US3350666A - Coaxial connector - Google Patents

Description

Oct 31, 1967 G. w. ZIEGLER, JR 3,350,666

coAxIAL CONNECTOR Filed April 50, 1963 4 Sheets-Sheet l Oct. 31, 1967 G. w. ZIE'GLER, .1R 3,350,665

, COAXIAL CONNECTOR v Filed April so, 19.63 4 sheets-Sheet 2 aww/MM JM Oct. 31, 1967 G. w. ZIEGLER, JR 3,350,666

COAXIAL CONNECTOR Filed April 5o, 1965 4 sheets-sheet s ci 0?; C Cil cg: ci ci ql @EL ci zOTz, [zafz3 Vz4 ZJZZJZZ. E20

S -1 @ya 87 93^ e7 IN VEN TOR.

GEORGE w Z wenn, JR.

United States Patent 3,350,666 COAXIAL CONNECTGR George W. Ziegler, Jr., Carlisle, Pa., assignor to AMP Incorporated, Harrisburg, Pa. Filed Apr. 30, 1963, Ser. No. 276,714 20 Claims. (Cl. S33-33) This inventionrelates to high performance coaxial connectors and to a method of compensating connectors for broad-band signal frequency use.

It is an object of the invention to provide an improved method of compensating coaxial connectors for operation with minimized signal distortion and loss over a broad range of signal frequencies.

It is a further object of the invention to provide improved coaxial connector constructions having a relatively low, flat VSWR over a broad range of signal frequencies.

The continuing and expanding development of communication equipments utilizing signal frequencies in the kilomegacycle range has produced a complementary need for high performance coaxial connectors of all types and sizes. The history of the technical development of coaxial connectors has been to adapt available connectors by modification to the extended signal frequency ranges and to the newly developed coaxial cables for such ranges. While the features of mechanical interconnection utilized in prior art devicesare often satisfactory with respect to connectors utilized in the higher `frequency ranges, the electrical characteristics of connectors designed for operation in use -with signal frequencies ranging from several hundred to several thousand megacycles have proven to be less than optimum or even inadequate due to limitations forced on the connector design by mechanical considerations.

The present invention provides a novel method of connector design and an improved connector construction related to such method and various parts thereof. In order to fully develop the novel method of the invention and the connector construction contemplated, a theoretical treatment of problems and the contemplated solution will first be given, followed by an exemplary connector construction utilizing aspects of the method of the invention.

In the drawings:

FIGURE l is a schematic diagram of a generalized lossless connector Vincluding examples of mismatch sources;

FIGURE 2 is a schematic diagram depicting characteristic impedance mismatches;

FIGURE 3vis a plot of standing wave ratios resulting from impedance mismatches over the frequency range -12; gHz.

FIGURE 4 is a schematic diagram representing compensation of a discontinuity capacitance;

FIGURE 5 is a schematicV diagram representing compensation with discontinuity capacitance pairs;

FIGURE 6 is a plot of standing wave ratios identified with different capacitance relationships for the capacitance pairs of FIGURE 5;

' F-IGURE 7 is a plot showing the effect of impedance on length of matching sections as related to an optimum standing wave ratio at a frequency of gHz.

FIGURE 8 is a plot showing the effect of impedance on standing wave ratio where the length of a given section is optimized in accordance with FIGURE 7;

FIGURE 9a is a schematic diagram of a connector including intermating halves and typical diameter changes necessary to achieve mechanical interconnection;

FIGURE 9b is an equivalent circuit diagram for the `connector of FIGURE 9;

3,350,666 Patented Oct. 31, 1967 ICC FIGURE 9c is a representation of impedance levels in different sections of an exemplary connector embodiment made in accordance with the invention and having the configuration of the connector of FIGURE 9;

FIGURE l0 is a connector cross-section showing radial laminations of various materials included to facilitate manipulation of the method of the invention;

FIGURE l1 is a plot of standing wave ratio versus frequency showing the performance of the connector example included in the specication as compared with a typical performance curve for the prior art; and

FIGURE 12 is a schematic diagram of a partly compensated connector.

