US3413683A - Annular bi-component spinerette assembly - Google Patents

Description

Dec. 3, 1968 R. 1.. YELVERTON, JR 3,413,683

ANNULAR BI-COMFONENT SPINNERETTE ASSEMBLY Filed Sept. 24, 1965 6 Sheets-Sheet l INVENTOR. ROY L. YELVERTO/V JR.

ATTORNEY Dec. 3, 1968 R. L. YELVERTON, JR 3,413,683

ANNULAR BI-COMPONENT SPINNERETTE ASSEMBLY 6 Sheets-Sheet 2 Filed Sept. 24, 1965 I L I INVENTOR. ROY L. YELVERTO/V JR.

ATTORNEY Dec. 3, 1968 R. L. YELVERTON, JR 3,413,633

ANNULAR ELI-COMPONENT SPINNERETTE ASSEMBLY Filed Sept. 24, 1965 6 Sheets-Sheet 3 INVENTOR. R0) L. YE L V5 R TON JR.

A TTORNEY R. L. YELVERTON, JR 3,413,683

ANNULAR BI-COMPONIZNT SPINNERETTE ASSEMBLY Filed Sept. 24, 1965 6 Sheets-Sheet 4 INVENTOR. ROY L. YELVERTON JR.

ATTORNEY 1968 R. YELVERTON, JR 3,413,683

ANNULAR BI-COMPONLNT SPINNERETTB ASSEMBLY Filed Sept. 24, 1965 6 Sheets-Sheet 5 INVENTOR. R0) L. YELVERTO/V JR.

ATTORNEY Dec. 3, 1968 R. L. YELVERTON, JR 3,413,633

ANNULAR BI'COMPONENT SPINNERETTE ASSEMBLY Filed Sept. 24, 1965 6 Sheets-Sheet 6 INVENTOR. ROY L. YE'LVERTON JR.

A 7' TOEWE Y United States Patent 3,413,683 ANNULAR BI-COMPQNENT SllNERETIE ASSEMBLY Roy Lee Yelverton, .Ir., Gulf Breeze, Fla, assignor to American Cyauamid Company, Stamford, Conn, a

corporation of Maine Filed Sept. 24, 1965', Ser. No. 490,651 2 Claims. (Cl. 18-8) ABSTRACT OF THE DISCL'BSURE For spinning economically large numbers of bicomponent fibers from each spinning position, an annular bicomponent spinnerette is provided having a novel spin dope distribution assembly associated therewith.

This invention relates to an improved circular extrusion head for multi-component extruded filaments and, more particularly, for bi-component extruded filaments.

Synthetic fibers, which are achieving rapidly increasing use, are made by extruding polymer into a coagulating or cooling medium. There are, in general, three processes. One is melt extrusion, where the materials which form the final filaments are melted and extruded through spinnerettes into a medium which is colder and which may be colder air, a coagulating bath, and the like. The second method is the so-called dry spinning method, in which the materials forming the fiber are dissolved in a solvent of suitable volatility and are then extruded through spinnerettes into a hot atmosphere which may be air, the solvent rapidly evaporating, and the filament solidifying after it leaves the spinnerette nozzles. The third method is the so-called wet spinning method, in which the polymer is dissolved in a solvent and extruded through a spinnerette into a coagulating bath, which causes the material to coagulate. The wet gel is washed free of solvent, stretched and dried to form the final filament. The coagulation may be by dilution of the solvent, chemical action on the materials extruded, or temperature. Very frequently the action of the coagulating bath may be a combination of dilution and temperature change. In the specific example of filament formation which will be described below and which deals with bi-component acrylic fibers, the COag' lating bath is much colder than the temperature of the extruded dissolved material and also results in dilution, because in this case the spinning solution is a hot aqueous salt solution of polymer and the coagulating bath is a cold salt solution with excess Water. Coagulation takes place, therefore, both by temperature change and by dilution of the solvent. Since the present invention relates to an apparatus and broadly also to an extruding process for wet spinning, the particular mechanism of coagulation in the bath into which the filaments are extruded is more or less immaterial; and the invention, therefore, is not limited to the wet spinning of any particular solutions into any particular coagulating bath. The specific example which will be described is only a typical use of the invention, although a very valuable one, and is intended to be illustrative only of the operation of the invention.

