ee 6.1 INTRODUCTION Microwave eystems normally consist of several microwave components including the source ang the load being connected to each other by waveguide or co-axial or transmission line eysten, ay these components must be built with low standing wave ratios, lower attenuation, lower: insertion Jonson and other desirable characteristics to achieve the desired transmission of microwave sign The rectangular and circular waveguides, cavity resonators etc. that were discussed in previous chapters arc also microwave components. In this chapter we study other components like waveguide junctions, joints, corners, drives, posts and screws, directional couplers, ferrite devices phase shifter, Gilterne 6.2 WAVEGUIDE MICROWAVE JUNCTIONS = i Ata certain position in a waveguide system, many a times it becomes necessary to split all or part of the microwave energy into particular directions. This is achieved by waveguides or in general by microwave junctions, These are combined to form coupler units that direct the energy as required Alternately the vame junction may be used to combine two or more signals. In general, a microwave junction ia an interconnection of two or more microwave components as shown in Fig. 6.1 ‘Thin junction has four ports similar to low frequency two-port networks. Fig. 6.2 shows a rowave wuree at port D and microwave loads at ports ®, @ and ®. wn he microwave junction is analogous to a traffic junction where a number of roads meet on which vehicles enter and Jeave the traffic junction. Ina similar manner, when input from microwave iz applied at port © a part of it comes out of port ® another part out of port @ some part out ours:secenen® 201 ® aa a uae ' part may come out of port © itself due to mismatch between port O = row" . joa Port @) Microwave or wave guide Por @ junction — T w w Hw PotD] source Junction Load |Por@ || ww Load Port @) Fig. 6.2 62.1 Scattering or (S) Parameters Low frequency circuits can be described by two port networks and their parameters such as Z, Y, H,ABCDetc. as per network theory. Here network parameters relate the total voltages and total currents as shown in Fig. 6.3. ao) 202 Inasimilarway at microwave frequencics, wo talk of travelling waves with aunociy instead of voltages and currents and the microwave junction can be dofined by what m Me Doon, ‘S parameters or scattering parameters (similar to H, ¥, Z parameters). Referring to relay can be sven that for an input at one port, we have four outputs ns discussed ontlier, Simi i apply inputs to all the ports, we will have 16 combinations, which aro ropresonted in a mn ity, and that matrix is called asa Scattering Matrix. It isa square matrix which gives all the combiner ips between the various input and output ports ofa microwave ju ttle " altering (8) parameters,” Mlerowave and Rada Enging rng of power relationshij ion clements of this matrix are called scattering coe/ficients or sc The Fig. 6.3 ‘Toobtain the relationship between the scattering matrix and the input/output powers at diffore ports consider a junction of'n' number of transmission lines wherein the ith line (can be any line from 1 ton) is terminated in a source as shown in Fig. 6.4. me line Leo Fig. 6.4 “i ae are at nn be terminated in an impextanco other than the characteristic impedance ec. Z,,# Z,) and all the remaining lines (from 2nd to nth line) in an Guimoaliies oqtall ag,a? s 203 the ith line, then it divides itself ‘There will be no reflections from inces are equal to characteristic wave by po tho ancient wave at tho junction duo toa source at 1fa0 umber of lines a8 a, ay,..d, a8 shown in Fig. 6.4. 4928 ne and the nekdont waves are absorbed since their impeda orth N (7). But, there isa mismatch atthe Lat line and hence there will be a reflected at ane ie petintotho junction. ie? e by is related toa, by, b, = (reflection coefficient) a, = Six“ pore, $3 7Fefleetion coefficient of Ist Hine, 1 = reflection from Ist line and “ i = source connected at ith line. tence, the contribution to the outward travelling wave in the ith line is given by 1 0; =Sa-a [ez bp= b= =.= 01 ] ase 2: Let all the (2 ~1) lines be terminated in an impedance other than Z,- (ie. Zy,# 2, for all J thelines) : “Then, there will be reflections into the junetion from every line and hence the total contribution tothe outward travelling wave in the ith line is given by j 0, = Sj,-41 + Si, "da + Sig 20% ~ +S;,°%% (6.1) | j=1 ton since ican be any line from 1 0 0 | Therefore, we have, i by = Sy a+ Siz 42+ S15 * + Sin Bn i by = Soy 4 + Sp A + S25 45 +» + Son On Sx 0 + Sy Ag + Sy3.03 * + Si On In matrix form, & Su Sz Sin a b, a. | os a . : (6.2) , Sy Sex San 0 Column Matrix [6] ‘Scattering Column Matrix (a] corresponding to ‘Matrix (S] Corresponding to Reflected waves of order nxn Incident Waves or Output or Input. (6.3) (0) = [S} [a]Microwave 10 Huy yy May, 204 i When a junction of n numbor of wavegtild a’a ropresont inputs to partioular ports, bia represent outputs out of various port, Si corresponds to seattoring cootfieionta rosuliing die to input at ith pork and ayy fer “4 out of jth port. §, denotes how much of powor it rofloctod baal from the Junction Inte the iW port aj, power is appliod al. the ith port. stnoll Jan In qonaiclored, Properties of (S) Matrix [S] is always as quare matrix of ordar (1% 1) L 2. [S]is a symmetric matrix. ie. Sy = Sr 3, [S]isa unitary matrix ie, [Ss] 1S]* =] where, [S] =complex, conjugate of [5] [1] =unit matrix or Identity matrix of the same order aa that of 1], 1 torm of any row (or column) multiplied by the comple, mu of any othor row (or column) in zero, 4. The sum of the products of on conjugate of the corresponding Lo ee WN, e. Sp Sj =O ie, 2 in Say oo (aia planon (4ay the /eth port) are moved away from the ch of the coofficients Sy involving, fe will be multiplied 5. If any of the terminal or reference junction by an electric distance [}yl), © by the factor ein, 6.3. MICROWAVE T-JUNCTIONS A T-junction is an intersection of three waveguides in Lhe form of Englinh alphabet “M, There are several types of Tee junctions. The following Tee junctions will he dincunned, 1 H-plane ‘Tee junction 2, E-plane Tee junction 3, E-H plane Tee junction (Hybrid junction) 4. Magic-T junction 5. Rat race junction 6.3.1. H-plane Tee Junction 2H plane Teo iene. is formed by cutting n rectangular slot along the width ofa main waveguide panto urna guide. the side arm — called the H-arm an shown in Mig. 6.6. The port e main waveguide are called collinear ports and port D is the Harm or side arm.tors 206 el sane Tee is 80 called because the axis of tho side nrm ia parallel to the planes of the main BP gsion line. Asall three arms of H-plano Tee lie in the plane of magnotic field, the magnetic or ivides itself into the arms. Therefore thia is alao called a current junction. sa ropertis ofa H-plane Tee can be completely defined by ita [5] matrix, The order of scattering, atti’ x 8 since there are three possible inputs and 3 possible outputs. Plano of Coplanar aymmotry ‘arms / Port 2 Port 1 Power cutront flow le WA ~ Harm (side arm) Pon, > Fig.6.5 Su Size Sis (5) =|S21 See Sap (6.4) Sor Sse Soa. Now we determine the S-parameters Sy, i-> 1, 2,3,j > 1, 2, 3by applying the properties of (S]. 