The objective of the coaxial connector is to provide a transmission path between coaxial cable paths, which is at least as good with respect to signal distortion and energy loss as that of the particular cable utilized. Ideally, a given connector should be physically identical to an incremental length of the particular cable to satisfy this requirement. Unfortunately, because of the requirement that cable halves be joined by a connection at least as strong, with respect to pull test, as the cable itself, connector design must deviate from the suggested ideal with various means of interlocking cable halves required. The more usual approach is to utilize threaded or bayonet type connector assemblies adapted to be crimped or otherwise aixed to each half of cables to be joined with some additional structure, including a threaded or bayonet type portion having intermating male and female parts to complete paths for the inner and outer conductors of the cable. The practical result of this requirement is that numerous changes in diameter and in conductive path spacing must be made along the length of the connector. An example of this is shown in U.S. Patent No. 2,540,012, granted Ian. 30, 1951, to Dr. O. M. Salati.

It will be realized that, for each change in diameter of conductive paths or dielectric material laminations, there may occur a change in the characteristic impedance of the connector representing a characteristic impedance mismatch.

Abrupt changes in conductor diameters represent discontinuities which behave as capacitances shunting the conductive paths at each point in a connector at the point at which diameter changes occur. Both factors act as defects which, in varying degrees, produce reflections resulting in interferences between incident and reflected waves carrying a standing wave pattern. This phenomena can seriously reduce the efiiciency of energy flow.

In FIGURE l a broad generalization of any connector is schematically represented to include any type of discontinuity or mismatch source. The length of the connector represented as S is shown to be made up of a series of incremental or finite lengths S1, S2, S3, Sn. Length S exists between a transmission path such as a coaxial cable having a characteristic impedance Z0 and a further transmission path which could be any suitable coaxial Connector interface including a further coaxial cable half having the characteristic-s impedance Z3. Along the length S of the connector are capacitances C1, C2, C3 Cn representing discontinuity capacitances occurring at the possible points of conductor diameter changes. Between each of these capacitances are impedance sections Z1, Z2, Z3, Zn representing the characteristic impedances of each of the sections 1, 2, 3, n. The section lines A-A and B-B represent possible points looking into the connector shown in FIGURE 1 from lengths ds and Sn, respectively.

In general, .the maximum standing wave ratio, SWR, resulting from m individual sources, of any nature, s the product of the individual Icontributions of each source. This is known as the combination principle as discussed book of Design and Performance of Cable Connectors 1 for Micro-Wave Use, Btu-Ships, Index No. N.E.-110718, May 1956. A number of small energy reflection sources may, through their phase relationship, produce a large SWR. This is expressed as:

m pmnx. il1pi l For an impedance match at B-B, FIGURE 1, any Z1, and C1 or any combination of elements may be considered to complete the match; and any is the equivalent of others. This is demonstrated as follows. Consider the length of the transmission line, Sn, of the connector of FIGURE 1, to be the matching device. With respect to admittance, Y, conductance, G, and susceptance, B, for the `admittance at point A-A in FIGURE 1, the general relationship is YA=GA+1`BA.

In terms of Sn and the phase constant ,Bn the foregoing -may be expressed as: 1

Y(Y--.7`Yu tan Bush) YA* GA-i-JBA- Yn jY0 tan nSn The real parts provide the conditions for a conductance match.

:mn-GA) B AYO Eq. 3 For a susceptance match, the imaginary parts provide the required condition.

tan uSn B AY.. tan SfGAYO- Y.; Eq. 4

For an admittance match BA:[(1--GAZ0)(GAYtr-Yr)l'A Eq- 5 The match is frequently sensitive except for singular points, as indicated by the relationship obtained from Eqs. 5, 6, and 7 Because of this, and in general, a perfectly flat line, SWR=1, as viewed from point B-B over a broad-band of frequency, cannot be obtained by compensating mismatches. In other words, by following the above relationships, compensation to substantially pB=1 may be obtained at few, if any, of the frequencies present in a broad-band frequency spectrum. However, within the rigorous restri-ction of Eq. a practical approach to a flat line may be made. The method of the invention contemplates achieving a connector having a relatively at standing wave ratio over a broad range of frequencies by the use of novel arrangements of compensating sections and spacing therebetween. This approach contrasts materially from that of the prior art, as for example, in the Salati patent above mentioned.