Extruded fibers are produced in two general forms: continuous filament, in which the extruded filament is formed in continuous lengths, though sometimes a thread may have a number of filaments; and the so-called spun threads, which are produced from relatively shorter pieces of filaments, which are frequently referred to as tow or staple. It is with the latter type of synthetic fiber threads that the present invention is most commonly used, but the extrusion of multi-cornponent, continuous filament yarn is not excluded. In the past, when staple or tow was produced the fibers were machanically crimped,

'ice

somewhat in the form of the crimps on a hobby pin, in order to produce a sutiicient amount of cohesion to permit textile processing. The spun thread or yarn is then made into fabrics and usually in the finishing of the fabrics they are subjected to heat, such as hot water, steam and the like, which takes out the remaining mechanical crim leaving the spun yarn with the individual fibers substantially straightened.

The yarns spun as described above have achieved widespread commercial use, but there is a demand for a bulkier yarn, for two reasons among others. First, the bulkier yarn has a softer feel or hand and, secondly, in many cases the bulkier yarn has greater covering power per unit weight, with of course a saving in cost. In order to produce a bullrier yarn it has been proposed to extrude multi-component and, preferably, bi-component fibers, the two components having slightly different rates of shrinkage. These multi-cornponent fibers, which in the balance of the specification will be referred to as bi-component fibers in accordance with the preferred form, tend to curl or crimp on relaxation from stretching and when such fibers are made up into yarns the curl results in a bulkier and softer yarn. For textile processing purposes, after the fibers have been extruded from a spinnerette into a suitable coagulating bath they are usually stretched. If they are allowed to relax and curl, the crimp frequency and amplitude usually are too extensive for satisfactory textile processing, and therefore it is common to remove the curl temporarily by stretching under suitable conditions and then mechanically crimping for spinning purposes. The spun yarn either before or after being formed into a fabric is then subjected to heat and cooling, for example by hot water, and on cooling the individual fibers curl and the bulky yarn results. The present invention deals only with the extrusion of the oi-component threads, and the further treatment, spinning and the like, is not significantly changed by the present invention. The extrusion of bi-component filaments, however, presented quite serious mechanical and operating problems, for the spinnerette rifices were extremely tiny, for example a small fraction of a millimeter for 3 denier filaments. To proportion accurately two streams of the different filament components was a formidable problem. This can be well appreciated when one considers that a commercial spinnerette for wet spinning tow may have from 5,000 to 20,000 or more orifices.

A successful approach to the problem above presented is described and claimed in the patent to Fujita Shimoda and Zoda, No. 3,182,106, May 4, 1965. In this patent each orifice in the spinnerette plate of the extrusion head was flared on its inner side to produce openings much larger than the final spinnerette orifice. This also had the additional advantage of spacing the orifices sufficiently so that the extruded filaments did not tend to stick to each other and the coagulating bath could reach all of the circumference of each filament. The back of the spinnerette plate showed the enlarged or flared orifices tangent to each other in rows. A guide plate back of the spinnerette plate had channels communicating with the rows of orifices in the spinnerette plate, and behind this were mounted septa along each row, which, in effect, with the plate defined a Y-sh-aped channel, the sharp edges of the septa of course being aligned with the center of the orifice openings in the back of the spinnerette plate. Supply manifolds were provided so that one solution of material to be extruded flowed down one leg of the Y and the other down the other and joined and flowed in laminar fiow down through the stem of the Y and into the orifice openings in the spinnerette plate. This laminar fiow produced no substantial mixing of the normally quite viscous spinning solutions, and so, a filament was extruded which had in cross section a semi-circle of one component and the other semi-circle the other.

The present invention is not concerned with the exact shape and size of the spinnerette orifices. The semi-circular cross sections are, of course, produced by round orifices. The size of the orifices is determined by a number of factors, including the cross sectional size of the fiber produced. For convenience the two different spinning solutions were referred to as A and B, and this same convenient designation will be followed in the present specification. The patent above referred to also describes a refinement in the form of an extremely fine mesh screen, for example less than 200 mesh, immediately back of the orifice openings in the spinnerette plate. This produced a very slight amount of local turbulence so that at the interface of the two semi-cylindrical filament components there was very slight mixing, which prevented the separation at the interface when the filament was coagulated.

