴⁰ A second set of contacts 104 of the control relay connects the fuse 94 to a bus lead 105, while the third set of contacts 106 connects the fuse 95 to a bus lead 107. The operating coil 108 of a pump motor control relay 109 (shown at the extreme right upper end of FIG. 8A) ₄5 is connected directly across bus leads 105 and 107, so that the relay is energized when the control relay 96 energizes. The pump motor control relay has three sets of contacts 110-112 which are operative when engaged to supply current from the three-phase source to the pump motor ₅₀ 43, appropriately through overload relays such as shown at 113 and 114. Therefore, when the “ON” switch is depressed, the pump motor begins operation. However, as noted above, electrolyte is merely pumped from the tank back into the tank until the solution valve solenoid 55 5C is energized.

The trip finger solenoid 77 (shown to the right and below control relay 96) has one terminal of its operating coil connected to the bus lead 107. The other terminal of the operating coil is connected through the normally00 open contacts 115 of a trip finger solenoid relay 116 to the bus lead 105. Therefore, when the relay 116 is energized, the trip finger solenoid 77 is energized to move the trip finger into the path of the work sheet 13 (FIG. 5). One terminal of the operating coil 117 of relay 116 is 05 connected to bus lead 105, while the other terminal is connected to one terminal 118 of a normally-open “START” switch 119. When the “START” switch is depressed, the contact 118 is connected to a contact 120 which in turn is connected to a terminal 121 of the rear70 ward limit microswitch 85. The microswitch 85 is of the double-pole, double-throw type, and, with the table at its forwardmost position, the switch is operated so that one movable contact 122 thereof engages the stationary contact 121. The movable contact 122 is in turn connected 75 to one stationary contact 123 of the trip finger micro

3,239,441 ¹⁵ switch 81. That switch is also of the double-pole, double-throw type and the switch is not operated at this time, so that one of its movable contacts 124 engages the stationary contact 123. Movable contact 124 is connected to bus lead 107, with the result that the operating Coil 117 of the trip finger solenoid relay 116 is connected across the power transformer secondary 93, and the relay energizes. The set of contacts 115 of the relay then close to complete the energizing circuit for the trip finger solenoid 77. That solenoid energizes to move the trip finger into the path of the work sheet 13. A holding circuit for the relay 116, by-passing the “START” switch, is completed through contacts 115A of the relay and a stationary contact 121A of microswitch 85 when the table moves forwardly to open the rearward limit microswitch 85.

One of the terminals 143 Of the variable speed reversible drive motor 21 (shown at the right of FIG. 8 A) which controls translating movement of the table 12, is connected to a normally-closed and a normally-open set of contacts 125 and 126, respectively, of a latching type relay 127. The latch coil 128 of the relay 127 has one Of its terminals connected through a set of normally-open contacts 129 of a forward control relay 130 to a stationary contact 131 of the “START” switch 119. When the “START” switch is depressed, the contact 131 is connected to another stationary contact 132 which is in turn connected to the stationary contact 123 of the trip finger microswitch 81. With this switch in the position shown, One terminal Of the operating coil of the latch relay is therefore connected to the bus lead 107, when the forward relay 130 is energized. That relay has an operating coil 133, one terminal of which is connected to bus lead 107. The other terminal is connected to a movable contact 134 of the double-pole, double-throw forward limit microswitch 83. In its normal position the movable contact 134 engages a stationary contact 135 which is connected to bus lead 105. As a result of these connections, the forward control relay 130 is energized when the control relay 96 is energized.

The other terminal of the latch coil of latch relay 127 is connected through the normally-closed contacts 136 of a protective relay 137 to the bus lead 105, so that when the protective relay 137 is de-energized, the latch relay 128 is energized.

The protective relay 137 is designed to be operated only when voltage is supplied to the variable speed reversible motor 21, to prevent opposite polarity voltages from being simultaneously, or nearly simultaneously, supplied to the motor. The operating coil 138 therefore has its terminals connected across the terminals 143 and 144 of the drive motor.

Supply of voltage to the motor is also controlled by a motor control relay 140 whose operating coil 141 has one of its terminals connected to bus lead 105. The other terminal of the operating coil is connected through a now-closed set of contacts 142 of the latch relay 127 to the contacts 129 of the forward control relay 130. Since these contacts are engaged at this time and connected to the bus lead 107, the motor control relay is energized.