In FIGURE 2 a schematic diagram of a connector is shown having a length S and an impedance Z to interconnect transmission line paths having characteristic impedances Z0. The standing wave ratio, pA as viewed at A-A may be expressed as:

PA Eq. 10

where IA represents the reflection coefficient at A-A. The characteristic impedance Z being greater than Z0 represents a mismatch when viewed from B-B such that FIGURE 3 shows pA and p15 in a plot of p versus j wherein f is in gHz.; Z0=50 ohms; Z=52 ohms; S=0.1f)\. As will be apparent from the curves in FIG- URE 3, when S-a a-l, 3,5, Eq. 12

then

PB=(Z/ZO)2=PA2, Z Zo Eq- 13 This is in conformity with Eq. 1 above. The variation of pB with frequpency is limited to lpp. As will be apparent from FIGURE 3, the envelope, pA2, is the locus of points satisfying Eq. 13 for all values of S. The contribution of a Z-type mismatch may be relatively lar-ge at low frequencies, especially in an electrically long connector wherein :a multicplicity of maxima may occur over a given frequency range. An electrically short connector is desirable for the fewest maxima and careful design with respect to the particular Z0 of the cable must be made to provide minimum values of pmx.

Considering now discontinuity capacitances, FIGURE 4 depicts `a schematic diagram with a connector of length S having characteristic admittance Y1 placed between transmission paths having characteristic admittances Y0 with C representing a particular discontinuity capacitance. Considering w=21rf and as the phase constant for the section S, the conditions for an 4admittance match at vB---B may be expressed as:

For a susceptance match, the imaginary parts of Eq. 14

where c0=2.99776 1010 centimeters per second and K1 represents the effective dielectric constant of connector section length S. This is known as the Griemsmann relationship. For a conductance match, the real parts of Equation 14 yield wCYO tan :0 Eq. 18

Unfortunately, this condition can be realized only in trite cases and the beneficial effect of obtaining a susceptance match from Eq. 15 above, can be largely offset by the conductance mismatch introduced by the compensating section of length S.

The solution to this is shown in FIGURE 5, which is similar -to the schematic diagram of FIGURE 4, but includes an additional capacitance C2 spaced from a discontinuity capacitance C1 by a distance S1. The two capacitances C1 and C2 can be selected to interact to produce an admittance match vat C-C by making YC=Y0, such that the relationship is For a conductance match, the real parts of Eq. 19 give C1=C2=C Eq. For a susceptance match, the imaginary parts of Eq. 19 give @icl-i-Cziyr YOz-Y12-i'0-,ZC1C2 For an admittance match, combining Eqs. 20 and 21 In applying Eq. 17 above, to the connector arrangement of FIGURE 5, it is the practice when C1 and C2 are not equal to merely add their values. Refer to the Griemsmann article, above mentioned land to the work by M. Pomerantz, Modification of Existing R.F. Coaxial Connectors, TM- 1896, September 1957, Fort Monmouth, NJ. This is a satisfactory approximation when C1 and C2 are not substantially different. It is important to note that the form (eC)2 in Eq. 22 does not appear in the Griemsmann relationship above expressed. The term (wC)2 can be much larger than (Y112-Y12), and thereby the controlling term, when: firstly, Y and Y1 do not differ greatly; secondly,

when C is large; thirdly, when w=21rf is large owing to V high frequency; and fourthly, by a combination of the first three effects. From FIGURE 5 the absolute quantity of the Vreflection coeicient may be expressed such that The interaction between C1 and C2 is shown in FIG- URE 6 wherein pis plotted against f in Gh.; C1=0.025/ntf; Z1=Z11=50 ohms; and S adjusted for admittance match at l0 Gh. As can be seen, the standing wave ratio.pA, in.

creases monotonically with frequency in contrast with the single Z-type discontinuity as heretofore indicated in FIG- URE 3; compare pA, FIGURE 3, with p11, FIGURE 6. The interaction of two equal capacitances with increasing frequency produces a series of minima as required by Eq. 22, in the vstanding wave ratio as viewed from C-C alternating with maxima/of increasing magnitude. By applying Eq. 22 at a high frequency, as for example, l() gHz. as indicated in FIGURE 6, a single maximum at an intermediate frequency may be obtained. The magnitude of the standing wave ratio maximum decreases with -a decrease of total capacitance C1|C2. From Eq. 20 and from FIG- URE 6, the provision of equal capacitances of the smallest possible value provide the lowest possible minima with respect to pc. In practice, -a twenty percent difference in the capacitances is acceptable for non-critical applications. For special high-performance applications wherein the cost of tolerance control is acceptable, efforts should be made to make the capacitances as close to equal as possible.