The spinning process described above was successfully operative, but the mechanical construction of commercial extrusion heads with many thousands of orifices presented a very severe problem. A major breakthrough to this problem was achieved by Douglas and Tonnies in their application, Ser. No. 249,203, filed Jan. 3, 1963, and assigned to the assignee of the present application. Here a large number of thin plates of different shapes in series of 3s or 4s were stacked to form the septa and the distribution of manifolds of solutions A and B to the two sides respectively of each septum. If one looks at the extrusion head lying on its edge, this meant that the thin plates were stacked vertically and they were generally referred to as a vertical stack head. For the first time, an economical practical commercial extrusion head for thousands of orifices was produced. However, important as the breakthrough was, problems were presented. The plates had to be aligned extremely accurately because, of course, their edges, forming the septa, had to be precisely aligned with the spinnerette orifices. Also, the very thin plates were capable of bending and a serious problem was presented when the head was disassembled for cleaning, which is a periodic necessity in a practical commercial machine.

The next step for commercial extrusion heads is described and claimed in the co-pending application of Sulich, Ser. No. 286,773, filed June 1 0, 1963, now US. Patent No. 3,245,113, issued Apr. 12, 1966, and also assigned to the assignee of the present invention. Here the channels, including spin solution manifolds, dividers, septa and the like, were arranged in horizontal plates parallel to the face of the spinnerette plate, and this construction is known as the horizontal stack head.

Problems of alignment of the various plates when the head is disassembled for cleaning were completely solved, but a problem remained of accidental slight bending or warping of the very thin septa when the head was disassembled for cleaning. This problem was at least as acute as in the vertical stack head, although the initial alignment of the plates presented no problem. The horizontal stack head, except for the great care needed to prevent bending of the septa in cleaning, overcame the disadvantages of the vertical stack; but now another problem arose.

With an extrusion head having many thousands of spinnerette orifices, the coagulating bath has to pass from the outside of the rope of extruded filaments into its center. In such passage the temperature of the coagulating bath changes, because in many cases the spinning solutions are quite hot, for example a temperature difference of 60 C. and more from the approximately C. coagulating bath in the case of acrylic fibers extruded into aqueous thiocyanate baths, as described in the specific example below. Also, in this case, and the same 1s true for many other wet spinning processes, the chemical nature of the bath changed, because in coagulating the thiocyanate content of the coagulating bath increased, and in the case of other coagulating baths different chemical changes may result. As a result, the filaments near the center of the rope did not coagulate at the same rate as those nearer the outside, and this problem is referred to in the art as white core.

The problem was not so severe that extrusion heads could not be used, but it required a wider tolerance for practical acceptance and so was a distinct practical disadvantage. It is more particularly with a solution to this problem and the elimination or great reduction of bending of septa during cleaning that the present invention relates. The present application may, therefore, be considered as an improvement on the Sulich application; but it should be noted that the problems solved are really acute only with extrusion heads for commercial machines which have thousands of orifices. The problems do not arise, at least in severe form, in laboratory machines which may have extrusion heads of only a hundred or a few hundred orifices. The present invention is, therefore, an improved head for practical commercial machines and not a new principle which is needed in all sizes of extrusion heads, even for laboratory machines.

In the horizontal stacked head of the Sulich application, it was necessary to provide certain'stitfening or supporting ribs to the plate in which the septa were formed. These resulted in some channels, as of course there were no orifices in the spinnerette plate opposite the ribs. To a very small degree, these channels helped with the white core rope problem, but the help was too small to be of any practical significance either in the rectangular form of extrusion head or in the circular form, which are both shown in the Sulich application. As a matter of fact, there is somewhat greater help in the white core problem in the rectangular head than in the circular head. However, even there the improvement is so slight that for practical purposes the difference is insignificant, so that in practical machines no distinction is made whatsoever between the two shapes of heads as far as white core problems are concerned.