1. Because of plane of symmetry of the junction scattering coefficients S,, and Sz, must be equal. 2 S13 = Sas 2, From the symmetric property, Sy = Sj: Siz = Sey, Sx = Soo = Stay Sis = Sar 3, Since port is perfectly matched to the junction Ssq =0 With these properties [S] matrix of Eq. 6.4 becomes, [Ss] = ie, We have four unknowns.Micrc 206 rowave and Radar hag 4, From the unitary property sls" = ‘ Su Sia Si] [Sn Siz Sis] [1 0 0 a ie, Sp Sep Sis ||Si2 Sez Sig] =]0 1 0 Ss Ss OJ|S, Si 70 oo1 Multiplying we get, : RCs Si Sin + Siz Sjo + Sig Sg = (RC, = row 1, column 1) or [Sy 12 + 1Si21? + [Syg1?=1 ~ Similarly R,C,: ISygl? + | Sol? + 1S1312=1 462) RC: 1S,g1?+ 1S,g17=1 ney RCy: S13 Sir + Sis St2= 0 (6.9) From Eq. 6.8 21Sj3/2=1 or Sys 6.10 : : Comparing Eqs. 6.6 and 6.7, we get ISI? = 1 Spo? - Str = Sap (6.11) From Eq. 6.9, Sis #0, (St + Si2=0, or Si, =— Siz or Sy =-Syz or Sp=-Sy (6.12) Using these in Eq. 6.6. 1 1 1 ISy 1? + 1Sy1? + See 1 or 21S) =5 oF Sy 2 «-(6.13) From Eq. 6.11 and 6.12, 1 i sls (6.14) Sw 2 (6.14) oT a and Sa = (6.15) Substituting for S13, S,1, Sy and Syo from Eq. 6.10 and Eqs. 6.13 to 6.15 in Eq. 6.5, we get 2.21 = Sai 1.4 - =|. 6) Sl 5 (6.16) aus y v2 VB We know that [6] = [S] [a] (from Eq. 6.3)resonators od oat 4 Loser 18] '(a; 2 2 fe &| fb, 2. A) fo 22 fa e |) |e 0} (a M2 a b, = 4¢,-20,+4 6.17) 5 1 = 9 ~ 52 + te AGS) b, 2 1 ABB) np 5h Fle + ete S638) = ia tha, ABIY case 1 , 40,4, =a,=0, ie., Input is given at port @ and no inputs at port © and port 2). Substituting these in Eqs. 6.17, 6.18 and 6.19, we get by =f -% and 6,=90 Let P, (corresponding to az) be the power input at t port @. Then this power divides equally setween ports © and @ in phase i.e., P, = P, (power outputs at the respective ports corresponding wo by and by). But Pz =P, + P,=2P\=2 Py ‘The amount of power coming out of port © or port ® due to input at port ©. B B, qa =10 logiy + = 10 logs = 10 loss, | 5) B10 809 p B0\ 5) 10 log =— 10(0.3010) =— dB Hence the power coming out of port © or port @ is 3 dB down with respect to input power at port ©, hence the H-plane Tee is called as 3-dB splitter. Further when TE, mode is allowed to propagate into port @, the electric field lines do not change their direction when they come out of port ® and ®, hence called H-plane Tee. i. ‘vaves that come out of ports © and ® are equal in magnitude and phase. Case 2: a =a,=4,0,=0 The208 Microwave and Raga. Engi ‘ ier ie, The output at port @ is addition of the two inputs at port © and port @ and these are i. phase. added in 6.3.2 E-Plane Tee ‘A rectangular slot is cut along the broader dimension of a long waveguide and a side ‘attached as shown in Fig. 6.6. Ports ® and @ are the collinear arms and port @ is the Rare is Pot @ E-arm Port @) Pot Fig. 6.6 E-plane Tee. When TE, mode is made to propagate into port @, the two outputs at port © and @ will have aphase shift of 180° as shown in Fig. 6.7. Since the electric field lines change their direction when they come out of port © and , it is called a E-plane Tee. E-plane Tee is a voltage or series junction symmetrical about the central arm. Hence any signals that is to be split or any two signal that are to be combined will be fed from the E arm. ® Fig.6.7209 jo” trix of an E-plane Tee can be wal! wing ma‘ “pl ‘ee can be used to describe its properties. In general, the me se eport @ (side or E arm) is proportional to the difference between instantaneous powers i: | ic fom ports © and @. oe om gneeffective vale of the power leaving the E arm is proportional to the phasor difference tne powers entering ports © and ®. When powers entering the main arms (ports © and a opposition, maximum energy comes out of port’@ or E-arm. "tis a three port junction the scattering matrix can be derived as follows: [s]is a 3 *3 matrix since there are 3 ports. SurSz Sis [5] =|S2, Sep So Ss Ssz Sss. z. The Scattering coefficient S23 =—Sy3 (6.20) since outputs at ports ® and © are out of phase by 180° with an input at port ®. 3. Ifport @ is perfectly matched to the junction. = (6.20) (6.22) «(6.23) | | - WY Si Sp Sy][Sn Sez. Sie] fro 0 ies Se Sa Sel|Se Sz Si|=|0 1 0 Sis Sis 0 |] St, Sis 0 ool RG ISy1? + [Spa]? + | Sial® RCo? ISy21? + [Soo]? + 1 Sig)? = 1 R,C3: 1Sy31? + 1 Sys} RC: (S13: Sii-Sia Sig=O > Equating Eqs. 6.24 and 6.25, we get = Sag ’ (6.28) (6.29) From Eq. 6.26,210 From Ka. he Using these values (Eqs. gsst0 6. 30) om 1m Ea. 6.24. , on Case 1: between the two outputs. Hence E-plane Tee alse act. Case 2: a, =a, =a,0;=0 Substituting again in Eqs. 6.34 to 6.36, we get ie., equal inputs at port © and port © result in no output at port ©a 214 ae ence gaitariy we ean have all combinations of inputs and outputs. -H Plane (Hybrid or Magic) Tee x slots are cut both along thé width and breadth ofa long waveguide and side arms port © is the H-arm. and 3 perectangulan iepttached ss shown in Fig. 6.84, Ports © and @ are collinear arms, vt @ is the E-arm. Port @) pot @ Pot ® Ley on (a) Signal into E=arm Output signal Signal into H=arm Fig. 6.8 (b)212 Microwave and Radar Engin Such a device became necessary because of the difficulty of obtaining a complete}, three port Tee junction. This four port hybrid Tee junction combines the power dividing, Matche, " ofboth H-plane Tee and E-plane Tee as shown‘in Fig 6.8) and has the advantage of being a “tea matched at all its ports. This has several useful applications as will be seen later, (u?tely properties of E-H plane Tee, its scattering matrix can be obtained as follows. ving the 1. [S]isa4 x 4 matrix since there are 4 ports Si Se Sis Sia . iy2|S2 52 Ss Se a Sa S2 Sop Su (63 Sa Se Sis Sus 2. Because of H-plane Tee section Sx = Sig : (6a 3. Because of E-plane Tee section Su =-Si4 (638 Because of geometry of the junction an input at port ® cannot come out of port ® since the, are isolated ports and vice versa. - a Su = Sig=0 +-(6.40) 5. From symmetric property, S;;= Sj Siz = Say» S13 = Sai} So = Spas Say = Sag 3 Soa = Sep i Sar = Sia * (6.41) 6. Ifports @ and @ are perfectly matched to the junction. Sug = Syq= 0 -=(6.42) Substituting the above properties from Eqs. 6.38 to 6.42 in Eq. 6.37, we get Si Sp Sis Sta Siz Sop Sos ~ Sia =| er eo (6.43) Sy Sy 0 0 7. From unitary property, [S] [S]* = [J] Su Sir Sis Sta 1000 ive,,| 5 =| Sz Sis -Siu} _]0 1 0 0 Sis Sty 0 0 0010 Sy Sy 0 0 oo0o01 F1SyP=1 (6.44) 45) tynse wa peat coor Veh 1g 1 ABAD) nics? I Ayal 4 [Ay l*= ABAD from Fiq. 6.46 and Keg. 6,47, ? Ay fi AOA) Aus 4 AAD Comparing Fas, 6.44 and 6.40, we gol i Ay * Fay AED) sing thiene values from Ka, (48 rnd 6.41) in Hay 6,44, we 18) 14 1Syyh* 4 het st 2 (Bh + 15h? #0 jie. By 2820 AGEN) : From Eq. 6.45, By = 0 