United States Patent [19J
`Collins et al.
`
`[11) Patent Number:
`[45) Date of Patent:
`
`5,065,118
`Nov. 12, 1991
`
`[75)
`
`[54) ELECTRONICALLY TUNED VHF/UHF
`MATCHING NETWORK
`Inventors: Kenneth S. Collins, Morgan Hill;
`Craig A. Roderick, San Jose, both of
`Calif.
`[73} Assignee: Applied Materials, Inc., Santa Clara,
`Calif.
`[21) Appl. No.: 558,290
`Jul. 26, 1990
`[22) Filed:
`Int. CJ.s ............................................... HOlP 5/04
`[51)
`[52) U.S. Cl ...................................... 333/33; 333/24.1;
`333/160
`[58) Field of Search .............. 333/160, 161, 156, 17.3,
`333/32, 33, 24.1, 205, 207, 22 F, 24.2
`
`[56]
`
`References Cited
`U.S. PATENT DOCUMENTS
`3,384,841 5/1968 Di Piazza ............................ 333/160
`
`FOREIGN PATENT DOCUMENTS
`1214333 4/1960 France ................................ 333/160
`
`21 33
`
`SOURCE
`
`882121 11/1961 United Kingdom ................ 333/160
`
`OTHER PUBLICATIONS
`Moreno, Microwave Transmission Design Data, Dover
`Pub!., N.Y., N.Y., 1948, title page and pp. 103-106.
`Primary Examiner-Paul Gensler
`Attorney, Agent, or Firm-Robert J. Stern; Douglas L.
`Weller; Paul L. Hickman
`
`ABSTRACT
`[57]
`A matching network matches an output impedance of a
`source with an input impedance of a load. The matching
`network includes a plurality of transmission line stubs.
`Each transmission line stub includes a first transmission
`line conductor, a second transmission line conductor
`running parallel to but not in electrical contact with the
`first transmission line conductor, and ferrite dielectric
`material between the first transmission line conductor
`and the second transmission line conductor. A magnetic
`field is used to vary the relative permeability of the
`ferrite dielectric material.
`
`21 Claims, 3 Drawing Sheets
`
`29
`
`22
`
`LOAD
`
`30
`
`DC
`POWER
`SUPPLY
`
`DC
`POWER
`SUPPLY
`
`Page 1 of 9
`
`Samsung Exhibit 1008
`
`

`

`U.S. Patent
`
`Nov. 12, 1991
`
`Sheet 1of3
`
`5,065,118
`
`21 33
`
`Figure 1 45
`
`29
`
`22
`
`41
`
`SOURCE
`
`LOAD
`
`30
`
`DC
`POWER
`SUPPLY
`
`DC
`POWER
`SUPPLY
`
`1
`
`RF
`POWER
`SUPPLY
`
`13
`
`14
`
`7
`
`5
`
`8
`
`10
`
`PLASMA
`CHAMBER
`
`6
`
`Figure 2
`
`3
`
`2
`
`Page 2 of 9
`
`

`

`U.S. Patent
`
`Nov. 12, 1991
`
`Sheet 2 of 3
`
`5,065,118
`
`54
`
`56
`
`51
`
`54
`
`52
`
`Figure 4
`
`Page 3 of 9
`
`

`

` a3DYNOS
`
`LO
`~ CD
`C\I
`~ "--
`::J
`
`C\I
`C\I
`
`~
`
`C»
`C\I
`C\I
`
`C\I
`0
`
`~
`
`co
`C\I
`
`0
`M
`
`~
`
`~
`
`M
`
`~
`
`~
`
`0
`
`~
`
`O> ·-LL
`
`~
`
`M
`(.\J
`
`(0
`Q)
`"--
`::J
`
`O> ·-LL
`
`0
`M
`(.\J
`
`(.\J
`0
`C\I
`
`~
`
`0
`C\I
`
`C\I
`M
`C\I
`
`w
`0 a:
`:::::>
`0 en
`
`~
`
`C\I
`C\I
`
`U.S. Patent
`U.S. Patent
`
`Nov. 12, 1991
`Nov. 12, 1991
`
`Sheet 3 of 3
`Sheet 3 of 3
`
`5,065,118
`5,065,118
`
`c
`ct
`0
`...J
`
`(.\J
`(.\J
`(.\J
`
`LO
`M
`
`~
`
`LO
`M
`
`~
`
`~
`
`(.\J
`M
`
`~
`
`c
`ct
`0
`...J
`
`w
`0
`a:
`:::::>
`0
`en
`
`Page 4 of 9
`
`Page 4 of 9
`
`