The now-closed contacts 126 of the latch relay 127 are connected between the terminal 143 of the drive motor and the bus lead 107, while the terminal 144 of the motor is connected through another set of now-closed contacts 145 of the latch relay and a set of now-closed contacts 146 of the forward control relay 140 to the cathode of a controlled rectifier 147. The anode of the controlled rectifier is connected through a diode rectifier 148 to the bus lead 105, so that the drive motor 21 is supplied with a direct current voltage of amplitude determined by the control characteristics of the controlled rectifier at this time.

The gate, or control element, of the controlled rectifier is connected through the secondary 149 of a transformer 150 to the cathode of the controlled rectifier, so that the voltage supplied to the control element is determined by the voltage available across the primary 151 of the transformer 150. The primary 151 is supplied with an alternating current voltage of adjustable magnitude by a source 152. The amplitude of: the supply voltage is itself determined by the input control of the source connected to input terminals 153-155 thereof. The function of the control elements will be described more fully hereinafter but suffice it to say that at this .time the maximum voltage is available across the primary of transformer 150 ; so that the variable speed reversible drive motor 21 is provided with its maximum voltage to drive the table 12 at high speed away from its forward position.

Referring to FIG. 5, when work , sheet 13 moves into engagement with the trip finger 72, the trip finger microswitch 81 is moved to its second position. In that position, which corresponds to a placement of the forward end of the work sheet 13 immediately underneath the drum 10, the microswitch 81 interrupts the energizing circuit for the motor relay 140, thus interrupting, the supply of voltage to the drive motor. The table therefore stops.; (In order to provide for nearly instantaneous stops, the motor 21 preferably is provided with a magnetic brake.) At the same time, the energizing circuit for the trip finger solenoid 77 is interrupted, by reason of de-energization of the control relay 116, so that the solenoid is de-energized and the trip finger 72 is moved back to its retracted position. .

With the trip finger microswitch 81 in this second position, its movable contact 124 engages a stationary contact 156 which is connected to one terminal of the operating coil 157 of another control relay 158. The other terminal of the operating coil 157 is connected to the bus lead 105, so that relay 158 is energized at this time. One set of contacts 159 of the control relay 158 are then engaged to connect one terminal of each of the drum solenoids 64 to the bus lead 107. . The other terminal of each of the drum solenoids is connected to the bus lead 105, so that the solenoids'are energized at this time to. permit the drum and flap assembly to move down to its operative position.

The control relay 158 further includes a set of nowengaged contacts 160 which operates to form a holding circuit for the coil 157 including a lead 161. , The lead 161 is connected (FIG. 8B) through a set of normallyclosed contacts 162 of a time relay 163 to the bus lead 107.

Finally, (FIG. 8A) the control relay 158 has a set of now-closed contacts 164 which are connected between leads 165 and 166. These leads are respectively connected (FIG. 8B) to the cathode of a triode vacuum tube 167 and to the center tap of the secondary 168 of a power transformer 169. The primary 170 of the transformer 169 is connected between the bus leads 107 and 105, so that the transformer is energized at the time the main control relay 96 is operated.

The plate of the triode 167 is connected through the operating coil 171 of a first timer relay 172 to one outer terminal 173 of the secondary 168 of the power transformer. The control grid of the triode 167 is connected through a shunt RC network 174 to the. movable contact of a potentiometer 175. The outside terminals of the potentiometer are respectively connected between the center tap and the terminal 173 of the power transformer secondary 168.

As a result of these connections, when the relay 158 (FIG. 8A) is energized, voltage is applied between the plate and cathode of triode 167 and, after a time delay determined by the setting of potentiometer 175 and the time constant of the circuit including RC network 174, the tube passes plate current to energize timer relay 172.: One set of the now-closed contacts 176 of the timer relay 172 is connected between the bus lead 107 and one terminal of the operating coil 177 of . a control relay 178. The other terminal of this control relay operating coil is connected to the bus lead 105, so that the relay 178 operates at this point. The now-closed contacts 179 of the relay 178 are connected between bus lead 107 and a.