ML) | Thin means ports © and @ are alno perfectly matehed to the junction, Hones in nny four port | junction, if any two ports re perfectly rntched to the junction, Uhen thes remaining two porte are rfectly matomatically matched to the junction, Such a junetion where in nll the four ports are 9 ‘matched to the junetion ia called a Magic Tee, ‘The [S] of Magic Tee is obtained by #ubstita 6.52 in Eq, 6.43. ithe nentloring, parameters from Kas, 6.4% i | "Mic 2a rowave and Radar Shaheen a = 3g a4); dy = Fg +4) ~ b= pm ~ ay); by = ~o =a) 654 Using Fa, 6.54, we look at the properties of Magie Tee for some important cases, 420.0 = =a Case 1: Substituting these in Bq. 6.54, we get = 1 y, = 2; by = b4=0 by = pie yy Oa M4 the property of H-plane Tee. a, 40,4, Thisis the property of E-plane Tee. a, #0, a2 Case 3: a, bg = Fb sn 2 V2 chen power is fed into port ©, nothing comes out of port ® even though they are collinear posts (Magic), Henee ports © and @ are called isolated ports. Similarly an input at port @ cannot come out at port ©. Similarly Band H ports are isolated ports. Case 4: Then 0, by = by ‘This is nothing but the additive property. Equal inputs at ports @ and @ result in an output at port © (in phase and equal in amplitude). Case that is equal inputs at ports © and @ results in an output at port @ (additive property) and no outputs at ports ©, ® and ®. This is similar to case 4. 6.3.4 Applications of Magic Tee A magic Tee has several applications. A few of them have been discussed here. (a) Measurement of Impedance : A magic Tee has been used in the form of a bridge, as shown in Fig. 6.9 for measuring impedance. Microwave source is connected in arm @ and a null detector in arm ®. The unknown impedance is connected in arm @ and a standard variable known impedance in arm ©. Using the propertiespesos 215 ow the power from microwave source (a) gots divided equally between arms ® and @ je TS “vet? ‘ oe socharacteristic impedance Z, and hence there will be reflections from arms © and @.1f py othe unknown impedanco and standard varinblo impedances). These impedances are not j ane the reflection coefficients, powers Set and P55 enter the Magic Tee junction from 7 ae @ and @ ae shown in Fig, 6.9. The resultant wave into arm @ i.., the null detect strated as follows: ' @ detoctor tor can be : Sender vant Ie ae impedance a We —@ Fig. 6.9 Magic Tee for measurement of impedances. ‘The net wave reaching the null detector (Refer Fig. 6.9) wn) eee) For perfect balancing of the bridge (null detection) Eq. 6.55 is equated to zero. 3 (Py ~ Pa) 1 fey 2s (1 ~ Pa: | | Pi—P2=0 or Py=P2216 WOM 50 Samy py, ; Pie "Wy Ant. arty ef Ty ty ly Uy ty =I ie, Ry + jXy = Nyt IX, 1 en Ry ah, and Xy=K, Thus the unknown impedance can he mennured by adjusting thie standord varisble gay till the bridge is balanced and hoth impedances hecome equal, 8 (b) Magic Tee as a Duplexer :'The transmitter and receiver are connected in gory © respectively, antenna in the B-arm or port @|and port @ of Magic Vee in tertninatuy 4 _ matehed load as shown in Fig. 6.10. During trandnjission half the power reaches, the q from where it is radiated into apace. Other half r¢ pyhen the matched Jond where it in shay without reflections, No transmitter power reaches the receiver since port D and & sre jgp ports in a Magic Tee. During reception, half of the fercen power goes to the reexiy other half to the transmitter are isolated during reteption ax well aa during, tranuy Fig. 6.10 Magic Tee as a Duplexer. (©) Magic Tee as a Mixer : A magic'Tee can also be used in microwave receivers as a mixer where the signal and local oscillator are fed into the E and H arms as shown in Fig. 6.11. Half of the local oscillator power and half of the received power from antenna goes to the mixer where they are mixed to generate the IF frequency. WP =fiy~ fy Magic Tee has many other applications such as a microwave discriminator, microwave bridge etc.247 ‘Antonia = Matchod load = AI Local oscilator Fig. 6.11. Magic Tee as a mixer. 6.3.5 Rat race junction ‘This is a four port junction, the fourth port being added to a normal three port Tee. A typical rat race junction is shown in Fig. 6.12, _— angl4 Pat f ea 1944 Fig. 6.12 Ratrace ring ‘The four arms/ports are connected in the form of an angular ring at proper intervals by means ofseries (or parallel) junctions. These ports are separated by proper electrical lengths to sustain standing waves. For proper operation, itis necessary that the mean circumference of the total race be 1.5 2, and that each of the four ports be separated from its neighbour by a distance of /4 When power is fed into port © it splits equally (in clockwise and anticlockwise directions) into ports @ and ® and nothing enters port ®. At ports @ and © the powers combine in phase but at port @ cancellation occurs due to 2/2 path difference. For similar reasons any input applied at port is equally divided between ports @ and ® but the output at port © will be zero. The rat race can\e 218 Microwave and Rader Ena : ano be uned for combining two signals or dividing a single signal into two equal hay," unequal signals nre applied at port. ©, an output proportional to their sum will emerge fo tng ®@ and ® while a differentin} output will appear at port ®, Tom Ports The scattering matrix of a rat race junction (also enlled hybrid junction) shown below in iden] conditions (i.c., neglect ng leakage coupling values), 0S. 0 Sy Sn 0 Sm 0 =F 0 Se 0 Sy 0 Ss, 0 ) can be Written, (656) 6.4 DIRECTIONAL COUPLERS Directional couplers are flanged, built in waveguide assemblies which can sample a small amoy of microwave power for measurement purposes. They ean be designed to measur incident reflected powers, SWR (standing Wave Ratio) values, provide a signal path to a receiver or Perform other desirable operations. They can be undirectional (measuring only incident power) orbi-dreeton (measuring both incident and reflected) powers. In its most common form, the directional coupler is four port waveguide junction consisting ofa primary main waveguide and a secondaty auilins waveguide as shown in Fig. 6.31a. Received power 1 PB Received WA . power B Pp 3 . Pr Coupled “WV a. y— “power Back power (b) Fig. 6.13 (a) A schematic of a directional coupler (b) Directional coupler in dicating powers.ws o> pst | matched terminations at all its ports, the properties of an ideal directional coupler can be rized as follows. "a portion of power travelling from port © to port @ is coupled to port @ but not to port @. ‘a portion of power travelling from port © to port © is coupled to port @ but not to port © idirectional case). Aportion of power incident on port @ is coupled to port @ but not to port and.a portion of the power incident on port © is coupled to port © but not to port ®. Also ports D and @ are decoupted as are ports @ and ©. | mall portion of input power at port © is coupled to port @ so that measurement of this small i power jis possible. Ideally no power should come out of port @,Fig. 6.136 indicates the various ntloutput powers. in P; = incident power at port ©. P, = received power at port ©. P; =forward coupled power at port ©. P, =back power at port @. ‘The performance of a directional coupler is usually defined in terms of two parameters which are defined as follows. Coupling Factor C:'The coupling factor ofa directional coupler (D.C.) is defined as the ratio of the incident power P; to the forward power ‘P measured in dB. ive A657) : Fy Directivity D: The directivity of a D.C. is defined as the ratio of forward power ‘P7 power P, expressed in dB. to the back . in D=10 061022 aB oe For a typical D.C. C =20dB, D=60dB B, . =20=10 log 2 og 2 *BR iw oc PB Sees TAMERS POWERS : i " pares x os defined 2 dese uy, se ameter called Isolation is semetimes defined a Jeol ation Ae ccovaal coupler. 