Flat multi-signal transmission line cable with plural insulation

Abstract

A multi-signal transmission line is formed of a flat cable having a plurality of generally parallel conductors embedded in a dielectric core material, with an insulator jacket encasing the flat cable and being made of a dielectric material having a higher dielectric constant than the dielectric core material of the flat cable. The resulting composite transmission line cable insures that substantially all of the transverse electromagnetic propagation field created by the passage of a fast rise time pulse in a signal conductor is confined to the geometric area of the cable and the cable functions in a manner to greatly reduce the far end line-to-line interference (crosstalk) between the signal conductor and adjacent quiet lines.

Description

This is , for example, by reference to FIG. 10 a geometry for the thickness of the polyethylene core in coordination with the conductor arrangement where the majority of the electromagnetic field will be caused to propagate within the low loss dielectric, and attenuation or propagation velocity will be affected very little by the surrounding jacket. In the prototype design, CA-490, approximately 98 percent of the field propagates in the polyethylene cable core wherein in accordance with FIG. 10, the ratio of the core thickness to the conductor pitch is about 2∅Iaddend..

However, the composite multi-signal transmission line cables of the subject invention may be designed with thinner cable cores with certain trade-offs on the propagation velocity and attenuation. In these designs the higher dielectric constant material of the insulator jacket wil contribute more to the composite dielectric constant by the percentage of the square root of its dielectric constant weighed by its field effect. Consequently, the propagation velocity in such cable will be somewhat slower; still the beneficial electrical effects achieved by the cable of the subject invention will apply and crosstalk will be limited to acceptable levels.

The prototype cable CA-490 was designed for unbalanced systems; however, the basic concept of the subject invention applies as well for balance pair designs and improves crosstalk in both cases.

CHARACTERISTIC IMPEDANCE

The Characteristic Impedance of signal transmission lines depends on the size, shape and location of the conductors and on the dielectric constant of the insulation material. In order to establish a formula with measurable properties, the following basic relationship was utilized:

(Z_o)(c)(C)=1

where

Z_o =characteristic impedance, ohms, in air

c=velocity of propagation, meter per second, in air

C=capacitance, Farad per meter, in air

A simplified form of this with practical units:

Z_o =1.016/C

where:

C=capacitance, picofarad per foot, in air

To account for the dielectric material:

Z_d =Z_o /(ε)^1/2

where:

Z_d =characteristic impedance in cable

Z_o =characteristic impedance in air

ε=the effective dielectric constant, relative to air

To ease the prediction and calculation of the prototype cable's characteristic impedance a formula was established based upon the dimensions of the cross-sectional geometry:

Z_d =1/(ε)^1/2 42 cosh-1 (.Badd.1.8ײ -(1.25ײ -1) (1.25ײ -1)^1/2

where:

x=pd

p=pitch, signal to ground

d=diameter of conductor.

The above formula was found to be in close agreement with actual measurements. It gives direct applicability for multi-conductor cables where both signal and ground conductors are round wires. Therefore, further explanation is needed for the characteristic impedance of flat-round-flat conductor arrangements, the design used for the prototype cable and illustrated in FIG. 1. Since the thickness of the flat conductors is equal to the signal wire diameter and the corners are radial, the following consideration was accepted:

p=s+d

where:

s=the separation between the round and rectangular conductors

Recognizing the dimensions for the conductor arrangement in the prototype cable:

d=.Badd∅0113 inch

s=0.00685 inch

p=0.01815 inch

the calculation gives Z_o =76.2 ohm characteristic impedance in air.

Having 98 percent of the field propogating in the polyethylene and 2 percent in the insulator jacket, results in a square root of the composite dielectric constant equal to 1.525; consequently, the characteristic impedance of the prototype cable:

Z_d =Z_o /(ε)^1/2= 76.2/1.525=50 ohms

PROPOGATION VELOCITY

The electromagnetic energy propogates in free space with a velocity of .Badd.3(10)⁸ meters per second. Expressing this in more practical units on a time delay base:

TP_o =1.016 nanosecond per foot, in air and

TP_d =TP_o (ε)^1/2 nanosecond per foot, in cable where:

TP_d =propogration time in cable.

The signal propogation time in the prototype cable CA-.Badd.490 is 1.55 nsec./foot. The polythylene cable core without the jacket yields 1.53

nsec./foot propogation delay. These measured results are comparable to calculated values based upon the assumption that 98 percent of the TEM (transverse electromagnetic) field propogates within the polyethylene. These readings were taken at the 10 percent level of the input pulse rise time, t_r =0.18 nanoseconds.