`

`ELECTRONICALLY TUNED VHF/UHF
`MATCHING NETWORK
`
`15
`
`20
`
`1
`
`5,065,118
`
`2
`power which is due to dithering of the first variable
`impedance element from the change in reflected power
`which is due to dithering of the second variable impe(cid:173)
`dance element. Using the components of change, the
`BACKGROUND
`S control circuit continuously varies the steady state im(cid:173)
`pedance of the first variable impedance and the steady
`The present invention concerns the connection of a
`state impedance of the second variable impedance in
`first electrical circuit to a second electrical circuit using
`directions which minimize the reflected power. The
`a matching network so as to provide maximum power
`transfer between the first electrical circuit (the
`dithered method of tuning and control always con-
`"source") and second electrical circuit (the "load").
`10 verges to a unique matching solution, even for non-lin(cid:173)
`Maximum power is transferred from the source to the
`ear, dynamic loads. Convergence can be very fast by
`load when the output impedance of the source is the
`using high dither frequencies and magnetic dithering.
`complex conjugate of the input impedance of the load.
`The use of saturable reactors allows the variance of
`In most cases the output impedance of the source is not
`matching network impedance elements quickly and
`naturally equal to the complex conjugate of the input
`without moving parts.
`impedance of the load; therefore, matching networks
`While the matching network discussed in U.S. Pat.
`are placed between the source and load when power
`No. 4,951,009 works well for signals in the radio fre(cid:173)
`control and efficiency are criti~al. A ~atching network
`quency range (frequency less than or equal to 30 Mega(cid:173)
`operat~ properly .when the mput ~pedance of the
`hertz), for high power signals in the very high fre-
`ma~hing network is the complex conjugate ?f the out-
`quency (VHF) range (30-300 megahertz) or in the ultra
`hi h f
`(UHF)
`h rt )
`(300-3000
`put impedance of the source, and the output Impedance
`. ~e~uency
`g
`. r:inge
`. mega e z •
`of the matching network is the complex conjugate of
`parasitic Impedan~es w1thm the magnet1cal~y saturable
`the input impedance of the load. In this way power may
`r~actors are suff_ic~ently large to cause non-ideal opera-
`be transferred from a source through a matching net-
`work to a load with minimal loss of power through 2S t1onal charactenst1cs.
`.
`.
`One al~ernate app~oach f?r matching networks wh1c?
`power reflection, heat dissipation, etc.
`In cases where the input impedance of the load varies
`handle ht~h ~wer signals m the VHF or UHF ra~g~ ts
`during operation it is necessary to make adjustments to
`to use a d1stnbuted parameter ap~r?ach._ In the ?tstnb-
`the matching network to maintain maximum power
`uted parameter approach transmtss1on !me sections or
`transfer from the source to the load. Typically, match- 30 stubs are used to match impedances. In the prior art, the
`impedance of each transmission line stub may be varied
`.ing networks are designed such that variations in the
`input impedance of the load will result in a variation of
`by mechanically moving a short circuit or tap which is
`the impedance of the matching network, the input impe-
`connected to the transmission line stub. However, when
`dance of the matching network being held constant.
`it is desired to quickly change impedances of a matching
`Further, in many applications the output impedance of 3S network, for example in a dithering process, such me-
`a source is an output resistance with a negligible imagi-
`chanical movement is unacceptably slow and unreli-
`nary component. Therefore, in some prior art applica-
`able.
`tions, the impedance magnitude and the impedance
`phase angle is measured at the input of the matching
`networks. Variable capacitors or inductors within the 40
`matching network are varied until the input impedance
`of the matching network matches the output impedance
`of the source network, that is until the impedance phase
`angle is zero and the impedance magnitude matches the
`magnitude of the output resistance of the source. The 4S
`variable capacitors or inductors are placed in the
`matching network so that for every predicted variance
`in the input impedance of the load there is a solution in
`which the variable capacitors are set to values so that
`for the input of the matching network the impedance so
`phase angle is zero and the impedance magnitude
`matches the magnitude of the output resistance of the
`source.
`In U.S. Pat. No. 4,951,009 by Kenneth Collins et al.,
`entitled "Turning Method and Control System for Au- SS
`tomatic Matching Network", techniques are discussed
`in which variable impedance elements are used to re(cid:173)
`place variable capacitors and variable inductors. The
`variable impedance elements are constructed using
`magnetically saturable reactors, such as a transformer 60
`composed of primary and secondary windings wound
`around a non-linear ferromagnetic core.
`Reflective power is removed by "dithering". What is
`meant by dithering is varying at a known frequency or
`frequencies the impedance through the first variable 6S
`impedance element and the impedance through the
`second variable impedance element. A control circuit
`separates out the component of the change in reflected
`
`SUMMARY OF THE INVENTION
`In accordance with the preferred embodiment of the
`present invention, a matching network is presented. The
`matching network matches an output impedance of a
`source with an input impedance of a load. The matching
`network includes a plurality of transmission line stubs.
`Each transmission line stub includes a first transmission
`line conductor, a second transmission line conductor
`running parallel to but not in electrical contact with the
`first transmission line conductor, and ferrite dielectric
`material between the first transmission line conductor
`and the second transmission line conductor. A magnetic
`field is used to vary the relative permeability of the
`ferrite dielectric material. Throughout the discussion of
`the present invention, the term ferrite dielectric mate(cid:173)
`rial means ferromagnetic or antiferromagnetic dielec(cid:173)
`tric material.
`In the preferred embodiment of the present invention
`the first transmission line conductor and the second
`transmission line are coaxial. These may be imple(cid:173)
`mented by electrically conducting pipes placed one
`inside the other. Deionized water may be flowed
`through the inner pipe to remove heat generated by the
`transmission line stub. Alternately, some other fluid,
`such as air, may be flowed through the inner pipe to
`remove heat generated by the transmission line stub.
`Similarly air (or some other fluid such as deionized
`water) may be flowed on the outside of the outer elec(cid:173)
`trically conducting pipe.
`
`Page 5 of 9
`
`