It is to be realized that avoidance of defects causing energy loss is the best practice, but with the practicalities of manufacturing a disconnect type coaxial connection, the best approach is to provide as few discontinuities -as possible with the individual discontinuities being as small as possible. It has been found that the closer the diameter of the center conductor of a connector is to that of the cable center conductor, the smaller the discontinuity capacitances of both inner and outer connector conductors. As can be seen from FIGURE 3 and its associated discussion, it is highly desirable to provide a Z0 match over almost the entire length of the connector. This necessitates :an outer con-ductor diameter discontinuity for every major center conductor diameter discontinu-ity. i'

Considering Z0=50 ohms, a plot may be made depict- Ving Z1 versus connector section length S1 at 10 Gh. for la 6 standing wave ratio pc=l with C2=C1=0.025/.t,uf; as viewed from C`C in FIGURE 5. S1 is expressed in terms of A11, the wave- .length in free space in centimeters. AS can be seen from FIGURE 7, very short sections S1 are required when Z1 is made to be much greater than Z0. Due to the slope of the curve it will be apparent that for Z1 l00 ohms the compensation will be sensitive to normal production tolerances for compensator length S1. In FIGURE 8 a plot is shown of the standing wave ratio pc versus f' for different values of Z1 with S1 adjusted according to FIGURE 7. It will be apparent that by increasing Z1, the v-alue of pc may be made to approach 1.000.

The preceding considerations delineate the general requirements for broad-band compensation. From the curves and equations above given, it should be apparent that the total discontinuity capacitance of the connector should be minimized and most of the total should be made to reside in four equivalent or nearly equivalent discontinuities. Two of these discontinuities representing a pair associated with one connector half are matched with a high Z spacer of a length computed at a high frequency giving a small mismatch maximum tat a lower frequency to achieve the improved p as indicated in FIGURE 8. The other pair of discontinuities are similarly matched. The pairs of discontinuity capacit-ances are matched against each other through being separated by a section having an effective characteristic impedance equal to that of the line Z0, with the section length computed to compensate at the lower frequency. These lengths are calculated from Eq. 21,

above given, by substituting the high or low frequency Values in accordance with w=21r,f, where f is the particular v frequency. The remaining, small discontinuity capacitances (those'omitted in obtaining FIGURE l2 from FIGURE 9b) located between the end pairs of discontinuities are matched out in pairs separated by sections of lengths computed at frequencies intermediate to the high and low frequencies when such capacitances are substantial. The foregoing may be more fully brought to mind by following an exemplary procedure in accordance with the method of the invention. Y Y

The example hereinafter given includes additional discontinuities to those which will be present in certain simpliied connector designs in order to illustrate the various aspects of the method of the invention. In certain instances, depending upon the physical characteristics of the cable to be interconnected, and the mechanical specification for the connector, little compensation will be necessary, whereas in other situations the various steps of the method of the invention will be mandatory to achieve superior electrical characteristics.

Consider now that a connector is to be designed for a coaxial 10, as shown in FIGURE 9a, having an outer conductor 12, Ia dielectric core 14 and a center conductor 16 coaxially disposed within 12. A typical cable of this type has a Z0=50i2 ohms with a center conductor OD dimension of 0.0359 inch and an outer conductor dimension equal to 0.1254 inch. The dielectric material has a constant K*=2.25.