3,239,441 lead 180 which is connected (FIG. 8A) to one terminal of the transmission shift solenoid 27. The other terminal of the shift solenoid is connected to bus lead 105, so that the shift solenoid is energized at this time. The solenoid then operates to shift the transmission to low speed 5 for subsequent translation of the table.

The solution valve solenoid 50 is connected directly across the shift solenoid 27, so that such solenoid is also energized at this time, to shift the solution valve to position such as to supply electrolyte from the pump to the jq inlet pipe 35.

Referring back to FIG. 8B, the timer relay 172 also has a set of contacts 181 which are now closed and which are respectively connected between the center tap of power transformer secondary 168 and the cathode of a second 15 triode vacuum tube 182.

The plate of this tube is connected through the operating coil 183 of a control relay 184 to the terminal 173 of the power transformer secondary. The control grid of the triode 182 is connected through an RC shunt circuit 20 185 to the movable contact of a potentiometer 186. The outer terminals of the potentiometer are connected between terminal 173 and the center tap of power transformer secondary 168. Thus, when the timer relay 172 energizes, the control relay 184 in energized through 25 plate current of the triode '182, after a time delay determined by the setting of the potentiometer 186 and the time constant of the circuit including RC network 185. One set of the now-closed contacts 187 of the relay 184 is connected between the bus lead 107 and one terminal 30 of the operating coil 188 of a control relay 189. The other terminal of the operating coil of this relay is connected to the bus lead 105, so that the relay operates at this time. One set of now-closed contacts 190 of the relay 189 is connected between leads 191 and 192. Re- 35 ferring to FIG. 8A, lead 191 is connected to one terminal of the operating coil 193 of the deplating rectifier control relay 194. The other terminal of this operating coil is connected to bus lead 107. Lead 192 is connected to movable contact 134 of the forward limit microswitch 83, 40 which engages stationary contact 135 at this time. That contact is connected to bus lead 105, so that the deplating rectifier control relay is energized. That relay contains three sets of normally-open contacts 195-197 which are connected between the three-phase alternating current 45 source 90 and a deplating rectifier 198. This rectifier may be of any suitable design operable to supply the low constant voltage and high current desired for operation of the electrolytic deplating process. The two output terminals of the deplating rectifier are appropriately con- 50 nected to the master 11 and the work sheet 13 with a shunt resistor 199 in the circuit for metering purposes. (A similar control circuit may also be provided for a plating rectifier to supply the plating tank 51 with voltage at this time.) 55

Leads 200 and 201 are connected to opposite terminals of the resistor 199 and are connected (FIG. 8B) to the opposite terminals of the movable coil 202 of an ammeter 203. This meter is of the conventional moving coil type, but also has a fixed contact 204 and a movable con- 60 tact 205 positioned therein. The movable contact is connected to the pointer 206 of the meter, to move therewith, and the stationary contact 204 is positioned such that the movable contact engages it only when the current through the movable coil 202 reaches a predetermined minimum. 65 Stationary contact 204 of the meter 203 is connected to one terminal of the operating coil 207 of a control relay 208. The other terminal of this operating coil is connected through the normally-closed contacts 209 of a control relay 210 and the normally-open contacts 211 70 of a delay switch 212 to the bus lead 105. The movable contact 205 of the meter is connected through a holding coil 213 to the lead 201 and is also connected to the junction between a pair of resistors 214 and 215. The other terminal of resistor 214 is connected to bus lead 105, 75 while the other terminal of resistor 215 is connected through a diode rectifier 216 and through the contacts 187 of control relay 184 to bus lead 107. As a result of these connections, the relay 208 is energized when the movable contact 205 of the meter engages the stationary contact 204 but only after a time delay following initial energization of the delay switch 212. This switch has a heater 217 which has one of its terminals connected through the contacts 187 of relay 184 to the bus lead 107, and its other terminal connected to bus lead 105. Therefore, the heater 217 is supplied with operating current at the same time that deplating current begins to flow from the rectifier 198. The delay switch 212 is appropriately of the well-known “Amperite” type and is operative to delay closure of its contacts until after the deplating current has built up to a point such that the movable contact 205 of the meter moves out of engagement with the stationary contact 204.

The network of resistors 214 and 215 and the rectifier 216, together with a shunt capacitor 218, form a direct current power supply for the meter control circuit.