ATTENUATION

Signals transmitted through flat cables are attenuated along the lines. This attenuation is due to conductor losses and insulator losses; both are frequency dependent. Copper losses are affected by the square root of frequency, and insulation losses by the frequency. The measure of attenuation is expressed in decibel/foot at sine wave frequencies and by the slope change of the rise time at pulse type signals.

The shape of a selected pulse rise time (1,5, or 10 nanoseconds, etc.) may be matched by the ascending half of an equivalent sine wave frequency through a two channel oscilloscope. Such measurements showed good agreement with the following formula:

t_r =0.295/f_o

where

t_r =pulse rise time 10% to 90%, seconds

f_o =corresponding frequency, in Hz

SINE WAVE ATTENUATION

Conductor losses may be expressed by the following formula:

A_c =0.75(f_MHz)/Z_d d

where:

A_c =attenuation of the copper conductor, dB/100 ft.

Z_d =characteristic impedance of cable, ohms

d=diameter of copper conductors, inches The formula for insulation losses:

A_d =2.78(ε) D_f f_MHz

where:

A_d =attenuation of the insulation, decibel/100 ft.

ε=dielectric constant, relative to air

D_f =dissipation factor

The calculated attenuation for the prototype cable in decibel/100 ft. units:

______________________________________
Frequency in MHz
10         100     1000
______________________________________
Ac = (1.327) (fMHz)
4.20       13.27   42.0
Ad = (0.008) fMHz
0.08       0.08    8.0
Af =      4.28       14.07   50.0
______________________________________

FIG. 7 shows the calculated and measured Sine Wave attenuation values on the prototype cable (CA-490).

PULSE RISE TIME ATTENUATION

Both the edge and magnitude of the pulse are attenuated through a length of signal transmission line. These losses are due to the cable conductors and insulator, and will become evident by comparing the input and output shape of the pulse. In a given cable these losses are affected by the input rise time of the pulse and also by the length of cable.

To calculate the rise time attenuation a formula was developed for coaxial cables:

T_o =4.56(10)-7 (A_o² /f_o)l²

where:

T_o =time in seconds, needed for the output pulse to reach the 50 percent reference level of the input rise time

A_o =A_c +A_d in decibel/100 Ft. units

f_o =0.295/t_v ; frequency in Hz

l=length of cable in feet

t_r =input pulse rise time in seconds.

This formula applies to a theoretically vertical pulse edge or at least to a very fast pulse rise time. In practice, however, it is necessary to deal with 1 or 5 nanosecond rise time pulses. When the actual input rise time is comparable or slower than this calculated rise time attenuation of the cable, both should be considered. Referring to the tabulation of FIG. 6, with t_r =5 nanosecond pulse input, the output rise time based on the T_o calculation is only 2.515 nanoseconds. It is obvious, however, that when the cable is input with a pulse edge of 5 nanoseconds, the output should be at least 5 nanoseconds, but cannot be less. Consequently, it seems necessary that the actual input rise time would be accounted for in the calculations. Using this method for predictions, a reasonable agreement can be found to the actual measurements. FIG. 4 is a tabulation showing the calculation of T_o. FIG. 5 is a tabulation showing the schedule of the different amplitude levels of the actual input pulse, normalizing the 10-90% rise time for the tabulation of FIG. 4.

The prototype cable CA-490 was tested for rise time attenuation with three different pulse rise times:

0.18 nanoseconds

PAC 1.0 nanoseconds

5.0 nanoseconds

Results are shown in the tabulation of FIG. 6 in comparision with the calculated values.

LINE-TO-LINE INTERFERENCE (CROSSTALK)

Bundled twisted pairs, triplets and conventional multi-signal flat cables generally give no particular difficulties with certain transmission line parameters, such as characteristic impedance, propagation velocity or attenuation; however, line-to-line interference or crosstalk becomes a problem with fast rise time pulses or high frequency signals; for such signals coaxial cables are used at present for adequate overall performance.

In describing crosstalk between closely located signal transmission lines, the generally used terminology throughout the industry is:

1. Signal line: consists of signal and ground conductors;

2. Active line: conducting signal line;

3. Quiet line: nonconducting signal line;

4. Near End crosstalk: interference measured in

Quiet line at the end where signals enter the Active line;

4a. with pulsed signals: Fast crosstalk and peak crosstalk;

4b. with sinusoidal signals: the maximum level;

5. Far End crosstalk: interference measured in the

Quiet line at the load end of the Active line;

5a. with pulsed signals: Differential crosstalk; peak crosstalk

5b. with sinusoidal signals: the maximum level;