The present invention utilizes horizontal stacking, but the plates are neither rectangular or circular, they are annular. The extrusion head, therefore, has a large central conduit through which fresh cold coagulating bath is continually circulating. The outer shape of the head would not theoretically need to be circular, it could be square; but as this would only increase manufacturing costs and slightly decrease efficiency, practical extrusion heads according to the present invention are made up of elements which are pure annuli with concentric inner and outer circular peripheries. However, in the specification and claims, the term annular will be used in a slightly broader sense as shapes in which the central opening and the outer periphery form concentric curves which, however, do not have to be pure circles, although for practical purposes nothing is gained by departing from the ideal perfect annular shape.

In addition to the presence of a central, relatively large conduit through which coagulating bath can flow, the present invention does have certain general, though not sharply critical, dimensional relationships. Thus an annular head with a minute central opening is no better than a head with no opening at all, and it is of importance that the distance from the outer periphery of the annulus to the inner be sufiiciently small so that the number of rows of orifices is not excessive. For practical operating machines the cross sectional area of the central channel should not be significantly less than 5% of the area of the annular part of the extrusion head in which the orifices are located. Excellent results are obtained with bi-component acrylic polymers when the area is from about 10% to about 12% of the annular area of the head containing the orifices. Theoretically there is no upper limit on the percentage, but as a practical matter, it will rarely be desirable to have an area more than about 15% of the annular area, because if it is too big the overall size of the head for a given number of extrusion orifices becomes uneconomically large, although of course perfectly operative.

The effectiveness of the introduction of coagulating bath through the center opening is so great that, as will be described at the end of the specification in specific tests, it is possible to introduce all of the bath through the central openings, with none being introduced around the extrusion head. It is, of course, preferable to have bath on both sides of the tubular rope of filaments which results, but it is quite surprising that, as an extreme test, the introduction through the central opening alone permits operation at speeds substantially as great as when the bath is introduced also around the extrusion head.

Another important advantage of the present invention is a greatly increased pull-away of the filaments from the extrusion orifices, which permits satisfactory operation at considerably higher speeds. This is an advantage which can be enjoyed in two difierent ways. Thus with the same quality of extruded filaments, higher speed can be used, or with the same speed, an enormously greater factor of safety can be enjoyed. In practical operating machines a compromise is usually chosen, increasing the speed markedly while at the same time not going to the limit, so as to increase relia ility of operation. The best particular compromise in this respect will be chosen in each case in conjunction with the operating requirements, which include not only the nature of the material extruded but also its cross-sectional size. Different sizes of filaments also have an effect on the range of central openings which are permissible. As a general thing, coarser filaments, such as for example carpet threads, present a more serious problem of white core than do finer threads, and for this reason, in the specii c tests described at the end of the specification, the use of relatively coarse threads permits quite drastic testing of the advantages of the present invention.

The movement of the filaments through the coagulating bath serves to pull the bath along with it. In other words, they exert a pumping action. it is, therefore, ordinarily not necessary to use pumps which circulate bath at high speeds. Generally it is sufilcient to pump in bath at a rate such that it will not be degraded by warming up, change of chemical concentration, and the like; and the flow through a conventional bath chamber is quite largely effected by the pumping action of the filaments. In this connection, it should also be noted that there is a much stronger pum ing action on the bath entering through the central opening than that around the outer surface of the tubular rope of filaments. This pumping effect increases markedly the reliability of operation and is one of the factors permitting operation at higher speeds.

It mi ht at first be thought that the present invention would drastically reduce the number of orifices in an extrusion head of given size. However, since the number of orifices is very much greater in the outer rows than in the center, this does not reduce their number significantly or, if it is essential that the same number of orifices be preseat, a head of only VCI'j moderately increased outer circumference is needed. That the great advantages of the present invention are obtainable without significant reduction in number of orifices or with only very slight increase in size constitutes an important advantage of the present invention.

The generally annular arrangement of the orifices is not critical. The rows may be perfect circles, each having a septum which is continuous around the row. Other designs are possible, such as radial septa and holes. In such cases the holes need not be exactly in perfectly concentric circles. It is sufiicient that the rows of holes form an annular band.