The speed of the reversible drive motor 21 is controlled, in future movement of the table, by the setting of a potentiometer 22®. The movable contact 221 of the potentiometer is connected through the now-closed contacts 222 of timer control relay 184 to a lead 223. This lead is connected (FIG. 8A) to the terminal 154 of the adjustable potential alternating current source 152. One of the outer terminals 224 of the potentiometer 220 is connected through lead 225 (FIG. 8A) to terminal 155 of the adjustable potential source 152. As shown in the drawing, that terminal is also connected to one of the output leads of the source. Terminal 224 of the potentiometer is also connected through normally-closed contacts 226 of the timer control relay 184 to lead 223 which, as explained above, also was connected to input terminals 154 of the adjustable potential source. However, at this time with the timer control relay 184 energized, this connection is interrupted.

The other terminal 227 of potentiometer 220 is connected through lead 228 (FIG. 8A) to terminal 153 of the adjustable potential source 152.

The setting of the potentiometer is controlled by a synchronous motor 230 (FIG. 8B) whose shaft is mechanically connected to the movable contact 221 of the potentiometer. That motor has a forward winding 231 connected in series with a reverse winding 232, with the series combination shunted by a capacitor 233. The point of junction between the forward and reverse windings is connected to bus lead 105, and this lead is also connected through a brake winding 234 to each of a pair of normally-disengaged and normally engaged contacts 235 and 236, respectively, of a motor control relay 237. One terminal of the operating coil 238 of the motor control relay is connected to bus lead 105, while the other terminal is connected through now-closed contacts 187 of timer control relay 184 to bus lead 107. Motor control relay 237 is therefore energized.

The outer terminal of the forward winding 231 of the synchronous motor is connected through the now-closed contacts 239 of the motor control relay and a forward limit switch 240 to a set of normally-closed contacts 241 of the control relay 210. Since the relay 210 is not energized at this time, the contacts 241 connect the forward winding of the motor and brake winding 234 across the bus leads 105 and 107, through the now-closed contacts 187 of the timer control relay 184. When the brake winding 234 is energized, the brake is released. The synchronous motor therefore begins rotation of its shaft to begin to move the movable contact of the potentiometer.

Deplating of the unprotected metal surface of the work sheet 13 proceeds until the unwanted metal has been removed down to the surface of the insulator. At that time, current drops off and the movable contact 205 of the meter

3,239,441

203 engages, the stationary contact 204 to complete an energizing circuit for relay 208. The holding coil 213 maintains the movable contact in engagement with the stationary contact until its operating circuit is interrupted. When the relay 208 energizes, its normally-open contacts 5 242 engage to make a connection between the bus lead 107 through timer control relay contacts 187 to one terminal of the operating coil 243 of the control relay 210. The other terminal of the operating coil of this relay is connected to bus lead 105, so that the relay then energizes and closes holding contacts 243A which bypass contacts 242. The operation of relay 210 causes disengagement of its contacts 209 to interrupt the energizing circuit for relay 208 and the holding coil 213 of the meter. Disengagement of normally-closed contacts 241 also interrupts the 45 energizing circuit for the forward winding 231 of the synchronous motor 230 and brake winding 234. Interruption of the brake winding circuit applies the brake. The motor therefore stops and the movable contact of the potentiometer remains in the position which it had reached at the 20 time decrease of deplating current to the predetermined minimum (“burn-down”) was achieved.

The adjustable potential alternating current source 152 may be a simple amplifier which supplies an output voltage determined by the setting of the potentiometer. This 25 voltage may appropriately be controlled by the level of a bias voltage supplied to the input of the amplifier, the portion of the total available bias voltage supplied to the amplifier being determined by the position of the movable contact of the potentiometer with respect to its terminal 30 224. Such a configuration is shown in FIG. 8A, the amplifier being identified by numeral 152A and the bias source by numeral 152B. The output voltage from the source 152 is then of amplitude determined by the time it has taken for the unwanted portions of the metal layer at 35 the forward end of the work sheet to be removed. Since the controlled rectifier operates similarly to a thyratron, this voltage controls the average voltage supplied to the input of the reversible speed drive motor 21. That input may be supplied to the armature of the direct current 40 motor. Therefore, when the drive motor is once again energized, it will rotate at a speed determined by the time taken for initial “burn-down,” to move the work sheet and the drum at a speed determined by the deplating speed most advantageous for complete removal of the unwanted 45 metal areas without excess time delay.