6. Fast crosstalk; reaches maximum magnitude when twice the propagation time of Quiet line is greater than input pulse rise time: this is a miniature replica of the input pulse; the width is equal to twice the propagation time of the Quiet line; same polarity as the input signal;

7. Peak Near End crosstalk: develops generally at the end of the Fast crosstalk; may be caused by fringing field effect or termination mismatch;

8. Differential crosstalk: spike shaped, opposite polarity than the input pulse; magnitude depends on fringing field effect at the cable's outskirt and on the length of cable;

9. Peak Far End crosstalk: either polarity;

10. Matched-terminated: the output impedance of the generator (pulse or signal) is matched to the characteristic impedance of the Active line (lines);

The input impedance of the measuring instrument (oscilloscope) is matched to the characteristic impedance of the signal line being tested (Active or Quiet); all other signal lines are terminated to loads equal to their characteristic impedances.

In conventional multi-signal transmission line cables the Differential Far End crosstalk causes the highest level of interference; at the same time this is the area where the composite multi-signal transmission line cable of the subject invention is most effective. To study the differences, test results on the prototype cable (CA-490) were compared to those measured on a basic polyethylene cable core (CA-489); the latter may be considered a good quality conventional flat cable. FIG. 8 shows oscilloscope traces of the crosstalk at the Quiet line output measured by utilizing the following pulse rise times: 0.18, 1, 2, 5 and 10 nanoseconds consecutively for the Active line input signal. For this test one Active line was driven adjacent to the Quiet line. This condition offers an optimum fidelity and clarity for obtaining and studying shapes of different crosstalk pictures because the characteristic impedance of the cable specimen can be matched to the impedances of the Pulse Generator and Oscilloscope (50 ohms) without special network means. The improvement exhibited by the jacketed, composite multi-signal transmission line cable of the subject invention may be concluded by the following tabulation:

______________________________________
Crosstalk in Percent
Basic      Prototype
tr  Polyethylene
Cable
nsec.    CA-489     CA-490        Improvement
______________________________________
10       0.3        0.05          6    times
5        0.6        0.1           6    times
2        1.5        0.1           6    times
1        3.0        0.5           6    times
0.18     8.6        1.1           7.8  times
______________________________________

FIG. 8 is a representation of actual photographs showing far end crosstalk with each of the five different pulse rise times on both pictures. On CA-489, the base lines are aligned for easier reading. The crosstalk control is clearly visible in the reduction of the spikes and are most evident at the fast rise time pulses.

The reason for this crosstalk reduction is the physical difference between the basic polyethylene cable CA-489 and the prototype cable CA-490; i.e., the jacket. The latter makes the cable thicker and it also has a higher dielectric constant and higher dissipation factor than the polyethylene cable core.

A reasonable question arises; was the crosstalk reduced by the thicker cable or by the different dielectric? For an answer to this query, another cable (CA-580) was built with a polyethylene jacket around the basic polyethylene cable core with dimensions identical to the prototype cable CA-490. The Far End crosstalk measured on this specimen is shown in FIG. 9.

Comparing these measurements to the previous two cables, the following conclusions may be reached:

1. Far End crosstalk improved with the thicker polyethylene cable (CA-580) compared to the thinner one (CA-489). However, the composite multi-signal transmission line cable of the subject invention utilizing different dielectric materials confined the crosstalk far greater.

2. The differential spike (opposite polarity than the signal pulse) at fast rise time pulses still characterizes the thicker polyethylene cable, while the same is diminishing in the composite multi-signal transmission line cable of the subject invention (with t_r =0.18 nsec., it is 2.6 percent in cable CA-580, while only 0.5 percent in prototype cable CA-490).

For the 10 foot prototype cable, the Near End crosstalk was measured with one Active line adjacent to the Quiet Line.

The above tests clearly demonstrate the crosstalk controlling features of the multi-signal transmission line cable of the subject invention. However, for quantitatively worst case results, tests were made driving four Active lines (two at each side of the Quiet line) on 10-foot and 50-foot long prototype cables. In each instance, the performance of the cable made according to the teachings of the subject invention were substantially better than conventional flat cable transmission lines.

It is the inventor's conclusion that the construction of the subject multi-signal transmission line cable, and in particular, the difference in the dielectric constants of the insulated jacket and the core material results in the extremely desirable electrical characteristics. The inventor is not apprised of a definitive explanation at this time as to why the provision of an outer jacket having a higher dielectric constant provides the surprising and unusual, and extremely desirable result, of reducing the differential far end crosstalk on adjacent quiet lines, when a fast rise time pulse is applied to an active signal line. It is believed that the basic TEM mode in its ideal form is affected adversely by the surrounding air at the boundaries of conventional multi-signal flat cables causing "Differential" cross-talk in the adjacent signal lines at the far end of the cable. Differential crosstalk is created by the transients (leading edge, trailing edge) of a pulse and always has a polarity opposite to the direction of the swing in the Active line pulse. The magnitude of the Differential crosstalk is increased by both: faster transient times; and length of cable. The term "first mode" of propogation may be used to describe this air-affected fringing mode.