The preferred pure annular shape of the plates has an additional advantage. The septa now are cylinders fastened to a rigid fiat plate. This is both easy to produce mechanically and it reduces the possibility of slight bending or warping of the septa when an extrusion head is disassembled for cleaning to a very significant extent. When supported at one end, a cylindrical surface represents a maximum of stiffness and so, the problem of septa bending which, although reduced over the vertical stacked extrusion head by the horizontal stacked head, still existed is solved in the present invention, and it is solved by reason of the shape and design of the invention and not by the addition of further elements. The stiffness of the cylindrical septa can be well visualized when one considers that these septa are very short, in other words they form cylinders which are extremely short in comparison to their circumference. As one end of each cylinder is rigidly mounted or even integral with a still flat plate, a very high degree of stiffness is obtained so that the possibility of septa bending on disassembling and cleaning is, for all practical purposes, completely eliminated.

The invention will be described in greater detail in conjunction with the drawings, in which:

FIG. 1 is an exploded view of the plates forming the stack of one modification;

FIG. 2 is a cross section through a stack taken on the line 22 of FIG. 6;

FIG. 3 is a section along the line 33 of FIG. 5;

FIGS. 4A and 4B are cross sections along the lines 4A 4A and 4B4B respectively of FIG. 5, both at right angles to the section of EEG. 3;

FIG. 5 is a plan view, partly broken away, of two plates forming the distribution element for different solutions;

FIG. 6 is an elevation of the front view of the primary distribution plate;

PEG. 7 is an exploded view of a stack similar to FIG. 1 but showing a modified final distribution plate, and

FIG. 8 is a rear view, partly brokn away, of a portion of the distribution plate of FIG. 7.

Turning to FIG. 1, it shows the principal elements of a stack in exploded form with the exception of a screen just back of the spinnerette orifices which does not show in this figure but can be seen in FIG. 2, as will be de scribed below. The stack is made up of six main elements, a primary solution distribution plate 1 provided with external threads 18, an intermediate plate 2 with two rows of staggered holes 22 and 23, a final distribution element in two pieces 3 shown partly broken away in FIG. 1 and in more detail in FIGS. 3, 4A and 4B, an orifice plate 4 with orifices 20, an outer clamp ring 5 with a tapered skirt 1?. with internal threads 19 and an internal clamping element 6 with a clamp flange d, separate body It and external threads 9. It will be seen that as all of the elements of FIG. 1 are annular, when the stack is clamped together, as is shown in FIG. 2, there is a central passageway, shown generally at 7, which flares at its rear end where the tapered portion of the body 16 of the element 6 screws into the internal theads 11 of plate 1. This will be seen best in FIG. 2.

The stack will be described in connection with its operation with two supplies of polymer solutions of different shrinkage characteristics. These solutions will be labeled A and B in all of the figures in addition to the numerical labeling of the actual elements of each plate. The two solutions A and B enter through pairs of pipes 13 and 15 in plate 1. This is best seen in FIG. 6, although FIG. 2 shows some of the elements in cross section. It will be seen that plate 1, looking at it from the front, is provided with two concentric grooves 14 and 16 which extend around a full circumference of the plate 1. It will be seen from FIG. 6 that solution A entering through the pipes 13 flows into the outer groove 14 whereas solution B, entering through pipes 15, flows out into groove 16.

Plate 2 mounts on the face of plate 1 with the outer ring of holes 22 registering with the groove 14 and the inner row of holes 23 registering with the groove 16. Plate 2 transforms the grooves 14 and 16 into conduits. Solutions A and B pass from the conduits through plate 2, through the holes 22 and 23 respecetively.

FIG. shows a front view, broken away to illustrate its three levels. The rear-most level, which is shown at the right, is provided with oblong openings 24 and 25. These openings alternate and the openings 24 extend radially beyond the ends of the openings 25. The alignment is such that the openings 24 communicate with the staggered holes 22 of plate 2 and at their opposite end extend sufficiently to communicate with elements in the middle level of plate 3, as will be described below. The openings 25 are shorter and communicate with the inner row of holes 23 of plate 2. Therefore, the openings 24- will be filled with solution A and the openings 25 with solution B. The solution labeling is repeated for the first four openings in FIG. 5. The openings 24 and 25 are really depressions in the rear-most portion of the plate 3 which is shown generally at 26 in FIGS. 3, 4A and 4B, as well as FIG. 5. It will be apparent that the plate 2 closes the rear of the depressions 24 and 25 transforming them into cavities with inlets from the holes 22 and 23 of plate 2 respectively.