Referring again to FIG. 8B, the relay 210 also includes a set of now-closed contacts 244 which is connected between a lead 245 and (through timer control relay contacts 187) the bus lead 107. Lead 245 is connected 59 through the now-closed contacts 129 of the motor forward control relay 130 and the now-closed contacts 142 of the latch relay 127 to· the operating coil 141 of motor control relay 140. With the relay 140 energized, operating voltage of magnitude deterimend by the setting 55 of the speed control potentiometer 220 is supplied to the drive motor 21 and the table starts movement forward once more.

Motor control relay 140 also establishes a holding circuit for itself including a set of now-closed contacts 246 · eo thereof. These contacts are connected through now-closed contacts 247 of the trip finger microswitch 81, now-closed contacts 130A of forward contact relay 130, and nowclosed contacts 127A of the latching relay 127 to operating coil 141 of relay 140. This holding circuit bypasses con- eg tacts 244 of relay 210. It will also be evident that it bypasses contacts 131 and 132 of START switch 119, so that the START switch need not be held down during the first part of the cycle before operation of the trip finger microswitch 81. 7θ

The sensing and control apparatus so far described operates to sense the time required for deplating a portion of the work sheet, and then controls the speed of deplating of the remainder of the sheet in accordance with that time. In the preferred embodiment, using a constant voltage 75 deplating source, the speed of movement of the deplating area is the characteristic controlled. However, it is possible that other characteristics which. affect the speed of deplating be controlled in accordance with the sensed “burn-down” time. That time is affected by the thickness of the metal layer on the work sheet, the density of that metal, and its conductivity, all of which may vary along a sheet as well as from sheet to sheet. Further, “burn-down” time is affected by the conductivity of the electrolyte, which in turn varies with temperature, the pH, and the. amount of deplated metal in solution. Also, variation in line loss between the constant voltage source, and the master and the work sheet, changing the effective voltage, at the deplating area, affects “burn-down”, time. A feature of extreme importance of the present invention is the determination of “burn-down” time by measuring the time required to deplate one limited area of the work sheet and automatically controlling the deplating speed in accordance with that time rather than any attempt to control all of the variables which affect that time.

With this feature of the invention it is possible not only to compensate automatically for variation of thickness and characteristics of the metal covering of work sheets of the same.nominal thickness, but even to run work sheets of different nominal thickness through the machine without manual adjustment.

It is possible that direct contact be made between the master 11 and the work sheet. 13 during the deplating operation. The short circuit detector 102 shown in FIG. 8A is provided to open the control circuit for control relay 96 in such case, thereby to stop operation of the entire apparatus.

When deplating of all of the unwanted metal areas on the work sheet has ben completed, the forward end of the table moves the operating member 84 into engagement with the forward limit microswitch . 83, thus moving that switch from its normal position to a second position. As· a result, the energizing circuit for the forward motor control relay 130 is interrupted, thus interrupting the energizing circuit for the motor control relay 140. The motor therefore stops. At the same time, operation of switch 83 interrupts the energizing circuit for the control relay 194 for the deplating rectifier. With the deplating rectifier disconnected from the source 90, voltage is removed from the electrodes and no further deplating occurs.

The microswitch 83 includes a movable contact 248 which now engages a stationary contact 249.^; This contact is connected to a lead 250 which (FIG. 8B) is connected to the cathode of a triode vacuum tube 251. The movable contact 248 is connected to lead 166 which, as above described, is connected to the center tap of the power, transformer secondary 168. The plate of the triode 251 is connected through the operating coil 252 of the timer control relay 163 to a second outside terminal 253 of the power transformer secondary 168. The control grid of the triode is connected through an RC shunt network 254 to the movable contact of a potentiometer 255 whose outer terminals are connected between the center tap of the transformer, secondary and the terminal 253.