The jacket of the invention alters this harmful "first mode" of propogation by changing the character and magnitude of the "Differential" crosstalk. The term "Second mode" of propogation may be used to describe the propogation of the subject jacketed cable. It is theorized that the higher dielectric constant of the insulator jacket may effectively prevent or effectively the nature of the "first mode" propogating fringing field of the input pulse signal so as to substantially cause attenuation of the far end crosstalk in the adjacent signal line.

It is believed that the provision of the higher dielectric constant outer jacket may excite a "second mode" of propogation in the Active signal line, and that may be of an opposite polarity to the interference or crosstalk created by the "first mode" electromagnetic field, whereby the resulting effect on the adjacent Quiet line at the far end is a substantially reduced far end crosstalk level.

Stated differently, the construction of the subject composite multi-signal transmission line cable, and specifically the arrangement of the higher dielectric outer jacket, may excite a second mode of propogation. The effect of this "second mode" on the adjacent quiet line may be of a positive interference or crosstalk effect, whereas the "first mode" crosstalk may be of a negative value, whereby the total of the impositions of the first and second modes on the adjacent quiet line is either a cancellation or a substantially reduced crosstalk signal which is below the allowable limits for operation of the circuitry.

It is anticipated that further experimentation may yield a definative answer as to the operation of the system. Nonetheless, it has been positively determined, that this specific construction of a flat cable encased in an insulation of a higher dielectric material than the core material provides extremely beneficial electrical characteristics and, when the outer jacket is made of certain desirable materials such as vinyl, it additionally provides mechanical properties which greatly enhance and increase the value of the resulting transmission line cable.

It is noted that the desirable characteristics of the subject composite cable appear to be further enhanced by the selection of insulation materials which provides that the lossiness or dissipation factor (i.e., the property of a dielectric material to dissipate energy) of the inner core material 13 is less than the lossiness or dissipation factor of the insulator jacket 16.

It is understood that the present invention is susceptible to various modifications, changes and adaptations, and the same are intended to be comprehended within the meaning and range of equivalence of the appended claims. For example, it is readily apparent that other dielectric materials may be employed in the subject invention, with the one limitation being that the other insulator jacket have a higher dielectric constant than the inner core material in which the conductors are embedded. Furthermore, although the invention has been described with respect to alternating flat and round conductors, it is also readily apparent that other conductor configurations, for example, two rectangular-shaped conductors interposed between adjacent round conductors, may also be employed, in which case the parameters "w" "s" and "p" will be adjusted accordingly. The parameter "w" would be the combined width of the several flat conductors disposed between the round signal conductors, whereas the parameters "s" would be the spacing between the signal conductor and the nearest rectangular conductor. Furthermore, the parameter "p" would likewise be the spacing between a signal conductor and the adjacent imaginery ground conductor disposed withih the adjacent flat conductor. Also, the composite multi-signal transmission line cable of the subject invention may be constructed so as to include only round conductors. Likewise, the conductors of the subject multi-signal transmission line cable may take the form of a plurality of generally parallel conductors made up of twisted pairs or twisted triplets, in which case the multi-component conductors will be embedded in a core having a lower dielectric constant, about which is disposed an insulator jacket made of a material having a dielectric constant greater than the dielectric constant of said inner core.

It has been empirically determined that the combined thickness of the jacket is preferably approximately two-thirds of the thickness of the cable core which is capable of confining within its cross-sectional area approximately 98 percent of the TEM field. Stated differently, the combined thickness (i.e. above and below the cable core) of the jacket should be 2/3 of the thickness of the hypothetical cable core capable of confining 98 percent of the TEM field, in those instances where the thickness of the cable core is insufficient to confine 98 percent of the TEM field.

Claims

1. A composite multi-signal transmission line cable for transmitting fast rise time electrical pulses with minimum far-end crosstalk comprising:

a flat multi-conductor cable including a plurality of generally parallel conductors embedded in a planar sheet of insulation material having a dielectric constant, adjacent conductors being spaced apart at a predetermined pitch with selected conductors adapted to be connected to ground, while remaining conductors are used as signal-carrying conductors the ratio of the thickness of the insulation material to said pitch being at least 2.0 such that at least about 98 percent of the transverse electromagnetic (TEM) field propagates within said insulation material; and

an insulator jacket surrounding the flat multi-conductor cable and in intimate contact with said insulation material; said insulator jacket being made of a dielectric material having a dielectric constant greater than the dielectric constant of said insulation material whereby the electrical effect of the composite of the insulation material and the insulator jacket is to establish a balance between the inductive and capacitive coupling coefficients between adjacent signal-carrying conductors thereby minimizing the far-end cross-talk.