1 .< defined as the ratio of the incid==s power power Plexprssed ind. There are several types of directional couples that have been developed Hie guide couplers with common broed wall-sections ig. 6.14). branching Eaide common wall instead of coupling holes (Fig. 6.15). shart slot couplers (Fiz. 6.36). (Fig. 6.17), Joop directional coupler (Fig. 6.18). couplers made from parallel aroun: rips running internally within the waveguide structure. It may be noted that in most of the directional couplers only three of the four ports are us-4. unwanted port is normally terminated in a matched load built into it. The two waveguides (primar; and secondary) share a common wall. This common wall has got gole or holes for coupling th? energy flowing into the main waveguide to the side waveguide and hence called a side hole coupl=r. A two hole directional coupler is most commonly used.onators fl 224 Major \ harmonics Py Fundamental Input oe Fig. 6.15 Two hole branching guide coupler. —, SS Fig. 6.17 Bifurcated coupler. Fig. 6.18 Loop directional coupler. 64.1 Two-hole Directional Coupler ‘The principle of operation of a two-hole directional coupler is shown in Fig. 6.19. It consists of two guides the main and the auxiliary with two tiny holes common between them as shown, The two holes are at a distance of A,/4 where 2, is the guide wavelength.Microwave and Radar Engnggy, 7 222 Port 2 V+ Pe, ya in Auxillary WG Po Py j Port4 \ Fig. 6.19 Two hole directional coupler. Port 3 hn in phase at the position of 2nd hole and hence they es are out of phase by 180° at the position of the 1st hole and therefore they cancel each other making P, = 0 (ideally). The magnitude of the power coming out of 2 holes depends upon the dimension of the two holes. Since the distance between holes is 4,/4, P, is made’0’ (since the incident power will have to travel a distance of (4 +i,/4) when it comes back from hole ® resulting in 180° phase shift. Compared to incident power leakage through hole © entering port ®. ‘The number of holes can be one (as in Bethe cro: f Multi-hole coupler). The degree of coupling is determined by size and location of the holes in the The two leakages out of holes © and ® bot add up contributing to P;, But the two leakag ssguide coupler) or more than two (as ina waveguide walls. ‘Although a high degree of directivity can be achieved as a fixed frequency, itis quite difficult over a band of frequencies. In this connection, it should be relaized that the frequency determines the separation of the two holes asa fraction of the wavelength. 6.4.2. Bethe or Single-hole Coupler ae ae A single-hole directional coupler is shown in Fig. ees 6.20. Here the directivity is improved as the Bethe Co*xial probe coupler relies on a single hole for coupling process, rather than the separation between two holes. The power entering port: is coupled to the co-axial probe output and the power entering port @ in absorbed by the matched load. The auxiliary guide is placed at such an angle that the magnitude of the magnetically excited wave a made equal to that of the electrically excited wave for improved directivity. In this coupler, the waves in the auxiliary guide are generated through a single hole which includes both electric and magnetic fields. Because of the phase relationships involved in the coupling process, the signals generated by the two types of coupling cancel in the forward direction and reinforce in the reverse direction. Fig. 6.20 Bethe or single-hole coupler.ae sonalor® nF » 223 gas, Scattering Matrix ofa Directional Coupler yout the propertios ofthe directional coupler to arrive at the [S] matrix. , Directional coupler is a four port network, Hence [S] is a 4% 4 matrix Su Se Sis Sy Su Sr. Spy 8, ose [sy =| 52 Sez Sos Say Ss Sez Sig Sy Sn Siz Sis Sig (6.60) 2, Ina directional coupler all four ports are perfectly matched to the junction. Hence the diagonal elements are zero. ies 8 a2 = Sgq = Syq= 0 (6.61) 3, From symmetric property, Sy= Sj, (6.62) Ideally back power is zero (P, = 0) i.e., There is no coupling between port ® and port ®. S13 = Sq, =0 (6.63) 4. Also there is no coupling between port ® and port © . Spy = Sin =0 (6.64) ‘Substituting in Eq. 6.55, the values of scattering parameters as per Eqs. 6.66 to 6.59, we get 0 S, 0 Sy Sz 0 Sy 0 (6.65) 0 Sy 0 Sy o ig 0 Oa 0 [J = 5. Since [S] [S*] = J, we get 0 S, 0 Suj[® Se © Sul [1 00 0 S. 0 Sy O||S, 0 Sy O|_jo 100 OS 0) Sal Ose 0 Se] 2 \0,0 1.0 Sy, 0 Sy 0 0001 S, 0 Sy 0 RC: [S17 + 1Syl?=1 RC: 1Syo1? + [Sag 1? = RiCy: [Sa |® + |Sggl?=1 RC, Siz Sia + Sia Sg“ slorowave and Radar f; 0 ng 224° Comparing Eq. 6.61 and 6.62, ‘ Su = Sia MGs Comparing Eq. 6.62 and 6.63, Sin = Suu thay Let us assume that S,» is real and positive = 4 Sos M04, P Soy + 8x3 P=O P [Sox + Sal = 0 P40, Sox + Sop =O om Siz = Soa = From Eqs. 6.69 and 6.72. i.e., Sp, must be imaginary. Let Sos =50 = Sis Sip =Sys=P (tranamission parameter) and Si, = St ia. Also, ptt gai ‘Substituting these values in Eq. 6.65, [S] matrix of a directional coupler is reduced to 0 PO iq P 0 jq 0 « = (6.74) =)o iq 0 P iq 0 PO 6.5 WAVEGUIDE JOINTS It is not possible to build a waveguide system in one piece and may require several sections connected by joints. These are the waveguide joints and must be constructed in such a way that a good connection is made between any two sections of a waveguide without any irregularities and without affecting the E and H-field patterns, Irregularities in a joint cause reflection effects, create standing waves and increase the attenuation. A rotating joint could be required as in a radar system where the transmitter/receiver is stationary and the antenna system is revolving. ‘There are several types of waveguide joints. Some of them are shown in Fig. 6.21. The semipermanent butt joints, a bolted flange (Fig. 6.20a) consists of two sections bolted together with a gasket to exclude moisture. For perfect mechanical alignment, it should be ensured that there are no bends or discontinuities and the ends of the waveguides and flanges must have mirror smooth finish to