As a result of these connections, when the forward limit microswitch 83 operates, the . triode 251 after a definite time interval determined by the RC time constant and the potentiometer setting, passes plate current to energize the timer control relay 163. The normally-closed contacts 162 of this relay then disengage, thereby interrupting the holding circuit for the control relay 158 (FIG. 8A). With the contacts 159 of the control relay 158. now disengaged, the drum solenoids are de-energized to cause return of the drum and flap assembly to their upper positions. Also, the connection between the cathode of triode 167 and the center-tap of the power transformer, secondary 168 is interrupted by disengagement of relay contacts 164, thus interrupting plate current of the triode 167. Energizing current for timer control relay 172 is therefore interrupted, and this relay and relay 178 both de-energize. As a re

3,239,441 suit, the solution valve solenoid 50 is de-energized and the valve returns to normal position to direct the electrolyte solution from the pump to the tank. Further, the shift solenoid 27 is also de-energized to shift the transmission to high speed for subsequent table return. Also, the 5 energizing circuit for the triode 182 is interrupted and relay 184, as well as relays 189, 21® and 237 are thereby de-energized.

The movable contact of speed control potentiometer 220, being connected to the now-open contacts 222 of de- io energized relay 184, is now disconnected from the adjustable potential source 152. The entire bias voltage is then connected across the amplifier input and the maximum voltage is delivered by the source to the controlled rectifier 147. 15

The synchronous potentiometer control motor 230 must now be energized to move the movable contact of the potentiometer 220 back to its normal position adjacent terminal 224. This is accomplished through energization of the reverse winding 232 and brake winding 20 234. The outer terminal of the reverse winding is connected through a reverse limit switch 257 and a set of normally-closed contacts 258 of control relay 184 to the bus lead 107. The other terminal of the reverse winding is of course connected directly to the bus lead 105, so that 25 the motor is energized to return the movable contact of the potentiometer to its initial position. When that position is reached, the reverse limit switch 257 opens to deenergize the reverse winding and brake winding to stop movement of the motor. 30

Timer control relay 163 also includes a pair of nowclosed contacts 259 which are connected between the cathode of a triode vacuum tube 260 and the center-tap of power transformer secondary 168. The plate of the triode is connected through the energizing coil 261 of a 35 timer control relay 262 to the terminal 253 of the power transformer secondary. The control grid of the triode is connected through a shunt RC circuit 263 to the movable contact of a potentiometer 264 whose outer terminals are connected between the terminal 253 and the center 40 tap of the power transformer secondary. Thus, with relay 163 energized, triode 260 passes plate current to energize relay 262, after a time delay controlled by potentiometer setting and RC time constant. The relay 262 has a pair of now-closed contacts 266 which are con- 45 nected between bus lead 107 and a lead 267. That lead (FIG. 8A) is connected through a set of normally-open contacts 268 of a reverse control relay 269 to one terminal of the unlatch coil 270 of the latch relay 127. The reverse control relay 269 has an operating coil 271 which ⁵⁰has one of its terminals connected through the movable contact 272 and stationary contact 273 of rearward limit microswitch 85 to bus lead 105. The other terminal of the operating coil 271 is connected to bus lead 107, so that the reverse control relay 269 is energized at this time. ⁵⁵

Since the other terminal of unlatching coil 270 of latch relay 127 is connected through contacts 136 of the protective relay 137 to the bus lead 105, the unlatching coil is energized at this time. The contacts of that relay therefore return to the condition shown in the drawing. 60

The operating coil 141 of the motor control relay 140 of course has one of its terminals connected to bus lead 105. The other terminal is connected through a set of now-closed contacts 275 of the latch relay 127 to the contacts 268 of the reverse control relay 269. The motor 65 control relay is therefore energized to engage its contacts 146 and 246. Terminal 143 of the drive motor 21 is then connected through contacts 125 of the latching relay and contacts 146 of the motor control relay to the cathode of the controlled rectifier 147. Terminal 144 of the drive 70 motor 21 is connected through a now-closed set of contacts 276 of the latching relay 127 to the bus lead 107. It will be noted that the cathode of the control rectifier 147 was connected to terminal 144 in the forward movement of the drive motor but it is now connected to ter- 75 minal 143. Therefore, the motor rotates in reverse direction, at high speed, to return the table to its initial position. When the table reaches this position, its switch operating member 86 engages the rearward limit microswitch 85 to return that switch to its initial position, thus de-energizing the reverse relay 269 and interrupting the energizing circuit for the motor control relay 140 to stop motor drive. The board may then be unclamped and the machine is ready for a new cycle of operations.