`

`5,065,118
`
`3
`The transmission line stubs may be organized in a
`variety of topologies. For example, a first transmission
`line stub may extend from the source to the load. A
`second transmission line stub may have one end con(cid:173)
`nected to the source and the other end terminated. Al- 5
`ternately, two or more transmission line stubs may be
`connected in series between the source and the load.
`Another topology is to connect the source directly to
`the load and connect a first end of one or more transmis(cid:173)
`sion line stubs to the connection between the source and 10
`the load. The unconnected end of each transmission line
`stub would be terminated by a short circuit, open cir(cid:173)
`cuit, or other circuit of known admittance.
`A matching network designed according to the pre(cid:173)
`ferred embodiment of the present invention may be 15
`tuned by minimizing phase and magnitude errors, as
`described herein, or may be tuned by minimizing the
`power reflected from the input of the matching network
`toward the source, as in the dithering technique de(cid:173)
`scribed in U.S. Pat. No. 4,951,009, referenced above.
`
`20
`
`4
`Transmission line stub 46 consists of a transmission
`line conductor 31 and a transmission line conductor 32
`separated by a ferrite dielectric material. A magnetic
`field is applied to transmission line stub 46 by a current
`supplied by DC power supply 43 through a wire 42
`wrapped around transmission line stub 46. Varying the
`current through wire 42, and thus the magnetic field
`applied to transmission line stub 46, varies the relative
`permeability of transmission line stub 46. A terminator
`39 of transmission line stub 46 may be, for example, a
`short circuit, an open circuit or some other circuit with
`a predetermined admittance. The admittance of termi(cid:173)
`nator 39 is herein referred to as termination admittance.
`For the purpose of calculating the input admittance, the
`termination admittance of terminator stub 45 is the ad(cid:173)
`mittance of load 22. Also, throughout the following
`discussion, values are given in terms of admittance,
`rather than its inverse (impedance), in order to simplify
`the formulas used in the discussion.
`The matching network is adjusted so that admittance
`(Yin) between a line 33 and a line 34 of the matching
`network is equal to the output admittance (Y g) of source
`21. This may be done in two steps. In the first step, the
`magnetic field applied to transmission stub 45 is varied
`so that the admittance of transmission stub 45 between
`line 33 and 34 is equal to Y g+ I -jB, where B is a con(cid:173)
`stant andj is the imaginary number (- l)i. In the second
`step, the magnetic field applied to transmission stub 46
`is varied so that the admittance of transmission stub 46
`between line 33 and 34 is equal to - I+ jB. Thus the
`admittance of transmission stub 46 cancels out the imag(cid:173)
`inary component of the admittance of transmission line
`stub 45 and leaves the total admittance of the matching
`network to be Y g·
`FIG. 2 shows a matching network of the type shown
`in FIG. 1 applied to a system which is used in a plasma
`process inside a plasma chamber 2. The load of the
`system is generated by the voltage across an electrode 5
`and walls 7 of plasma chamber 2, which act as a separate
`electrode. A capacitor 6, typically between IO and 100
`picofarads isolates a de voltage component on electrode
`5 from the matching network. The matching network
`matches the output admittance between a line 13 and a
`line 14 of a power supply 1. A first transmission line stub