avoid reflection effects,wonntor® 296 | “ 1b shows a quarter wavelength flange joint which uses no mechanical connection and > pet rout at point B creates a short circuit at point A due to the standing wave distribution these points. ‘The field patterns are not disturbed and no discontinuity exists but there is eat feakage of energy through the open flange. ie a a @) (b) Flanges © Fig. 6.21 (a) Bolted flange (b) 24/4 Flange joint (c) Choke joint. ‘The choke joint (Fig. 6.21c) is superior to butt joint and is very widely used. It consists of flanges that are connected to the waveguide at its center. The right side flange is flat and the ft side one is slotted, 2,/4 deep from the inner surface of the waveguide. This is positioned at a datance of 4/4 from the point where the flanges are joined. Due to the 2/2 length taken together, ‘short circuit is created at the place where the walls are joined together resulting in an electrical Sort, Mechanically they are separated by as much as one tenth of a wavelength and the area canbe sealed with a rubber gasket for avoiding moisture.Microwave and Radar ng ‘Waveguide bend, comer and twist are shown in Fig: 622. These components are usefal fora, the direction of the guide by desired angle. ‘The bends can be H bend or E bend (Fig. 6.2! dimension the H lines are affected (Hbend) and: ifth the Elines are affected (E bend). ‘The bending radius than 1.05 and mean length as long as possible. (Rmin 2a). If the bend is in the direction e bend isin the direction ofnarson the wigg must be at least 22, to avoid SWp, inn, =1.5 ban E bend and Ruin =1.5 9 f° Satara Radius must be (a) ef gH eat (b) —T 1. Circular-Rect esc lar ©90" Twist \ a8 Twit @ fig22. (@)H bend and E bend, (b) H-plane corer and E-plane comer, (c) Circular to rectangular ane rectangular to circular transitions (Tapers) (d) 90° twist and 45° tiwsts. tend for small reflections). Sharp 90° bends create total reflection resulting in infinite SWR. ‘Therefore bends have to be gradual. “at lower frequencies a bend may have to be very long and in such cases, a corner would be cred. A mitered 90° bend (shown in Fig. 6.226) is a corner. “Mitered bends or corners are normally H-plane corners where d= 0.65 D as shown in Fig. 6.2c- Smaller values of din a E-plane corner have the danger of arcing and hence not common: Tnorder to minimise the reflections, the mean length L must be an odd. number of quarter wavelength, 80 that reflected wave from both ends of the waveguide are completely cancelled. iii 0, 1,2, - a, ie, L=(Qn+1)7£ where | Flexible waveguides are special bends that are made of ribbon brass ich are edge interlocked so that the internal dimensions are preserved in any angular position. The exterior of such waveguides is coated with rubber to avoid dangers of oxidation and humidity. Figure 6.21e shows waveguide transitions or tapers. These are required whenever itis required tojoin two waveguide sections that have different shapes for their cross-sectional areas, For: ‘example xtends over a distance of acircular to rectangular waveguide transition has a gradual taper that nore than 2, Similarly rectangular to circular transition is also possible. Ifthe circular waveguide tarries the TE, mode the rectangular waveguide will operate in the dominant TE,)mode and vice- versa, We can also have two rectangular waveguide tapers of different dimensions provided that there is a match between their characteristic impedances. ‘ Waveguide twists such as 90° and 45° twists shown in Fig. 6.21d are helpful in converting vertical to horizontal polarizations or vice-versa. Twists can be incorporated along with bends also. ) 67. WAVEGUIDE IRISES be any waveguide system, when there is a mismatch thete will be reflections. In transmission es in order to overcome this mismatch lumped impedances or stubs of required value are placed at precalculated points. In waveguides too, some discontinuities are made use of for matchingMicrowave and 220 nd Radar Engineering it tho guide, causing mismatch (and productio: wrponos, Any auscoptance apponring ACrO®S 4 vavel) vent to be cancelled by introducing another susceptance of the same magnitude nat opposite nature, Iron (also called windows, apertures, diaphragms or obstacles) shown in Fig g ie are made une of for the purpose, I) ES OR ie wave | Vy wei) guide | | | . oa Equivalent: see : ‘rout i | 4 @ w Fig. 6.23 Waveguide irises. ‘An inductive iris (Fig. 6,28a) allows a current to flow where none flowed before. The iris is placed in a position where the magnetic field is strong (or where electric field is relatively weak), Since the plune of polarisation of electric field is parallel to the plane of iris, the current flow due to iris causex a magnetic field to be set up. Energy storage of magnetic field takes place and there is iin increase in inductance at that point of the waveguide. : In capacitive iris (Fig, 6.230), itis seen that the potential which existed between the top and bottom walls of the waveguide now exists between surfaces which are closer and therefore the capacitance has increased at that point. The capacitive iris is placed in a position where the electric field is strong. ‘The inductive and capacitive irises if combined suitably (correctly shaped and positioned) the inductive und capacitive reactances introduced will be equal and the iris becomes a parallel resonant ircuit (Fig. 6.23c). For the dominant mode, the iris presents a high impedance and the shunting effect for this mode will be negligible. Other modes are completely attenuated and the resonant iris acta ai 4 band pass filter to suppress unwanted modes. Figure 6.23d, shows a series resonant iris which is supported by a non-metallic material and is transpresent to the flow of microwave energy. ' 6.8 POSTS AND TUNING SCREWS . i When u metallic cylindrical post is introduced into the broader side of waveguide, it produces the same effect as un iris in providing lumped reactance at that point. If the post extends only a short dintunce (< 2,/4) into the waveguide, it behaves capacitively (Fig. 6.24a), and this capacitiveer “ 229: ot Resonators prance increases Wi __ 0 esonant circuit (Fig. 6.240), ifitis > Ay, the post behaves i the post acts as ith depth of penetration. When the depth is equal to,/4, inductively (Fig. 6.24c) and 2 . i a saauctive susceptance decreases when the post is moved further away from the centre of the ave ide. When the post is extended completely across the waveguide, the post becomes: inductive 24d). The susceptance vs penetration (/t) characteristics is shown in Fig. 6.25. fis Fig. 6.24 Waveguide posts.Na 230. Microwave and Radar Eng, ‘The amount of susceptance decreases as the diameter of the post is reduced. Ifthe post, . thicker, effective Q will be lowered and can act as a band pass filter similar to an iris, The big advantage of the post over an iris is that it is readily. adjustable. An adjustable Post j or slug. The adjustable or tuning screws are shown in Fig. 6.26. Ag in cans known asa screw t " I A posts depending upon the depth of penetration, the tuning eerew may introduce inductiy capacitive susceptance. i8 made q | a ava) | 4 + ' ® © @ Fig. 6.26 @ Acombination of two screws. a4 apart can be used to match a waveguide to its load similar to ‘use of two fixed stubs in a transmission line. A very effective waveguide matcher can be realised when two tuning screws are placed in close proximity separated by 3 4,/8 as shown in Fig. 6.27, ‘This is almost similar to the double stub matching in transmission lines. 