The connector 18 shown in FIGURE 9a, is `comprised of h-alves` 20 and 60, adapted to intermate through cornplementary male and female parts, The connector half 20 is comprised of a stepped metallic shell 22radapted to be threaded onto a forward metallic shell 34. Shell 22 includes a shell extension 30, which may be crimped or otherwise secured to cable 10 in a standard fashion. Shells 22 and 34 include a number of internal diameter changes of the type used to secure the internal parts. The forward internal end of shell 34 includes typical spring finger members 38 which serve to complete the outer path electrical connection between half 20 and half 60. Included in each the center path electrical connection with complementary portions of the contact member 62 through lingers 64 thereof.

The particular means employed to interconnect the connector halves to the cable 10 and the center contact members to the central conductor thereof may be achieved by a wide variety of connector designs. The important consideration with respect to connector 18, is depicted by the schematic diagram thereunder representing the various lengths Si, discontinuity capacitances C1 and impedance values Z, for each section =l,2,3, n.

In accordance with the theoretical procedure heretofore given `and the step-by-step example to follow, the connector shown in FIGUIE 9a was dimensioned and compensated to produce the following values where n represents the stations shown in FIGURE 9a and the diameters are related to FIGURE 10. FIGURE 9b shows roughly the impedance levels existing along the length of the connector of FIGURE 9a.

NOMINAL DIMENSIONS IN INCHES there is a suitable change in half 20 in accordance with Eq. 26 and identical changes in half 60. In essence, the halves 20 and 60 should be arranged to be mirror images with respect to electrical characteristics.

Part of this step is to arrange the discontinuities of the connect-or to facilitate compensation. This is achieved by concentrating seventy-live percent or more of the total discontinuity capacitance of the connector in four capacitances (C1,C2), locating a pair at each end of a connector half. Referencing FIGURE 9b, it will thus be seen that the discontinuity capacitances C1 and C2 are, respectively, represented by a change in the diameter of the outer connector shells with respect to the center conductor 16 to provide C1 and a change in the outer conductor connector shells with a change in -the effective diameter of the center conductor at the largest portion of contact members 42 and 62, to provide C2. invention is to space the discontinuity capacitance pairs (C1,C2) by a high Z section, Z1.

Conductor Dielectric Bead Diameters Diameters Lengths n Sn' Dn du Dn dn Dn l dn' 1 0. 2G13 0. 0359 0. 2140 0.1350 0.1254 0.0359 0.0210 .2943 0900 2923 .0910 d2' D2' 0300 2613 .0800 2593 0810 d3' Da' 2650 2542 .0800 2522 1940 1840 0810 0970 2542 0400 2522 1940 1840 0810 0080 ELECT RICAL PARAMETE RS I1 Zn Cn pn qn Kn t2 nf (B=pt mhos) (S=q t A n) A connect-or so arranged produced a standing wave ratio,

pW from W-W, FIGURE 9a, 0.2 gHZ. intervals over a frequency range, f', of 0-12 gHz., as shown in FIGURE 1l. The pw computed is well below 1.10 at all fre- 45 quencies throughout the chosen range. Also shown in FIGURE 1l is a specitication for a type-N connector.

As a helpful adjunct to the practice of the method of the invention, there exists a procedure which simplies the calculation of the effective dielectric constant through 50 any section of a connector or cable of the type shown in FIGURE 9a. This is shown in FIGURE 10, wherein various radial laminations of dielectric material are repsented by Ka, Kb, Kc, and the diameters of the nth section are Dn, Dn', dn', Dn", dn, dn. From this the 55 effective dielectric constant of the nth section, Kn, may be computed from the expression:

l --K,(Kb-K) 10g g-Kbug-KQ 10g Using this value for Kn, the characteristic impedance of the nth section may be computed from:

KbKo log The next step is to determine the smallest diameter of the contact member portions 44 and 64, that will provide adequate attachment to the cable center conductor and suicient spring force, respectively. While the con. tact members 42 and 62 are shown enlarged to exaggerate the relative dimensions, the shoulder diameters joining the various portions should be minimized or eliminated if possible. The diameter of the most forward portion of 42, shown as pin 46, should be maximized such that the diameter d5 approaches the diameters d3=d4. The enlargement of the dia-meter d5 is limited by the thickness of the spring members of 64 which must provide spring force for the inner contact between the contact members 42 and 62.