`of the matching network includes a coaxial transmission
`line conductor 9 enclosed by a coaxial transmission line
`conductor 8. A second transmission line stub of the
`matching network includes a coaxial transmission line
`conductor 11 enclosed by a coaxial transmission line
`conductor 10. As shown in FIG. 2, coaxial transmission
`line conductor 9 is electrically connected to capacitor 6,
`coaxial transmission line conductor 11 and line 13 of
`power supply 1. Also, coaxial transmission line conduc(cid:173)
`tor 8 is shown electrically connected to coaxial trans(cid:173)
`mission line conductor 10, line 14 of power supply 1,
`and walls 7 of plasma chamber 2. A short circuit, such
`as a disk 12, is used to electrically connect coaxial trans(cid:173)
`mission line conductor 11 to coaxial transmission line
`conductor 10 as shown. A vacuum pump 3 is used to
`pump out gasses within plasma chamber 2 through a
`conduit 4.
`FIG. 3 shows a cross-sectional side view of a possible
`implementation of the matching network shown in
`FIG. 2. Coaxial transmission line conductor 9 may be
`implemented using a copper tube 49 having an outside
`radius of approximately 0.0095 meters and a length as
`calculated below. Coaxial transmission line conductor 8
`may be implemented using a copper pipe 50 having an
`
`25
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`FIG. 1 shows an electronically tuned VHF/UHF
`matching network in accordance with the preferred
`embodiment of the present invention.
`FIG. l shows a electronically tuned matching net(cid:173)
`work as shown in FIG. 1 used in a system which
`supplies power to a plasma process in accordance with
`the preferred embodiment of the present invention.
`FIG. 3 and FIG. 4 show cross-sectional views of a 30
`design implementation of an electronically tuned
`matching network as shown in FIG. l in accordance
`with the preferred embodiment of the present inven(cid:173)
`tion.
`FIG. 5 shows an electronically tuned VHF /UHF 35
`matching network in accordance with an alternate pre(cid:173)
`ferred embodiment of the present invention.
`FIG. 6 shows an electronically tuned VHF/UHF
`matching network in accordance with another alternate
`preferred embodiment of the present invention.
`
`40
`
`DESCRIPTION OF THE PREFERRED
`EMBODIMENT
`In FIG. 1, a source 21 is shown connected to a load
`l2 through an electronically tuned VHF /UHF match- 45
`ing network. The electronically tuned VHF/UHF
`matching network consists of a transmission line stub 45
`and a transmission line stub 46, arranged in the shown
`topology.
`Transmission line stub 45, in general, may be of any 50
`transmission line type with two conductors separated
`by a ferrite dielectric material. Throughout the discus(cid:173)
`sion of the present invention, the term ferrite dielectric
`material means ferromagnetic or antiferromagnetic di(cid:173)
`electric material. For example transmission line stub 45 55
`may be twin leads with a ferrite dielectric material be(cid:173)
`tween each of the twin leads. Alternately transmission
`line stub 45 may be coaxial.
`In FIG. 1, transmission line stub 45 is shown to con(cid:173)
`sist of a transmission line conductor 29 and a transmis- 60
`sion line conductor 30 separated by a ferrite dielectric
`material. A magnetic field is applied to transmission line
`stub 45 by a current supplied by DC power supply 44
`through a wire 41 wrapped around transmission line
`stub 45. Varying the current through wire 41, and thus 65
`the magnetic field applied to transmission line stub 45,
`varies the relative permeability of transmission line stub
`45.
`
`Page 6 of 9
`
`