Turning now to FIGS. 3 to 5, it will be seen that there are concentric interrupted openings 27 and 28, as will be seen from FIG. 5. These grooves are flared, FIGS. 4A and 4B. The appearance on FIG. 5 that the grooves would extend into two chambers 24 and 25 is because the flared portion is seen, but the grooves do not extend into these two chambers, since each only communicates with a single chamber.

The front level of plate 3 is shown at 29 in FIG. 3. It also appear in the section in FIG. 2, but the illustration is clearer in FIG. 3, which is to a larger scale. The openings 27 are opposite solid portions 30, and corresponding solid portions 31 are opposite the openings 28. These solid portions are staggered, as appears more clearly from the left hand portion of FIG. 5 and FIGS. 4A and 4B.

In the front plate 29 are also machined septa 32 which are in the form of short concentric cylinders. Each septum is opposite the center of an annular row of spinnerette holes 20 of spinnerette orifice plate 4. It will be seen from a consideration of FIGURES 2, 3 and 5 that different streams A and B of polymer solutions flow on different sides of each septum. With the orific plate 4, the septa define annular channels through which the different polymer solutions flow. The particular solutions A and B are shown for a few septa in FIGS. 3, 4A and 4B, and the flow of solution B appears most clearly from FIG. 3, whereas the upper portion of FIG. 2 shows the flow for solution A. For clarity, only a few septa are shown in FIG. 2. It will be seen from FIG. 2 that a very fine mesh screen 21, for example a ZOO-mesh screen, is between the edges of the septa and the orifices in the back of plate 4. This screen produces slight turbulence at the interfaces of the streams of the two polymers.

The spinnerette stack, as shown in FIGS. 1 and 2, ex trudes into a coagulating bath, as has been described above. It will be seen, particularly from FIG. 2, that there is a fairly wide central passage through the spinnerette stack. Through this passage coagulating bath flows. The filaments are extruded in the form of a hollow tube, coagulating bath flows between them both from the outside and from the inside, as described above. The travel transversely through the tube of filaments is sufficiently short so that there is no substantial change in the characteristics of the coagulating bath for different fibers, and therefore the problem of White core is solved. The great effect of the short travel will be discussed below in connection with specific tests.

It will be seen from FIG. 3 that the septa are machined into the solid plate 29. They are, therefore, in the form of low, concentric cylinders with one end part of a rigid backing plate. This gives the septa the maximum strength against bending and greatly reduces or completely eliminates problems of bending of septa when the head is disassembled for cleaning, which, as has been pointed out above, must be effected periodically in practical machines. At the same time, the automatic alignment of septa with rows of orifices is fully maintained. In other words, all of the advantages of the horizontal stack spinnerette heads of the Sulich application referred to above are shared by the present invention. This represents an unusual and advantageous situation, because generally when apparatus is modified to eliminate one drawback, this is at the expense of certain other features. In other words, usually a compromise results. In the present case the important advantages of the solution of the white core problem and maximum insurance against bending of septa on cleaning are obtained without any offsetting disadvantages.

FIGS. 1 to 6 represent a very economical form of the invention in which the final distribution plate or element is made up of several portions and lends itself to cheap and simple fabrication. FIGS. 7 and 8 illustrate a modification which operates in the same way with maximum rigidity of the final distribution plate but which presents the disadvantage of somewhat more expensive machining or fabrication. As the plates or elements 1, 2, 4, 5 and 6 are the same as in FIG. 1, they bear the same reference numerals. The flow of solutions A and B through them is also the same, and the description will not be repeated. However, the final distribution plate or element 33 is different and is illustrated in detail in FIG. 8, which shows part of the element broken away, exposing three different levels. Contrary to the element 3 of the earlier figures, this plate is in one piece, and FIG. 8 shows a view from the rear and not from the front, as is the case in FIG. 6. The rear face of the plate is provided with depressions 34 and 35 which, though slightly different in shape, perform exactly the same functions as the corresponding elements 24 and 25 of FIG. 5. In other words, with the plate 2 they form conduits or chambers which communicate with the different staggered holes 22 and 23 of plate 2. In the same manner, the resulting chamber 34 receives solution A and 35 solution B.