The non-compensating sections Z2, Z3 and Z4 may be made to closely approximate the characteristic impedance dll n Z0 of the line. Considering Z0 to be 50 ohms and using Eqs. 25 and 26, above given, the diameters D2, D3, D4, may be calculated by successive approximation to provide the proper Z values. With respect to the air gaps existing in the sections 2, 3, and 4, nominal spacings may be included such as 0.0020 inch between the head and outer conductor and 0.0010 inch between the bead and center conductor with 0.0100 inch being provided at the bead overlap.

With respect to the compensating sections Z1 and Z5, such should be adjusted to a suitable value higher than the characteristic impedance of the cable. For many applications, ZnLlZo is sufficient. For the general case ECI. 25

9 an adjustment wherein LIZOSZnSZZO is preferable. In special applications wherein cost is secondary to performance, a choice of Zn 2Z0 may be made and suitable tight tolerances imposed.

The next step is to calculate values for all discontinuity capacitances using the method of Whinnery and Jamieson; Proceedings of IRE, vol. 32, pp. 98-114 and 695-709, 1944. The provision of C1=C2 may be made by successive approximation altering relative diameters.

Experience has shown that the section S2 should be made as short as is possible. With respect to S5, owing to the interaction of discontinuities of the reentrant type, the value of C is dependent upon S5. When C5 decreases almost linearly with S5. It is preferred to malte S5 as small as is permitted by the accumulation of tolerance in length of the outer conductor parts, allowing bead resilient to take up any remaining tolerances. If the dmeter of 46, d5, has been made as large as possible relative to the portion 44 (diameter=fd3=d4), then C5 should be a negligibly small value. If for design reasons it is not possible to so size portion 46 and the length S5, other procedures hereinafter given may be carried out.

If C4 is small, as it frequently may be, the length S' may be made any arbitrary value vto provide bead overlap. If C4 is appreciable, then an alternate approach may be taken to determine S'. This will be hereinafter explained relative to compensation at intermediate points of the -frequency range.

In many applications it has been satisfactory to make S=1/2)\ at 5 gHz. Alternatively, S may be computed in the manner hereinafter given for the computation of S. The compensation for the major discontinuity capacitances C1 and C2 may be made by adjusting the length S1 in accordance with Eq. 2l, above given, wherein the frequency substituted in the w term of the equation is the highest frequency of the range for the connector; i.e., in the example above given, gHz. In the event that the C3 cannot be made negligible, a different choice of S1 will compensate. Viewing now FIGURE 12, which includes the partinent portions of the schematic diagram of FIGURE 9b, it may be considered that the omitted discontinuities are compensated or other-Wise reduced to a negligible effect. For a susceptance match at B-B with an admittance match at E-E, the relationship is 360q1f degrees Eq. 28

Asvabove indicated, S5 is best minimized. For compensation of the di-scontinuity capacitances C5, a rst procedure is to solve by succes-sive approximation, which is necessary due 4to the interaction previously described. lA graphical method plots C5 vs. S5. over the.practical range of S5 together with C5 vs. an S5 which is computed 10 using Eqs. 3l, 32 and 33. The point of intersection of the curves will provide the desired S5 directly.

If the grhaphical method for solving for C5 and S5 produces curves that do not intersect with values of S5 which may 'be achieved in practice, the curves can be made to intersect by changing the diameter of the pin 46 (diameter d5). It matters not which direction this diameter is changed as long as intersection is obtained and practical dimensions are achieved. Following the change in diameter, the curves may be replotted to locate a point of intersection to yield S5.

The following method is sure to produce compensation but has the disadvantage of complicating the affected connector parts. A simultaneous decrease in Z5 and increase in C5 result from decreasing the diameter D5 over the length S5 until a suitable plot intersection is obtained as above. This procedure provides an annular ring as in the Series N connector. Application of this method to the Series N connector requires, for a `match to 20:50.00, that the ring length be 0.0100 inch. The actual specified value is 0.026 inch, which is thus in disagreement with the instant method.

As heretofore indicated, the major discontinuities of the connector are compensated at the maximum frequency and about half the maximum frequency; in the example shown, 10 gI-Iz. and 5 gHZ. This compensation is at least effective at intermediate frequencies. The remaining small discontinuities may be spaced to compensate the small mismatch from the compensated major discontinuities at about 7.5 gHz., or spaced to compensate each other as follows. Again, from FIGURE 9b, assuming negligible mismatch elsewhere in the connector, S' and S4 may be computed.