`

`5,065,118
`
`6
`r0 =radius of the outer conductor (meters)
`r;=radius of the inner conductor (meters).
`Once the characteristic admittance of the transmis(cid:173)
`sion line stub is calculated, the input admittance of the
`transmission line stub (Y;n) may be calculated using the
`following formula:
`
`Y;n= Yo[(Y,+jY0 tan (a))l(Y0 +jY, tan (a))]
`
`20
`
`5
`inside radius of approximately 0.0159 meters and a
`length as calculated below. A ferrite dielectric 53 is
`placed between copper tube 49 and copper pipe 50.
`Ferrite dielectric 53 may be for example, a plurality of
`ferrite toroidal cores, available as part number M3-665 5
`from National Magnetics Group, Inc. having a business
`address of 250 South Street, Newark, N.J. 07114. A
`magnetic field, used to vary the relative permeability of
`ferrite dielectric 53, is generated by current through a
`solenoid coil 55 wrapped around copper pipe SO as 10 where:
`Yr= termination admittance of the transmission line
`shown. Solenoid coil 55 is made of, for example,
`enamel, insulated, tightly wove 8 A WG copper wire.
`stub (Siemens)
`During operation the transmission stub is cooled by
`j = ( - 1 )!
`deionized water flowing within a center region 59 of
`a=2'1T(µ*E)!(d/f..)
`copper tube 49, and by air flow through an open region 15
`d=length of the transmission line stub (meters)
`57 between copper pipe 50 and solenoid coil 55. While
`!..=free space wavelength of the signal through the
`solenoid coil 55 may be placed directly upon copper
`transmission line stub (elf) (meters)
`pipe SO, in the preferred embodiment open region 57
`c=free space speed of light (300X 106 meters/-
`exists to allow fluid to pass between solenoid coil 55 and
`second)
`copper pipe 50.
`f=frequency of oscillation of the signal through the
`Coaxial transmission line conductor 11 may be imple-
`transmission line stub (Hertz).
`mented using a copper tube 51 having an outer radius of
`A three step formula may be used in the designing of
`approximately 0.0095 meters and a length as calculated
`a matching network such as that shown in FIG. 3. In the
`below. Coaxial transmission line conductor 10 may be
`first step, ferrite dielectric material for constructing
`implemented using a copper pipe 52 having a inner 25 ferrite dielectric 53 and ferrite dielectric 54 is chosen. In
`radius of approximately 0.0195 meters and a length as
`selecting the ferrite dielectric material it is important to
`calculated below. A ferrite dielectric 54 is placed be(cid:173)
`select material which will have a low loss and a high
`tween copper tube 51 and copper pipe 52. Ferrite di(cid:173)
`relative permeability in the operating frequency range
`electric 54 may be for example, a plurality of ferrite
`of the matching network. Solenoid coil 55 and solenoid
`toroidal cores, available as part number M3-665 from 30
`coil 56 will respectively vary ferrite dielectric 53 and
`National Magnetics Group, Inc. having the business
`ferrite dielectric 54 between a minimum relative perme(cid:173)
`address given above. A magnetic field, used to vary the
`ability (µm;n) and a maximum relative permeability
`relative permeability of ferrite dielectric 54, is gener(cid:173)
`(µmax).
`ated by current through a solenoid coil 56 wrapped
`In the second step, the length of copper tube 49 and
`around copper pipe 52 as shown. Solenoid coil 56 is 35
`copper pipe 50 of the first transmission line stub is deter(cid:173)
`made of, for example, tightly enamel, insulated wove 8
`mined by iteration so that the real component of Y;n of
`A WG copper wire. During operation the transmission
`the first transmission line stub may be made equal to Y g
`stub is cooled by deionized water (or some other fluid
`over the range solenoid coil 55 varies the relative per(cid:173)
`such as air) flowing within a center region 60 of copper
`meability of ferrite dielectric 53. Then the length of
`tube 51, and by air flow (or the flow of some other fluid 40
`copper tube 51 and copper pipe 52 of the second trans(cid:173)
`such as deionized water) through an open region 58
`mission line stub is determined by iteration so that the
`between copper pipe 52 and solenoid coil 56. While
`imaginary component of Y ;n of the first transmission line
`solenoid coil 56 may be placed directly upon copper
`stub may be canceled over the range solenoid coil 56
`pipe 52, in the preferred embodiment open region 58
`varies the relative permeability of ferrite dielectric 54.
`exists to allow fluid to pass between solenoid coil 56 and 45
`When making the iterative determination of the
`copper pipe 52.
`length (dt) of the first transmission line stub a good first
`FIG. 4 shows another cross-sectional front view of
`guess is three fourths the wavelength of the signal
`the matching network shown in FIG. 3. The cross-sec(cid:173)
`tional view shown in FIG. 4 is along an axis perpendicu(cid:173)
`through the first transmission line stub when the rela(cid:173)
`lar to the axis of the cross-sectional view shown in FIG. 50
`tive permeability of ferrite dielectric 53 is at its maxi(cid:173)
`3.
`mum value. That is when:
`
`In general, for each of the transmission line stubs
`shown in FIG. 3, the characteristic admittance of the
`transmission line (Y 0 ) may be calculated by the follow-
`ing formula:
`
`55
`
`where:
`µ. 0 =permeability of free space (4'1TX lQ-7 Henry/m- 60
`eter).
`µ=relative permeability of the ferrite dielectric be(cid:173)
`tween the inner and the outer conductors.
`E0 =permittivity offree space (8.86 X lQ-12 Farad/m-
`eter)
`E=relative permittivity of the ferrite dielectric be(cid:173)
`tween the inner and the outer conductors (dielec(cid:173)
`tric constant).
`
`65
`
`d1 = i(A/(£*µmax)I) (meters)
`
`When making the iterative determination of the
`length (d2) of the second transmission line stub a good
`first guess is one half the wavelength of the signal
`through the second transmission line stub when the
`relative permeability of ferrite dielectric 54 is at its
`maximum value. That is when:
`
`In the third step, the number of turns of solenoid coil
`55, the number of turns of solenoid coil 56, the maxi(cid:173)
`mum current through solenoid coil 55 and the maximum
`current through solenoid coil 56 is determined so that a
`magnetic field of sufficient strength may be generated
`
`Page 7 of 9
`
`