It will be obvious that the initial movement of the table 12 away from its forward-most position causes reversal of the position of forward limit switch 83, thus de-energizing timer relays 163 and 262. However, the motor control relay 140 is not de-energized by the resultant disengagement of timer relay contacts 266. Rather, a holding circuit bypassing those contacts maintains the relay energized. That circuit is composed of now-closed contacts 246 of the motor control relay 140, now-closed contacts 281 of reverse control relay 269, and nowclosed contacts 282 of latching relay 127.

While the filament circuits for the triode tubes of FIG. 8B are not shown in the drawing, it will be understood that they may be conveniently energized through connection to an additional secondary winding of power transformer 169.

Since the operation of the machine has been described in detail hereinabove during the description of the electrical controls, the detailed operation will not be repeated. For convenience, however, the operation of the apparatus in general function will now be described.

When an electroconductive pattern such as a printed circuit is to be duplicated from a master 11, an appropriate metal-clad insulating sheet 13 is placed in position under the clamp 28 and the handle 29 is moved to clamp the sheet or board in position on the table. The “ON” switch 99 is then operated to furnish power to the electrical apparatus and, after a time delay provided foi warming up of the various filament circuits of the tubes employed, the “START” switch 119 is depressed. The motor 21 then operates at relatively high speed to drive the table forwardly until the forward end of the sheet 13 is directly beneath the drum. The leading edge of the work sheet 13 actuates the trip finger 72 which initiates the cycle wherein forward movement of the table 12 is arrested, the drum 10 is lowered to contact the work sheet 13, electrolyte flow is initiated and deplating current is applied. Deplating then starts automatically through connection of voltage to the master 11 and the sheet 13, while electrolyte is forced through the narrow passage 40 between these two members. When all of the unwanted metal has been removed from the forward end of the work sheet, the deplating current drops to a predetermined minimum. During this time interval between the start of deplating current and its reduction to this minimum, the synchronous motor 230 has been moving the movable contact 221 of the speed control potentiometer 220 to a position determined by that time interval.

When the ammeter 203 senses such decrease of the deplating current, movement of the potentiometer contact is stopped. The table then begins movement forwardly at a reduced speed determined by this “bumdown” time, and removal of the unwanted areas of metal from the sheet 13 proceeds as the table and the master drum move. When the unwanted metal has been removed from the rear end of the work sheet, forward movement of the table is stopped, the drum and flap assemblies are moved to their upper positions and the table is returned automatically at high speed to its initial position. When that position is reached, the controls are automatically reset for the next cycle of operations, the board is unclamped from the table and a new board may be placed in position.

With the spacing bands 33 and 34 in contact with the outer edges of the work sheet, there naturally remains a longitudinal strip of metal along each edge. These

3,239,441 strips of course may be removed during the usual subsequent cutting and blanking operations.

It is highly desirable that the deplating current density remain constant during operation of the apparatus, to insure complete removal of the areas of the metal layer 5 58 of the work sheet 13 which are not protected by the insulating segments 32 on the master sheet 11 (FIG. 1), at a constant speed of travel of the work sheet and drum relative to each other. The control system of FIG. 9 is designed to provide for this constant current density. jq

In the apparatus of FIG. 9, the clamp 28 is shown as connected to the three phase power supply 90 through three diode rectifiers 285, and three magnetic amplifier output windings 286. The anodes of the rectifiers 285 are each connected through a different one of the output 15 windings to a different phase output terminal of the power supply, but the cathodes of the rectifiers are connected together to the clamp 28. In similar fashion, the drum shaft 14 is connected to the power supply 90 through a set of rectifiers 287 and a set of magnetic amplifier output 20 windings 288.

The voltage supplied between the clamp 28 and the shaft 14 is controlled by the magnetic amplifier, which in addition has a feedback winding 289, a control winding 290, and a bias winding 291. 25

The control winding responds to the voltage between the anode and the cathode of the deplating apparatus and is designed to vary the impedance of the output windings 286 and 288 (in normal magnetic amplifier fashion) to maintain this voltage substantially constant. While this voltage might be sensed by connection directly between the clamp 28 and the shaft 14, the voltage therebetween is responsive not only to the IR drop across the gap between the drum and the work sheet, but also to the IR drop through the area of the metal layer 58 between the clamp and the gap. Since the latter IR drop depends on the resistance of the copper path, which may vary considerably, such connection would not furnish an accurate indication of the gap voltage. Therefore, the gap voltage is instead measured between the shaft 14 and the metal contact wheel or band 34, which is in direct contact with the portion of metal layer 58 directly beneath the master. (Of course, the wheel 34 is insulated from the drum, so as not to short the deplating system.)