S=tan-1[Wi lo-CT)2]=3G0QT degrees Eq. 34 S=qf)\4 Eq. 32 q=0.084730(S)"\/4 Eq. 36 S4=1/z(S-S5)=q4f)\4 Eq. 37 As will be apparent from FIGURES 6 and 8, maxima occur above 5 gHz. for the capacitance pairs involved. The exact frequency at which the maxima occur can be determined for a given ensemble of discontinuities and compensation can be made at that frequency. As a practical matter, 5 gHz. is often satisfactory. Reasonable deviation from the optimum frequency provides flexibility in adjusting the overall length of the connector.

Referring to FIGURE 12.:

Following the arrangement of the various connector lengths, discontinuity capacitances and impedance sections, the standing wave ratio for the connector, pw, may

1 1 be computed at several frequencies. This may be followed by a computation at short frequency intervals such as 0.2 gHz. The result of this is indicated in FIGURE 11.

By following the above method, connectors may be designed having compensation to provide a low, flat standing wave ratio over a broad range of frequencies.

Changes in construction will occur to those skilled in the art and various apparently different modifications and embodiments may be made without departing from the scope of the invention. The matter set forth in the foregoing description and accompanying drawings is offered by way of illustration only. The actual scope of the invention is intended to be defined in the following claims when Viewed in their proper perspective against the prior art.

I claim:

1. An improved broad band compensated coaxial connector for cable of characteristic impedance Z0, including a connector body having inner and outer conductive portions of different diameters and sections along the length of the body forming distinct discontinuity capacitances and having a given total discontinuity capacitance, the said body having a length S comprised of a number of sections of different lengths with one or more sections comprising nearly the total length S of a characteristic impedance Z0, and with the remaining length of S of characteristic impedances each being substantialy greater than Z and with the said distinct discontiuity capacitances being located at predetermined points along the length of said body.

2. The connector of claim 1, wherein most of the total discontinuity capacitance of the connector resides in spaced pairs of discontinuity capacitances.

3. The connector of claim 1, wherein most of the total discontinuity capacitance of the connector resides in spaced pairs of discontinuity capacitances with pair members being of substantially equal discontinuity capacitance.

4. The connector of claim 1, wherein seventy-five percent or more of the total discontinuity capacitance of the connector resides in four substantially equal discontinuity capacitances.

5. The connector of claim 1, wherein seventy-five percent or more of the total discontinuity capacitance of the connector resides in six substantially equal discontinuity capacitances.

6. The connector of claim 1, wherein the connector is comprised of at least two portions, each of which includes discontinuity capacitances and section lengths having impedance values such as to be substantially mirror images.

7. The connector of claim 1, wherein the connector includes two halves with most of the total discontinuity capacitance of the connector residing in at least three pairs of discontinuity capacitances with the pairs positioned such that one pair is proximate the physical center of the connector length and effected by the difference in diameters required to intermate the halves, and one of the other pairs positioned at the major diameter discontinuities in each half.

8. An improved transmission path for coaxial cable having a characteristic admittance Y0 or characteristic impedance ZU where Z0=1/ Y0 comprising in combination, an outer conductive path and an inner conductive path disposed coaxially therein, a dielectric therebetween, discontinuity capacitances C1 and C2 spaced by a section of length S1, having a phase constant 1, a characteristic admittance Y1 or characteristic impedance Z1 where the length S1 being of a value determined at a specific frequency f within the band of frequencies for which the transmission path is to be utilized such that 9. The transmission path of claim 8, wherein C1 is substantially equal to C2.