2 T > Blkm re ATT cmb oS Fig. 6.27 6.9 COUPLING PROBES AND COUPLING LOOPS When a short antenna in the form of a probe or a loop in inserted into a waveguide, it will radiate and if it is placed correctly, the wanted mode will be set up. Figure. 6.28 shows the correct positioning of the coupling probes for launching dominant TE,, mode. The probe is placed at a distance of 2,/4 from the shorted end of the waveguide and the centre of broader dimension of the waveguide because at that point electric field is maximum. This probe will now act as an antenna which is polarized in the plane parallel to that of electric field.= mode | Fig. 6.28 Coupling probe and loop. | mrecoupling loop placed at the centre of shorted end plate of the waveguide can also be used to " jgunch TEyo mode i.e., coupling is achieved by means of a loop antenna located in a plane 3 dicular to the plane of the probe. It is thus seen that probes couple primarily to the: electric _ Feld and loops to a magnetic field but in each case both are set up because electric and magnetic - felds are inseparable. 1 | tna waveguide system itis not possible to attach a fixed resistive load as a termination. Graphited. | and at the end of the waveguide as shown in Fig. 6.290 can dissipate energy to achieve SWR's - ifless than 1.01. Alternatively a resistive rod placed at a point in the waveguide where electric feld strength is maximum can aleo do a similar job (Fig. 6.296). Fig. 6.29¢ shows a wedge of resistive material (powdered iron or carbon mixed with a binder deposited on a dielectric strip) inthe form of a taper that can act as a termination. All these are matched terminations and | almost no reflection occurs. | Apermanent metal plate welded at the end of the waveguide (a short circuit) may be employed | forcomplete reflection. This could be made movable (similar to a choke joint discussed earlier) by | having an adjustable plunger as shown by Fig. 6.29d and e. Lr Ses == 1, Waveguide Grdphited sand Waveguide ~ Resistverod @) (b) Fig. 6.29232° ' Permanent plate at end @ 6.11" FERRITE DEVICES Ferrites are non-metallic materials with resistivities (p) nearly 10" times greater than metal, and with dielectric constants (,) around 10-16 and relative permeabilities of the order of 1009, ‘They have magnetic properties similar to those of ferrous metals. They are oxide based compounds having general composition of the form MeO. Fe, i.e., a mixture of a metallic and ferric oxide where MeO represents any divalent metallic oxide such as MnO, ZnO, CdO, NiO or a mixture of these. They are obtained by firing powdered oxides of materials at 1100°C or more and pressing them into different shapes. This processing gives them the added characteristics of ceramic insulators s0 that they can be used at microwave frequencies. Ferrites have atoms with large number of spinning electrons resulting in strong magnetic properties. These magnetic properties are due to the magnetic dipole moment associated with the electron spin. Because of the above properties, ferrites find application in a number of microwave devices to reduce reflected power, for modulation purposes and in switching circuits. Because of high resistivity they can be used upto 100 GHz. Ferrites have one more peculiar property which is useful at microwave frequencies i.e, the non- reciprocal property. When two circularly polarised waves one rotating clockwise and other anticlockwise are made to propagate through ferrite, the material reacts differently to the two rotating fields, thereby presenting different effective permeabilities to both the waves. i.€. 1 Hen p; for left circularly polarised wave and ¢,2, ,2, pz for the right circularly polarised wave. 6.11.1 Faraday Rotation in Ferrites Consider an infinite lossless medium. A static field B, is applied along the z-direction. A plane TEM wave that is linearly polarised along the x-axis at t= 0 is made to propagate through the ferrite in the z-direction. The plane of polarisation of this wave will rotate with distance, a phenomenon known as Faraday Rotation,een za gos neatly polarined wave can be regarded as the vector sam ck two cates kato crcaiathy Tae wave (Ex/2 vectors shown in Pig. 62). The lerrite material cflers different characteristics ipeee waves, with the result that the phase change for one wave is larger than the other wave Resting in rotation of the linearly polarized wave, at 2=L. {tis observed that a rotation of 100 degrees or more per om of ferrite length is typical for gertes ata frexqueney of 10 GHz. I the dinection of propagation is reverand, the plane of larication jnues to rotate in the same direction iz., z=1to2=0, the wave will arrive back at==0 polarised at an angle 26 relative tozazxis, wha tbls diliaci ieatatis (+ In fact, the angle of rotation is given by oe ~ (6.75) o= 38 BL) ( 5) where, [= length of the ferrite rod £. = Phase shift for the right circularly polarised (component in clockwise direction) wave with respect to some reference. phase shift for the left circularly polarised (component in anticlockwise direction) j BL ‘wave with respect to the same reference. Ina practical ferrite medium, there will be finite losses. The propagation constant for ge eee ee creularly polarised wave will have unequal attenuation constants and unequal phase port t Port2 comstant. Due to this, the direction of Faraday rotation will be different in the two regions call oi above and below the resonant frequency (,). Atwo port ferrite device is shown in Fig. 6.31 when a wave is tranamitted from port © port 2, it undergoes rotation in the anticlockwise Fig. 6.31 direction as shown. Even if the same wave is allowed to propagate from port © port ©, it will undergo rotation in the same direction (anticlockwise). Hence the direction of rotation of linearly polarised wave is independent of the direction of propagation of the wave.234 Microwave and Radar Eine 6.11.2 Microwave Devices which make use of Faraday rotation We discuss three important devices which make use of faraday rotation @ Gyrator (b)Isolator (© Circulator (a) Gyrator It is two port device that has a relative phase difference of 180° for transmission from port =~} oe ais © port @ and ‘no’ phase shift (0° phase shift) Port = for transmission from port @ to port @ as 4 | Radians = shown in Fig. 6.32. The construction of a