The gap voltage between the shaft 14 and the wheel or band 34 is both sensed by a voltmeter 292 and split between a very low resistance meter shunt 295 and the entire resistance of a relatively higher resistance potentiometer 296. The entire resistance. of a much higher resistance potentiometer 297 is connected across the portion of potentiometer 296 determined by the setting of the movable contact of potentiometer 296, for isolation purposes. The movable contact of potentiometer 297 is in turn connected to one terminal of the control winding 290. The other terminal of the control winding 290 is connected through a bias voltage supply 298 to the junction between the meter shunt 295 and the potentiometer _:296.

The resistance of potentiometer 296 is sufficiently greater than that of the meter shunt 295 that practically all of the voltage across the gap between the shaft 14 and the metal contact wheel 34 is developed across the potentiometer 296. A selected portion of this voltage is connected in series with the bias supply voltage to the control winding. It will be noted that the gap voltage is in opposition to the bias supply voltage, so that an appropriate operating level for the magnetic amplifier may be maintained.

The magnetic amplifier also includes the feedback winding 289, which is provided with a voltage determined by the amount of current flowing from the rectifier system. This current is detected by connecting the feedback winding 289 across the meter shunt. Since the deplating current is very high, a satisfactory amplitude of voltage for the feedback winding may be developed even though the resistance of the shunt is very low.

²⁴ ’

The bias winding 291 is supplied with voltage from the bias supply 298 and a potentiometer 300 which is connected in series with the bias winding. The movable tap on the potentiometer of course selects the amount of the bias supply voltage which is provided to the bias winding and therefore sets the operating level of: the magnetic amplifiers.

FIGS. 9α and 9b indicate schematically or diagrammatically the connections of the various windings of the magnetic amplifier system. FIG. 9a shows the various windings of one phase, indicating the common control, system by which the feedback winding, the bias winding,, and the control winding, operate to vary the impedance of the output windings, by controlling the degree of saturation of the cores. FIG. 9b merely indicates how three different cores may be electrically connected together to provide for control of the impedance of the output windings for the three different phases. It will be evident that, the magnetic amplifier could be of any suitable well known type which is operative to vary the impedance of output windings in accordance with the current flow through control windings.

In operation of the system of FIG. 9, any change in the voltage across the gap between the shaft 14 and the surface of the metal layer 58 will be immediately sensed and will vary the current through the control winding 290 to adjust the impedance of the output windings of the magnetic amplifier. Such adjustment will result in the voltage supplied to the deplating system being compen20 sated in accordance with the change in gap voltage. Similarly, any change in the deplating current will immediately be sensed by the· feedback winding and the magnetic amplifier output windings will have their impedances correspondingly adjusted to compensate for such ³⁵ change.

It will be evident that many changes could be made in the particular machine disclosed, and in the details and operation of the process. Some of these changes have been referred to specifically hereinabove. Others will ⁰ occur to anyone of skill in the art to which the subject matter of his application pertains. In particular it will be evident that the speed control system specifically described is not essential to the invention, though some correlation between the speed of work sheet advance and the time required for deplating is very desirable, particularly ⁴⁵ when work sheets of different conductor thickness are to be used. Further, though the use of a cylindrical master and a moving limited deplating area have been emphasized herein, the importance of the concept of this invention to printed circuit production with a planar master and a planar work sheet should not be minimized, particularly for circuits of short length, such as those now called “micro-miniature” circuits. In . fact, circuits of greater length could be manufactured through use of a planar _5g master if the moving “line” of electrolytic removal were provided for, for example by use of a dielectric sheet between master and work sheet, with a slit extending through the dielectric sheet and with provision for relative movement between the dielectric sheet and the work sheet, as by movement of the work sheet. In such case, the electrolyte could be pumped through the slit, which of course would move relative to the work sheet.

It will be understood that this invention is not to be considered limited to the preferred embodiment herein described, but rather only by the scope of the appended claims.