10. The transmission path of claim 9, wherein Z1 is greater than Z0.

11. The transmission path of claim 9, wherein Z1 has a value in the range of 1.1 to 2.0 Z0.

12. The transmission path of claim 9, wherein Z1 is greater than 2 Z0.

13. The transmission path of claim 8, wherein CZ is not equal to C1.

14. The transmission path of claim 13, wherein Z1 is greater than Z0.

15. The transmission path of claim 13, wherein Z1 has a value in the range of 1.1 to 2.0 Z0.

16. The transmission path of claim 13, wherein Z1 is greater than 2 Z0.

17. An improved broadband compensated coaxial connector for a cable of characteristic value Y0 including a connector body having inner and outer conductive portions of different diameters and sections along the length of the body forming distinct pairs of discontinuity capacitances of values C and Cl spaced apart by a section of length S of characteristic value Y and phase constant determined at a frequency f through the relationship where the said distinct pairs of discontinuity capacitances with given values of Y and existing between pair members are spaced by a length S determined at a high frequency f and the lumped discontinuity capacitance of each distinct pair is compensated against the lumped discontinuity capacitance of another pair with given values of Y and existing between the pairs spaced by a length S determined at a lower frequency f.

18. The connector of claim 17 wherein the said distinct pairs of discontinuity capacitances have given values of Y and existing between pair members spaced by a length S determined at a high frequency f substantially equal to the design maximum frequency for which the connector is to be used and wherein the lumped discontinuity capacitance of each distinct pair is compensated against the lumped discontinuity capacitance of another pair with given values of Y and existing between the pairs spaced by a length S determined at a frequency f substantially equal to half the said design maximum frequency.

19. The connector of claim 17 including other pairs of discontinuity capacitances located between said distinct pairs or adjacent thereto with an ensemble comprised of a said other pair and a distinct pair compensated by a length S at values Y and at the high frequency f and with the resulting ensemble compensated by a length S at values Y and at a lower frequency f to give overall broadband compensation matching to the cable utilized.

20. In a broadband compensated coaxial connector for use with cable of characteristic value Y0, a connector body having inner and outer conductor portions of different diameters in sections along the length of the body forming distinct discontinuity capacitances of values C' and C" spaced apart by a 4section of length S of characteristic value Y and phase constant determined at a frequency f from the relationship.

where the said pairs of discontinuity capacitances have values of Y and therebetween spaced by a length S determined at a high frequency f, a further pair of discontinuity capacitances with further values of Y and existing between said further pair spaced by a further length S determined at same or different high frequency f, the said distinct pairs and said further pair being compensated by spacing the lumped discontinuity capacitance of each distinct pair with respect to a member of said further pair by a length S of characteristic value Y and phase constant determined at a lower frequency f so as to provide overall broadband compensation of the connector to the cable.

References Cited UNITED STATES PATENTS 2,127,408 8/ 1938 Karr 333-97 2,305,456 12/1942 Okable S33-82 2,443,635 6/1948 Morris 339-177 14 2,518,665 8/1950 Collard 333-97 2,540,012 1/1951 Salati 333-97 2,689,294 9/1954 Weber et a1. 333-81 FOREIGN PATENTS 124,582 6/ 1947 Australia.

HERMAN KARL SAALBACH, Primary Examiner.

L. ALLAHUT, Assistant Examiner.

Claims (1)

1. AN IMPROVED BROAD BAND COMPENSATED COAXIAL CONNECTOR FOR CABLE OF CHARACTERISTIC IMPEDANCE Z0, INCLUDING A CONNECTOR BODY HAVING INNER AND OUTER CONDUCTIVE PORTIONS OF DIFFERENT DIAMETERS AND SECTIONS ALONG THE LENGTH OF THE BODY FORMING DISTINCT DISCONTINUITY CAPACITANCES AND HAVING A GIVEN TOTAL DISCONTINUITY CAPACITANCE, THE SAID BODY HAVING A LENGTH S COMPRISED OF A NUMBER OF SECTIONS OF DIFFERENT LENGTHS WITH ONE OR MORE SECTIONS COMPRISING NEARLY THE TOTAL LENGTH S OF A CHARACTERISTIC IMPEDANCE Z0, AND WITH THE REMAINING LENGTH OF S OF CHARACTERISTIC IMPEDANCES EACH BEING SUBSTANTIALLY GREATER THAN Z0 AND WITH THE SAID DISTINCT DISCONTINUITY CAPACITANCES BEING LOCATED AT PREDETERMINED POINTS ALONG THE LENGTH OF SAID BODY.

Images