Looking at the portion of FIG. 8 immediately to the left of center, it will be seen that there are a radial series of slots 36 and 37. The metal between the slots in the rear portion of the plate, shown to the right of the central break in FIG. 8, assumes the form of a sinuous baffle 38. It will be seen that this bafiie directs solution A from a chamber 34 into the slots 36 and solution B into the slots 37.

The front face of the plate 33, which appears at the extreme left of FIG. 8, has concentric septa machined in, the septa being the same short cylinders as appear in FIG. 5, but they are machined integrally in a single plate instead of a separate element. As the septa constitute the same type of element and perform the same function as in FIGS. 2, 3 and 5, they are designated by the same reference numeral 32.

As the alternate slots 36 and 37 receive solutions A and B respectively, as has been described above, it will be apparent that the two solutions flow on the different sides of each septum in exactly the same manner as in the modification shown in FIGS. 1 to 6. In other words, plate 33, although of different mechanical construction, performs exactly the same function as the multiple element, distribution plate 3 of FIGS. 1 to 6. While plate 33 is illustrated as a single, unitary structure, it can be fabricated as a multiple structure element. The feature of the sinuous baffie would still be retained. While the modification of FIGS. 7 and 8 may be somewhat more expensive to make, when in the single, unitary form it operates with the same efficiency and, in fact, as far as bending of the septa is concerned, it is, if anything,

9 more rigid than the modification of FIGS. 1 to 6. As, however, the rigidity of the septa in the earlier modification is entirely adequate, multiple element plates may be preferable wherever they are cheaper to manufacture.

It should be noted that in its broader aspects the present invention requires means for distributing the two polymer solutions on either side of each cylindrical or concentric septum, and the final distribution plates 3 and 33 represent only two possible mechanical distribution means. They are, therefore, illustrative only, though in a more specific aspect they constitute desirable mechanical structures.

While the apparatus of the present invention, as such, is not concerned with the exact chemical nature of the solutions of filament-forming material, except insofar as they must have suitable viscosities, a typical illustration of the process aspect of the present invention is as follows:

Using the apparatus of FIGURES 1 to 6, dissimilar solutions of acrylonitrile polymer dissolved in hot concentrated aqueous sodium thiocyanate (solutions A and B, previously identified) were pumped into pipes 13 and 15 to extrude bi-component filaments from orifices as previously described. Cold dilute aqueous sodium thiocyanate flowing through channed 7 and around the extrusion head rapidly coagulates the fresh extrudate by cooling and dilution to form wet gel bicomponent filaments.

Because of the relatively small distance through the moderate number of rows of extruded filaments, they are uniformly bathed in the cold coagulating bath. The length of travel is insufiicient to change significantly the temperature of the bath or to produce too high a thio cyanate content. After the filaments are coagulated they are washed in water, stretched, and dried in the usual manner. As this portion of the operation is not significantly changed by the present invention, it will not be further described in detail. When the present invention is used to produce other multi-component filaments, the nature of the solutions and of the filament-forming materials will of course be different. However, in every case the advantages of relatively short travel of the coagulating bath through the tubular rope of filaments are obtained.

In order to test out the effects of varying proportions of cross section of the central conduit and the annular face of the extrusion head, a head for very coarse carpet thread was used, which represents a severe problem in coagulation. T o produce a tow of about 16.5 denier carpet fiber, a spinnerette having 3108 orifices each having a diameter of 150 was used. The cross-sectional area of the whole box containing the spinnerette and coagulant bath was 79.1 in. the annular portion of the spinnerette head 36.8 in. and the center hole 3.8 in. In other words, the ratio of area of the center opening to annular face was just over 10%. No problems of white core resulted and spinning was excellent.

When the ratio dropped below 4.6%, satisfactory operation no longer took place. In other words, under the conditions of the test, for safe, practical operation the ratio should not be significantly below 5%.

The etfect of fiow through the central orifice was tested qualitatively by blocking off any flow around the outer periphery of the extrusion head. Because the back of the bath container overflowed under such conditions, it was not possible to maintain the full fiow through the central orifice. Nevertheless, although the flow was substantially less than used previously, spinning conditions were normal, there were no broken filaments or capstan Wrap-ups. This was, of course, an extremely drastic test, showing the enormous effect of the bath going through the center opening. For practical operating machines, such conditions would be ridiculous and so, the extrusion heads of the present invention will normally be used in practical operation with bath flowing both through the central opening and around the edges of the head.