`

`5,065,118
`
`7
`8
`Applicant's invention may be applied to a variety of
`to reach the desired minimum relative permeability of
`matching network topologies. For example, in FIG. 5, a
`ferrite dielectric 53 and ferrite dielectric 54.
`matching network, consisting of a transmission line stub
`What follows is an example of the design of a typical
`101 and a transmission line stub 102 connected in paral(cid:173)
`system in which Yg=0.02 Siemens, the maximum signal
`lel, is shown between a source 121 and a load 122.
`frequency is 100 megahertz and the load admittance has S
`Transmission line stub 101 includes a transmission line
`a real component only which varies from 0.05 Siemens
`conductor 131, a transmission line conductor 132 and a
`to 0.5 Siemens. For both the first transmission line stub
`terminator 133. Transmission line stub 102 includes a
`and the second transmission line stub r0 is selected to be
`transmission line conductor 129, a transmission line
`0.0159 meters; and r;is selected to be 0.095 meters.
`Ferrite dielectric 53 and ferrite dielectric 54 are se- 10 conductor 130 and a terminator 128. A transmission line
`135 connects source 121 to load 122.
`lected to be made from a plurality of ferrite toroidal
`cores, available as part number M3-665 from National
`For best operation of the two stub system shown in
`Magnetics Group, Inc. This material has a maximum
`FIG. 5, the part of transmission line extending from
`permeability (µ.max) of 20 when there is no magnetic
`source 121 to the connection of stub 101 should have a
`field applied to it and a minimum permeability (P.min) of IS characteristic admittance equal to the output admit-
`5 when an external magnetic flux density (B) of 1000
`tance of the source. Further, for best tuning range, the .
`length of transmission line 135 measured from the con-
`·
`f
`b 01
`h
`·
`f
`b 102 h
`Id
`gauss is applied.
`nection o . stu 1
`s ou
`to t e connection o ~tu
`The length of the first transmission line stub is chosen
`be three-eighths of a wavelength of the s1~n~l ge?erated
`to be three-fourths the wavelength of the signal through
`20 by source 121, and the length of transm1ss1on !me 135
`th fi e Irst transm1SS1on me s u w en e re a 1ve perme-· · 1. 1 t'
`
`t b h
`th
`
`.
`
`f fi
`measured from the connection of stub 102 to load 122
`'t d" I
`t · 53 ·
`t "t
`·
`1 e
`bil.
`h
`d b h
`· h h
`f
`h
`f h
`·
`I
`a 1ty o cm e 1e ec nc
`IS a I s maximum va u
`1
`s oul
`e t ree-e1g t s o a wave engt o t e s1gna
`20) · thi
`th 1
`th
`ual 0 185
`t
`1..
`generated by source 121.
`s case e cng
`s · . me ers.
`cq
`• ~n
`""'max=
`.
`.
`.
`A check IS made to see that at the maximum fre-
`. t"
`Alternat1vely, m FIG. 6, a matchmg network, con-
`r
`d fi
`h
`f
`(loo
`ah
`)
`· ·
`· ·
`f
`d
`· ·
`b 201
`meg ertz an or t e range o vana ion
`qucncy
`f fi
`an a transm1ss1on
`·
`b"I'
`"t d' 1 t . 53 2S s1stmg o a transm1ss1on me stu
`f I d ad
`~ttance, permea 1 ity 0 em e
`ie ec nc
`o oa
`line stub 202 connected in series, is shown between a
`may be vaned between P.max (20) and P.min <5)_ to. gcn~r-
`source 221 and a load 222. Transmission line stub 201
`ate a real component ofY;n.of the first transmission .Ime
`includes a transmission line conductor 231 and a trans-
`stub that equals Yg (0.02 Siemens). The check venfies
`mission line conductor 232. Transmission line stub 202
`that the length is acceptable because at the minimum 30 includes a transmission line conductor 229 and a trans-
`mission line conductor 230.
`load admittance of0.05 Siemens, when permeability(µ)
`offerritedielectric53is 13.5, Y;nequals0.019-j(0.015);
`We claim:
`and, at the m~um load ad~it~ce of. 0.5 S~emens,
`1. A transmission line stub in a matching network
`when permeab1ht~ (µ) of femte d1electnc 53 IS 10.3,
`coupled between a power supply and electrodes used in
`Y;n equals 0.019-J(0.092).
`. .
`.
`3S a plasma process, the transmission line stub comprising:
`. Next, the length of the second transm1ss1on lme. stub
`a first electrically conducting pipe;
`ts chosen to be one-half the. ~avel~ngth of the signal
`a second electrically conducting pipe, the second
`thro~gh the sec~~d transm1~s1on . lme s~ub w~en t~e
`electrically conducting pipe being placed inside the
`relative permeability of femte d1electnc 54 1s at its
`first electrically conducting pipe;
`maximum value (JLmax=20), in this case the length 40
`ferrite dielectric material between the first electri-
`equals 0.125 meters.
`cally conducting pipe and the second electrically
`A check is made to see that at the maximun:i ~re-
`conducting pipe, the ferrite dielectric material
`quency (100 megahertz) and for the range of vanat10n
`being in thermal contact with the first electrically
`of load admittance, permeability of ferrite dielectric 54
`conducting pipe and the second electrically con-
`may be varied between JLmax (20) and JLmin (5) to gener- 4S
`ducting pipe;
`ate an imaginary component of Y ;n of the second trans-
`variance means for changing the relative permeabil-
`ity of the ferrite dielectric material; and,
`mission line stub that cancels out the imaginary compo-
`nent of Y;n of the first transmission line. The check
`cooling means for cooling the second electrically
`verifies that the length is acceptable. The imaginary
`conducting pipe and the ferrite dielectric material,
`component of Y;n of the first transmission line varies so
`the cooling means including fluid flowing through
`between -j(0.015) when the load admittance is 0.05
`the second electrically conducting pipe.
`Siemens and -j(0.092) when the load admittance is 0.5
`2. A transmission line stub as in claim 1, wherein the
`Siemens. This range is covered. Specifically, when per-
`fluid flowing though the second electrically conducting
`meability (µ) of ferrite dielectric 54 is 7.0, Y;n of the
`pipe is deionized water.
`second transmission line is + j(0.012). Similarly, when SS
`3. A transmission line stub as in claim 1 wherein the
`permeability (µ.) offerrite dielectric 54 is 16.5, Y;n of the
`variance means includes a magnetic field generator
`second transmission line is + j(0.103). Thus the chosen
`which comprises:
`lengths of the first transmission line stub and the second
`wire wrapped around the first electrically conducting
`transmission line stub are acceptable.
`pipe, the second electrically conducting pipe and
`In order to generate a magnetic flux density of 1000 60
`the ferrite dielectric material; and,
`gauss when ro is 0.0159 meters and r; is 0.095 meters,
`a current generating means for generating current
`solenoid coil 55 and solenoid coil 56 are required to
`through the wire.
`produce a maximum magnetic field intensity (H) of
`4. A transmission line stub as in claim 3 additionally
`96,000 amperes-turns per meter. This may be accom-
`comprising flow means for flowing fluid between the
`plished, for example, by setting current through sole- 6S wire and the first electrically conducting pipe.
`5. A transmission line stub as in claim 4 wherein the
`noid coil 55 and solenoid coil 56 to have a maximum
`value of 40 amperes, and constructing solenoid coil 55
`fluid flowing between the wire and the first electrically
`and solenoid coil 56 to have 2400 turns per meter.
`conducting pipe is air.
`
`Page 8 of 9
`
`