gyrator is as shown. Fig. 6.32 in Fig. 6.32. It consists of a piece of circular 7 ” waveguide carrying the dominant TE,, mode with transitions toa standard rectangular waveguide with dominant mode (TEj0) at both ends. A thin circular ferrite rod tapered at both ends ig located inside the circular waveguide supported by polyfoam and the waveguide is surrounded by a permanent magnet which generates de magnetic field for proper operation of ferrite. To the input end a 90° twisted rectangular waveguide is connected as shown. The ferrite rod is tapered at both ends to reduce the attenuation and also for smooth rotation of the polarized wave. Operation : When a wave enters port © its plane of polarization rotates by 90° because of the twist in the waveguide. It again undergoes Faraday rotation through 90° because of ferrite rod and the wave which comes out of port @ will have a phase shift of 180° compared to the wave entering port ©. But when the same wave (TEyp mode signal) enters port ®, it undergoes faraday rotation throvgh 90° in the same anticlockwise direction. Because of the twist, this wave gets rotated back by 90° comes out of port © with 0° phase shift as shown in Fig. 6.30. Hence a wave at port © undergoes a phase shift of x radians (or 180°) but a wave fed from port @ does not change its phase in a gyrator. (b) Isolator An isolator is a 2 port device which provides very small amount of attenuation for transmission from port © to port ® but provides maximum attenuation for transmission from port @ to port ©. This requirement is very much desirable when we want to match a source with a variable load. In most microwave generators, the output amplitude and frequency tend to fluctuate very significantly with changes in load impedance. This is due to mismatch of generator output to the load resulting in reflected wave from load. But these reflected waves should not be allowed to reach the microwave generator, which will cause amplitude and frequency instability of the microwave generator.235 k ert ie oa WG — Transition Fig. 6.33 When isolator is inserted between generator and load, the generator is coupled to the load with ero attenuation and reflections if any from the load side are completely absorbed by the isolator without affecting the generator output. Hence the generator appears to be matched for all loads in the presence of isolator so that there is no change in frequency and output power due to variation in load. This is shown in Fig. 6.34. oe foe ee | ake EsMicrowave and Radar Engi 236 * Eninerng 6.36) is similar to gyrator except that an j, : ction of isolator (Fig. < Construction : The construction o! Se eet ov" iid) aad op Re ist ti Jar waveguide on toa makes use of 45° twisted rectangul egu 7 a rotate saete rod Gnetead of 90° in gyrator), a resistive card is placed along the larger dimension or” famgular wavogntide, soe to absorb an wave whose plane of polarisation is parallel tothe plan® tf resistive card, The resistive card does not absorb any wave whose plane of polarization perpendicular to its own plane. Circular wavegu! en Me oe 0 4s" Twist Resistive card card Fig. 6.35 Constructional details of isolator. Operation : A TEyo wave passing from port © through the resistive card and is not attenuated. After coming out of the card, the wave gets shifted by 45° because of the twist in anticlockwise direction and then by another 45° in clockwise direction because of the ferrite rod and hence comes out of port @ with the same polarization as at port © without any attenuation. But a TE,9 wave fed from port ® gets a pass from the resistive card placed near port ® since the plane of polarization of the wave is perpendicular to the plane of the resistive card. Then the wave gets rotated by 45° due to Faraday rotation in clockwise direction and further gets rotated by 45°pty Resonators 237 i ed fe ty se twist in the waveguide. Now the plane of polarization of the wave 3h peparalel with that ofthe resistive card and hence the wave willbe completely absorbed by sesistive card and the output at port © will be zero, This power is dissipated in the card as __ Matin practice 20 to 30 4B isolation is obtained for transmission from port ® to port O. cscs 7 @ Circulator circulator is a four port microwave device which pas a peculiar property that each terminal is fonnected only to the next clockwise terminal. ie., port © is connected to port © only and not to port @ and @ and port ® is connected only to port ® atc. This is shown in Fig. 6.36. Although there is no restriction on the number of ports, four ports ‘are most commonly used. They are useful in parametric amplifiers, tunnel diode, amplifiers and duplexer in radars. Construction : A four port Faraday rotation circulator is shown in Fig. 6.37. The power entering port © is TE, mode and is converted to TE,, mode because of gradual rectangular to circular transition. ‘This power passes port @ unaffected since the electric field is not significantly cut and is rotated through 45° due to the ferrite, passes port © unaffected (for the same reason as it passes port ®) and finally emerges out of port ®. Power from port © will have plane of polarization already tilted by 45° with respect to port ©. This power passes port @ unaffected because again the electric field is not significantly cut. This wave gets. rotated by another 45° due to ferrite rod in the clockwise direction. ‘This power whose plane of polarization ig tilted through 90° finds port ® suitably aligned and emerges out of it. Similarly port @ in coupled only to port ® and port ® to port ©. Port @ Line paratel Fig. 6.37 Four port circulator.Nae 238 Microwave and Radar aha re ed as a duplexer for a radar antenna system as shown j 1. Acirculator can be used as a dup! iNFig ag Matched termination Fig. 6.38 ‘Transmitter feeds the antenna while the received energy is directed to the receiver. The powerful radar transmitter is isolated from the sensitive receiver and also the same antenna can be used for both transmission and reception. This is the duplexer action being performed by a circulator. 2. Wecan have three port circulators, strip line circulators that can have several applications. ‘Two three port circulators can be used in tunnel diode or parametric amplifiers as shown in Fig. 6.39. 