When the speed of extrusion was increased to 125% of that used in the prior tests, excellent results were obtained, other conditions remaining constant, whereas with an ordinary head, as described in the Sulich or Douglas and Tonnies applications referred to above, such increased speed resulted in unsatisfactory operation.

Comparison tests were conducted to see whether the salt concentration in the coagulation bath could be increased above that previously used. It was found that the salt concentration could be increased as much as 50% without encountering white core problems. Operation of the processing equipment with increased salt concentration in the coagulant reduces the steam load on the evaporators by reducing the amount of water to be evaporated in reconcentrating the coagulant for reuse in preparing spinning solutions. This is an important economic advantage of the present invention.

When coagulant flows into a bundle of freshly extruded filaments, the concentration of the coagulant rises as it extracts salt from the coagulating filaments. In order to avoid the white core problem, the coagulant concentration should not be permitted to rise above a certain maximum so no single filament gets exposed to too high a concentration of coagulant. Since the coagulant concentration rises as the coagulant penetrates the fiber bundle, the coagulant fed to the spinning apparatus must have a low enough concentration to allow for this rise without reaching that level at which the white core problem exists. Since the flow path of coagulant through the fiber bundle is greatly shortened by the present invention, the concentration rise in the coagulant is greatly lessened. Thus, the coagulant fed to the spinning apparatus using the present invention can be much higher (up to 50% higher) without defective filaments due to improper coagulation.

Comparison tests were made to show the effect of eliminating the solution distribution plates and as, of course, bi-component filaments cannot be produced without distribution plates, the spinning was of single-component filaments. This does not change significantly the results. The spinnerette used had 10,308 holes of p. diameter. First, spinning was effected with the distribution plates eliminated and then with them added. The addition increased the maximum speed at which it was possible to spin by approximately 50%.

A comparison test was made with a vertical stacked ead according to the Douglas and Tonnies application referred to above, and the maximum speed permissible when making bicomponent fibers was only a little over half as great as when making bicomponent fibers using the present spinnerette assembly.

It is not known why the distribution elements of the present invention also permit higher speeds, and it is not desired to limit the present invention by any theoretical explanation. However, the results are practically valuable and show that the present heads, even if used with a single spinning solution, though introduced of course as two streams uniting in the orifices, improved results are obtained, which is a practical advantage as it is possible to shift from a run with bi-component threads to one with mono-component where the exigencies of a particular operation make this additional versatility desirable. The improved results are, of course, not as spectacu- Iar or as important as when bi-component filaments are spun, but it is an additional advantage of the present invention that the heads are also better even for monocomponent operation.

I claim:

1. In an annular spinnerette assembly comprising an annular spinnerette having a plurality of orifices disposed in concentric circular rows, means to introduce two separate spinning solutions thereto, and an annular distributor mounted adjacent said spinnerette to keep separate said two spinning solutions until just prior to flowing both said spinning solutions side-by-side through each orifice of said spinnerette; the improved distributor comprising an annular plate provided with (a) a plurality of annular septa defining a plurality of annular spinning solution channels in one face thereof, (b) a plurality of slots in the base of each such channel extending through such distributor, such slots being disposed in radial arrays with each such slot in any single such spinning solution channel being adjacent to and substantially coextensive with the slots in the next adjacent spinning solution channels, and (c) means to direct the two spinning solutions alternately into slots in alternate spinning solution channels.

2. A spinnerette assembly as defined in claim 1 wherein said last means comprises recesses in the other face of said distributor separated by sinuous bafiies which snake around each said radial array of slots so that alternate slots connect with one recess and the remainder of said slots connect with the adjacent recess, and means to separately supply different spinning solutions to each pair of such adjacent recesses.

References Cited UNITED STATES PATENTS 2,047,313 7/ 1936 Dreyfus 18B 2,536,092 1/1951 Roberts 188 3,245,113 4/1966 Sulich 188 3,248,466 4/1966 Woodell 264168 X 3,341,645 9/1967 Horiuchi et a1. 264-181 X JULIUS FROME, Primary Examiner.

J. H. WOO, Assistant Examiner.

Images