`

`5,065,118
`
`9
`6. A matching network with a plurality of transmis(cid:173)
`sion line stubs, the matching network being coupled
`between a power supply and electrodes used in a plasma
`process, wherein each transmission line stub comprises:
`a first electrically conducting pipe;
`a second electrically conducting pipe, the second
`electrically conducting pipe being placed inside the
`first electrically conducting pipe;
`ferrite dielectric material between the first electri(cid:173)
`cally conducting pipe and the second electrically 10
`conducting pipe;
`variance means for changing the relative permeabil(cid:173)
`ity of the ferrite dielectric material; and,
`cooling means for cooling the second electrically
`conducting pipe and the ferrite dielectric material, 15
`the cooling means including fluid flowing through
`the second electrically conducting pipe.
`7. A matching network as in claim 6 wherein the
`variance means includes a magnetic field generator
`which comprises:
`wire wrapped around the first electrically conducting
`pipe, the second electrically conducting pipe and
`the ferrite dielectric material; and,
`current generating means for generating current
`through the wire.
`8. A matching network as in claim 6 wherein wherein
`a first transmission line stub from the plurality of trans(cid:173)
`mission line stubs is coupled between the power supply
`and the the electrodes, and wherein a first end of a
`second transmission line stub from the plurality of trans- 30
`mission line stubs is coupled to the power supply and a
`second end of the second transmission line stub is cou(cid:173)
`pled to a terminator.
`9. A matching network as in claim 8 wherein the
`terminator is a short circuit electrically coupling a first 35
`electrically conducting pipe of the second transmission
`line stub to a second electrically conducting pipe of the
`second transmission line stub.
`10. A matching network as in claim 6 wherein a first
`transmission line stub from the plurality of transmission 40
`line stubs and a second transmission line stub from the
`plurality of transmission line stubs are coupled in series
`between the power supply and the the electrodes.
`11. A matching network as in claim 6 wherein a first
`end of a first transmission line stub from th