8. Circulators can be used as low power devices as they can handle low powers only. Match 3 port circulators: lead Output Amplified signal Fig. 6.39“239 See filters are frequency selective circuits that are made up of reactive components. We know that shore are basically four filter types the low pass, the high pass, the band pass and the band glimination filters. ‘The attenuation characteristic of these are well known. A microwave filter is quilt using & high Q cavity in contrast to low frequency filters that use lumped inductors and capacitors. Also there contain more irregularities compared to their low frequency counter parts. Normally more than one cavity will be cascaded to get the desired response. Such a filter is termed as n Multicavity filter. Coupling between cavities is accomplished by means of irises or slots and tuning is done by capacitive screws. Depending upon the depth of penetration of screw, it is possible to tune the cavity above or below the designed frequency. However to avoid lowering @ of cavity, depth of penetration is kept minimum. A typical multicavity filter is shown in Fig. 6.40. Since the effect of tuning screw is to lower the resonant frequency, each cavity is 1. Sey pe 30 , higher the dielectric constant ofa medium the moro slowly a microwave rf sleavels sys it, Since most of the microwave signal in a waveguide travels through it. a2 goat of te microwave signal in a waveguide travels through the center of the waveguide, sovement ofa a lectric slab with ¢, > 1 towards the centre of the waveguide means that | Beramave signal moves more slowly the nearer the slab gets to the center. The electric field Seeibution in the broader dimension ofthe waveguide will be modified by the dielectric slab #0 tic itis distorted from sinusoidal to that indicated in Fig. 6.47. ifthe vane is inserted deeper, there is more change in the medium and there is a greater phase if, The electric field distribution also shows that the dielectric has the same effect of increasing spe broader dimension of the waveguide which reduces the wavelength in the waveguide. ‘The Srount of phase shift is maximum when the slab is at the centre and minimum when its adjacent he wall of the waveguide if the dielectric vane is placed such that the vane's inside dimension is parallel tothe direction of the electric flux lines. Low loss. material ¢,> 1 Fig. 6.48 Electric field distribution. Inthe above dielectrie vane variable phase shifters the phase shift is changed continuously From one value to another and hence they are also termed as analog phase shifters. Ifa fixed (rather ‘hen variable) phase shift is produced, then we name them as digital/diserete phases shifters Digital phase shifters are mostly used in phased array antennas whereas analog phase shifters are used in bridges and instruments. {A precision phase shifter can be realised by a rotary phase shifter useful in microwave measurement. It basically consists of three circular waveguide sections all of which contain one delete vane. The contre section is rotatable providing the required phase shift. It works on the principle of converting a linearly polarised TE, mode into a circularly polarised mode. Alternately, ferrite phase shifters utilise faraday rotation for providing the necessary amount of hase shift ag in the case of Gyrator (discussed earlier) which provides 180° phase shift in one direction and 0° phase shift in the reverse direction.244 6.15 MICROWAVE ATTENUATORS For perfect matching sometimes we require that the microwal i ae nee Ty el ca completely without any reflection and also insensitive to frequency. we m, attenuators. is gain or loss in 4Bs, for providing: Attenuators are commonly used for measuring power gain © » fOr Providing ing daar simevhoee eae input to a particular stage to prevent oye pte i is er between instruments, for reducing the pow' 2 oe ting their ou . erloadine and also for providing the signal generators wit ; that precise measurement could be made. Attenuators can be classified as fixed or rae (continuous or step variation) types. Fixed Attenuators are used where fixed amount of attenuation is to be provided. If such a attenuator absorbs all the energy entering into it, we call it as a waveguide terminaton net normally consists of.a short section of a waveguide with a tapered plug of absorbing materia a, end. The tapering is done for providing a gradual transition from the waveguide medium, ae absorbing medium thus reducing the reflection occurring at the media interface. Figure ¢ shows such a fixed attenuator where a dielectric slab consisting of glass slab coated with aquadey or carbon film has been used asa plug. Here the lossy dielectric or vane shown is V-shaped and can occupy the whole of the wav: Variable attenuators provide continuous or step wise variable attenuation. For rectangular waveguides. these attenuators can be flap type or vane type. For circular waveguide type is used. Shorting plate Microwave Dielectric slab | | | Fig.6.49 Fixed attenuator.evans iia lap tyne attonte ton alton 10 Hij6: Ch60, condita i randitive edument on dae inmecrteed inte ainval alot curt atone the contye OF the whiter dinuenatan of tie quides, he flag iw mienented fn anal davon attoysinye tt (a dercout Ito the cones oF Che waveguide, he dageucet atbenustion be penne at ty the depth of asertion of the flap, eset Hanintive Gor Looking ft re te) ye r Mf | a om Maximum attenuation Minin attenuation Fig. 6.60 Flap atlonuator However the flap attenuator dial noods to bo calibrated agiinalcatandird sit brnota preepion attenuator, Thevane type attenuator, (Hig. 6.61) basically conuiataof pliuw yim with a contig of aqudoye oe carbon similar toa fixed vane attonuator, [thin vans anol al Leones in mds movable, it ea bee used ast variable attomuttor, The vane Positioned at tho contro of (he waveguide can be moved ‘aterally from the contro, whore it provider maximum attenuation (o tho odyemwhore the attondation “considerably reduced sinco th olectrie fol ting avo alway conontrated at the conte of the r |Microwave and Radar Eng; 246 Ere ends for matching the attenuator to the wavegui made equal to A,/2. The amount. ofattenuarg is waveguide, Tho vane is tapered at both i ined i this adequate match is obtained if the taper long! @ to ponte sonsitive and also has to be calibrated against a precision (standard) attenuato, Movable vane | —— 3 ‘Minimum altenation — Maximum altenation Fig.6.51 Movable vane attenuator. tive rotary vane attenuator provides precision attenuation with an accuracy of 2.1 %o¢ the indicated attenuation over the operating frequency range. It consists of three vanes. The central vane rotating type placed in the central section of a circular waveguide arrangement tapered at both ends. The other two vanes are in the rectangular sections as shown in Fig. 6.52. Are! Resistive vanes, @ @ ® | Fig. 6.52 Rolary wave precision attenuator. When all the three vanes are aligned their planes are at 90° to the direction of electric field. Hence there is no (or zero) attenuation. Vane I prevents any horizontal polarisation and hence electric field at the output of vane 1 is vertically polarised. The centre vane 2 is rotating type and if it as rotated by an angle 0, the [sin 0 component is attenuated and E cos 6 component is present at the output of vane 2 and the final output of the attenuator becomes E'cos28, which has the same polarisation as the input wave, The attenuation duc to this rotary vane attenuator is then equal to 20 log, con%0 = 40 log cos 0 that is independent of frequency and is precise. |