2. A composite multi-signal transmission line cable as in claim 1 wherein the insulation material is polyethylene, and the insulator jacket is made of vinyl.

3. A composite multi-signal transmission line cable as in claim 1 wherein the flat cable includes alternating flat and round conductors, with the round conductors being the signal carrying conductors, and wherein the characteristic impedance of the cable (Z_d) is expressed by the following relationship:

Z_d =(1/.sqroot.ε) 42 cosh^{-1 (1.8×2 -(1.25×2 -1) )}

Z_d =1/(ε)^1/2 42 cosh^{-1 (1.8×2 -(1.25×2 -1) 1/2)}

where

d=diameter of round conductors;

p=spacing between a round conductor and an imaginary round conductor falling within the confines of the adjacent flat conductor;

x=p/d; and ε=relative dielectric constant of insulation dielectric material to air dielectric.

4. A composite multi-signal transmission line cable as in claim 1 wherein the signal conductors are round in cross-section, while the ground conductors are rectangular in cross-section.

5. A composite multi-signal transmission line cable as in claim 1 wherein the insulator jacket is extruded over the flat cable.

6. A composite multi-signal transmission line cable as in claim 1 wherein the thickness of the jacket is approximately two-thirds that thickness of a theoretical cable core capable of confining approximately 98 percent of the TEM

field. 7. A composite multi-signal transmission line cable as in claim

4 wherein the insulator jacket is bonded to the flat cable. 8. A composite multi-signal transmission line cable as in claim 1 wherein the total thickness of the insulation material of the flat cable is sufficient to contain substantially all of the transverse electromagnetic field area

generated by a signal passing through a conductor. 9. A composite multi-signal transmission line cable as in claim 1 wherein the insulator jacket is made of a material which is self-extinguishing when exposed to a

flame. 10. A composite multi-signal transmission line cable for transmitting fast rise time electrical pulses with minimum far-end crosstalk comprising:

a flat multi-conductor cable including a plurality of generally parallel conductors alternating in cross-section between round and rectangular, and embedded in a planar sheet of insulation material having a dielectric constant and a low dissipation factor, the center to center distance between a round conductor and the center of an imaginary round conductor tangent with the facing wall of an adjacent rectangular conductor defining a pitch therebetween, the ratio of the thickness of the insulation material to said pitch being at least 2.0 such that at least about 98 percent of the transverse electromagnetic (TEM) field propagates within said insulation material; and

an insulator jacket extruded over said flat cable, and completely surrounding and in direct contact with said flat cable, said insulator jacket being made of a material having a dielectric constant greater than the dielectric constant of said insulation material, whereby the electrical effect of the composite of the insulation material and the insulator jacket is to establish a balance between the inductive and capacitive coupling coefficients between adjacent signal-carrying

conductors thereby minimizing the far-end cross-talk. 11. A composite multi-signal transmission line cable as in claim 10 wherein the insulation material is made of polyethylene, and the insulator jacket is formed of

vinyl. 12. A composite multi-signal transmission line cable as in claim 10 wherein the characteristic impedance (Z_d) of the cable is expressed in the following relationship:

Z_d =(1/.sqroot.ε) 42 cosh^{-1 (1.8×2 -(1.25×2 -1) )}

Z_d =1/(ε)^1/2 42 cosh^{-1 (1.8×2 -(1.25×2 -1)1/2)}

where

d=diameter of round conductors;

p=spacing between a round conductor and an imaginary round conductor falling within the confines of the adjacent flat conductor;

x=p/d; and

ε=relative dielectric constant of insulation dielectric material to air dielectric.

13. A composite multi-signal transmission line cable as in claim 10 wherein the thickness of the jacket is approximately two-thirds that thickness of a theoretical cable core capable of confining

approximately 98 percent of the TEM field. 14. A composite multi-signal transmission line cable as in claim 10 wherein the insulator jacket is

formed of a material which is self-extinguishing. 15. A composite multi-signal transmission line cable as in claim 10 wherein the insulator

jacket is bonded to the insulation material. 16. A composite multi-signal transmission line cable as in claim 10 wherein the dissipation factor of the dielectric material of the insulator jacket is greater than the dissipation factor of the insulation material of the flat cable.